WO2017011414A1

WO2017011414A1 - Bispecific molecules comprising an hiv-1 envelope targeting arm

Info

Publication number: WO2017011414A1
Application number: PCT/US2016/041809
Authority: WO
Inventors: Chia-Ying Kao LAM; Jeffrey Lee NORDSTROM; Barton F. Haynes; Mattia Bonsignori
Original assignee: Duke University
Priority date: 2015-07-10
Filing date: 2016-07-11
Publication date: 2017-01-19

Abstract

The invention is directed to multispecific molecules comprising an HIV-1 envelope targeting arm and an arm targeting an effector cell, compositions comprising these molecules and methods of use.

Description

BISPECIFIC MOLECULES COMPRISING AN HIV-1 ENVELOPE TARGETING ARM

[0001] This invention claims the benefit of and priority to U.S. Serial No. 62/191,095 filed July 10, 2015, U.S. Serial No. 62/222, 115 filed September 22, 2015, and U.S. Serial No. 62/301,993 filed March 1, 2016, and International Application No. PCT/US 16/23355 filed March 21, 2016 the entire contents of each of which are hereby incorporated by reference in their entireties.

[0002] This patent disclosure contains material that is subject to copyright protection. The copyright owner has no objection to the facsimile reproduction by anyone of the patent document or the patent disclosure as it appears in the U.S. Patent and Trademark Office patent file or records, but otherwise reserves any and all copyright rights.

[0003] All patents, patent applications and publications cited herein are hereby incorporated by reference in their entirety. The disclosure of these publications in their entireties are hereby incorporated by reference into this application in order to more fully describe the state of the art as known to those skilled therein as of the date of the invention described herein.

FIELD OF THE INVENTION

[0004] The invention is directed to multispecific molecules, such as but not limited to bispecific and trispecific molecules (e.g. bispecific antibodies, bispecific diabodies, and trivalent binding molecules) comprising an HIV-1 binding domain and an effector cell binding domain, and their uses.

BACKGROUND

[0005] Highly Active Antiretroviral Therapy (HAART) has been effective in reducing the viral burden and ameliorating the effects of HIV-1 infection in infected individuals. However, despite this therapy the virus persists in the individual due to latent reservoir of HIV-1 infected cells which evade this treatment. Thus, there is a need for therapeutic agents for treatment of HIV-1 infected individuals, as well as agents that target virus infected cells and have the potential to reduce the latent reservoir of HIV-1 infected cells.

SUMMARY OF THE INVENTION

[0006] In some aspects the invention is directed to multispecific molecules, such as but not limited to bispecific and trispecific molecules (e.g., bispecific antibodies, bispecific diabodies, trivalent binding molecules, etc.) which comprise epitope-binding fragments of antibodies (e.g., VL and VH Domains) that enable them to coordinately bind immunospecifically to at least one target on HIV-1 envelope (e.g. but not limited to a V3 glycan or a CD4 binding site epitope) and at least one epitope of a second molecule that is not HIV-1 Env, for example but not limited to an effector cell which expresses CD2, CD3, CD8, CD16, T-Cell Receptor (TCR), KG2D, etc. epitope.

[0007] Selection of the VL and VH Domains of the polypeptide domains of the multispecific molecules of the invention is coordinated so that the polypeptides chains that make up such multispecific molecules assemble to form at least one functional epitope-binding site that is specific for at least one epitope of HIV-1 Env and at least one functional epitope-binding site that is specific for at least one epitope of a molecule that is not HIV-1 Env. In some embodiments, the multispecific molecules of the invention comprise an Fc Domain (Fc bearing multispecific molecules of the invention).

[0008] In non-limiting embodiments, the multispecific molecules comprise 1, 2 or all 3 of the CDR_Hs of a VH Domain with the specificity of the V3 glycan binding antibody DH542 (also referred to as DH270.6), a variant of DH542 called DH542 QSA, DH542 L4, and/or other antibodies from the DH542 lineage (DH542-like antibodies, e.g, DH270.UCA, DH270.IA4, DH270.IA3, DH270.IA2, DH270.IA1, DH270 (also referred to as DH270.1), DH473 (also referred to as DH270.2), DH391 (also referred to as DH270.3), DH429 (also referred to as DH270.4); DH471 (also referred to as DH270.5)), and/or 1, 2 or all 3 of the CDR_Ls of a VL Domain of the V3 glycan binding antibody DH542 (also referred to as DH270.6), a variant of DH542 called DH542 QSA, DH542 L4 (the VL of DH542 L4 comprises the VL from DH429 (also referred to as DH270.4)), and/or other antibodies from the DH542 lineage (DH542-like antibodies, e.g., DH270.UCA, DH270.IA4, DH270.IA3, DH270.IA2,

DH270.IA1, DH270 (also referred to as DH270.1), DH473 (also referred to as DH270.2), DH391 (also referred to as DH270.3), DH429 (also referred to as DH270.4); DH471 (also referred to as DH270.5)).

[0009] In non-limiting embodiments, the multispecific molecules comprise the VH Domain with the specificity of the V3 glycan binding antibody DH542, a variant of DH542 called DH542 QSA, DH542 L4, and/or other antibodies from the DH542 lineage (DH542-like antibodies , e.g., DH270.UCA, DH270.IA4, DH270.IA3, DH270.IA2, DH270.IA1, DH270 (also referred to as DH270.1), DH473 (also referred to as DH270.2), DH391 (also referred to as DH270.3); DH429 (also referred to as DH270.4); DH471 (also referred to as DH270.5)), and/or the VL Domain, of the V3 glycan binding antibody DH542, a variant of DH542 called DH542 QSA, DH542 L4 (the VL of DH542 L4 comprises the VL from DH429 (also referred to as DH270.4)), and/or other antibodies from the DH542 lineage (DH542-like antibodies, e.g., DH270.UCA, DH270.IA4, DH270.IA3, DH270.IA2, DH270.IA1, DH270 (also referred to as DH270.1), DH473 (also referred to as DH270.2), DH391 (also referred to as DH270.3), DH429 (also referred to as DH270.4); DH471 (also referred to as DH270.5)).

[0010] In non-limiting embodiments, the multispecific molecules comprise 1, 2 or all 3 of the CDR_Hs of a VH Domain with the specificity of the CD4 binding site antibody CH557 (also referred to as CH235.12) or any of the antibodies of the CH235 lineage (CH557-like antibodies, e.g., CH556 (also referred to as CH235.11), CH555 (also referred to as

CH235.10), CH493 (also referred to as CH235.9), CH492 (also referred to as CH235.8), CH491 (also referred to as CH235.7), or CH490 (also referred to as CH235.6)), and/or 1, 2 or all 3 of the CDR_Ls of a VL Domain with the specificity of the CD4 binding site antibody CH557 (also referred to as CH235.12) or any of the antibodies of the CH235 lineage (CH557- like antibodies, e.g., CH556 (also referred to as CH235.11), CH555 (also referred to as CH235.10), CH493 (also referred to as CH235.9), CH492 (also referred to as CH235.8), CH491 (also referred to as CH235.7), or CH490 (also referred to as CH235.6)).

[0011] In non-limiting embodiments, the multispecific molecules comprise the VH Domain with the specificity of the CD4 binding site antibody CH557 (also referred to as CH235.12) or any of the antibodies of the CH235 lineage (CH557-like antibodies, e.g., CH556 (also referred to as CH235.11), CH555 (also referred to as CH235.10), CH493 (also referred to as

CH235.9), CH492 (also referred to as CH235.8), CH491 (also referred to as CH235.7), or CH490 (also referred to as CH235.6)) and/or the VL Domain with the binding specificity of the CD4 binding site antibody CH557 (also referred to as CH235.12) or any of the antibodies of the CH235 lineage (CH557-like antibodies, e.g., CH556 (also referred to as CH235.11), CH555 (also referred to as CH235.10), CH493 (also referred to as CH235.9), CH492 (also referred to as CH235.8), CH491 (also referred to as CH235.7), or CH490 (also referred to as CH235.6)).

[0012] In certain embodiments an antibody (or a molecule comprising the CDRs, or the variable domains of such antibody) binds specifically to a particular target, peptide, or polysaccharide (such as an antigen present on the surface of a pathogen, for example gpl20, gp41, or CD3), even where the specific epitope may not be known, and do not bind in a significant amount to other proteins or polysaccharides present in the sample or subject.

Specific binding can be determined by methods known in the art. Various competitive binding assays are known in the art. With reference to an antibody antigen complex, in certain embodiments specific binding of the antigen and antibody has a KD of less than about 10⁶ Molar, such as less than about 10⁶ Molar, 10⁷ Molar, 10⁸ Molar, 10⁹, or even less than about 10¹⁰ Molar.

[0013] In some aspects, the present invention is directed to bispecific molecules, e.g.

covalently linked polypeptide chains to form bispecific antibodies, covalently linked diabodies and/or trivalent binding molecules and their use in the treatment of HIV-1. In certain aspects, the bispecific molecules of the present invention can bind to two different targets or epitopes on two different cells wherein the first epitope is expressed on a different cell type than the second epitope, such that the bispecific molecules can bring the two cells together. In certain aspects, the bispecific molecules of the present invention can bind to two different cells, wherein the bispecific molecules comprises an arm with the binding specificity for an HIV-1 envelope, for example as provided by the binding specificity of the V3 glycan binding antibody DH542, a variant of DH542 called DH542 QSA, DH542 L4 and/or other antibodies from the DH542 lineage, and/or provided by the binding specificity of the CD4 binding site antibody CH557 (also referred to as CH235.12) or any of the antibodies of the CH235 lineage, which arm binds to the HIV-1 envelope expressed on a first cell, e.g. HIV-1 infected cell, and a second arm with the binding specificity for an epitope expressed on a different cell type than the first cell, such that the bispecific molecules can bring the two cells together. In certain embodiment, the second cell is in effector cell which expresses CD2, CD3, CD8, CD16, T-Cell Receptor (TCR), KG2D, etc. epitope.

[0014] In certain aspects, the invention provides a bispecific molecule comprising a first polypeptide chain and a second polypeptide chain, covalently bonded to one another, wherein:

(I) the first polypeptide chain comprises in the N- to C-terminal direction:

(i) a domain (A) comprising a binding region of the light chain variable domain of a first immunoglobulin (VLl) having the binding specificity of a CD4 binding site HIV-1 antibody or a V3 glycan binding antibody (1);

(ii) a domain (B) comprising a binding region of a heavy chain variable domain of a second immunoglobulin (VH2) specific for an epitope (2); and

(iii) a domain (C) comprising a heterodimer promoting domain;

(II) the second polypeptide chain comprises in the N- to C-terminal direction:

(i) a domain (D) comprising a binding region of a light chain variable domain of the second immunoglobulin (VL2) specific for the epitope (2); (ii) a domain (E) comprising a binding region of a heavy chain variable domain of the first immunoglobulin (VH1) having the binding specificity of a CD4 binding site HIV-1 antibody or a V3 glycan binding antibody (1); and

(iii) a domain (F) comprising a heterodimer promoting domain; and wherein:

the domains (A) and (B) are linked so that they do not associate with one another to form an epitope binding site;

the domains (D) and (E) are linked so that they do not associate with one another to form an epitope binding site; and

the domains (A) and (E) associate to form a binding site that binds the HIV-1 envelope like a CD4 binding site HIV-1 antibody or like a V3 glycan binding antibody (1); and

the domains (B) and (D) associate to form a binding site that binds the epitope (2).

[0015] In certain aspects the invention provides such bispecific molecules wherein:

(i) domains (A) and (B) are separated by a peptide linker 1 ;

(ii) domains (C) and (B) are separated by a peptide linker 2;

(iii) domains (D) and (E) are separated by a peptide linker 1 ; and

(iv) domains (F) and (E) are separated by a peptide linker 2.

[0016] In certain aspects the invention provides bispecific molecules comprising a first polypeptide chain and a second polypeptide chain, covalently bonded to one another, wherein:

(I) the first polypeptide chain comprises in the N- to C-terminal direction:

(i) a domain (A) comprising a binding region of the light chain variable domain of a first immunoglobulin (VLl) having the binding specificity of the V3 glycan binding antibody DH542, a variant of DH542 called DH542 QSA, DH542 L4 and/or other antibodies from the DH542 lineage, and/or the binding specificity of the CD4 binding site antibody CH557 (also referred to as CH235.12) or any of the antibodies of the CH235 HIV-1 antibody lineage (1);

(ii) a domain (B) comprising a binding region of a heavy chain variable domain of a second immunoglobulin (VH2) specific for an epitope (2), wherein domains (A) and (B) are separated from one another by a peptide linker 1 ; and

(iii) a domain (C) comprising a heterodimer promoting domain, including but not limited to a K coil or E coil; wherein domain (C) and domain (B) are separated by a peptide linker

2;

(II) the second polypeptide chain comprises in the N- to C-terminal direction: (i) a domain (D) comprising a binding region of a light chain variable domain of the second immunoglobulin (VL2) specific for the epitope (2);

(ii) a domain (E) comprising a binding region of a heavy chain variable domain of the first immunoglobulin (VH1) having the binding specificity of the V3 glycan binding antibody DH542, a variant of DH542 called DH542 QSA, DH542 L4 and/or other antibodies from the DH542 lineage, and/or the binding specificity of the CD4 binding site antibody CH557 (also referred to as CH235.12) or any of the antibodies of the CH235 HIV-1 antibody lineage (1), wherein domains (D) and (E) are separated from one another by a peptide linker 1 ; and

(iii) a domain (F) comprising a heterodimer promoting domain, including but not limited to a K coil or E coil; wherein domain (F) and domain (E) are separated by a peptide linker 2; and wherein:

the domains (A) and (B) do not associate with one another to form an epitope binding site; the domains (D) and (E) do not associate with one another to form an epitope binding site; and

the domains (A) and (E) associate to form a binding site that binds the HIV-1 envelope like the V3 glycan binding antibody DH542, a variant of DH542 called DH542 QSA, DH542 L4 and/or other antibodies from the DH542 lineage, and/or like the CD4 binding site antibody CH557 (also referred to as CH235.12) or any of the antibodies of the CH235 HIV-1 antibody lineage (1); and the domains (B) and (D) associate to form a binding site that binds the epitope (2).

[0017] In certain aspects the invention provides such bispecific molecules, wherein the first or second polypeptide chain further comprises an Fc Domain. The invention also provides such bispecific molecules wherein the first or second polypeptide chain further comprises an Fc Domain and the bispecific molecule further comprises a third polypeptide chain.

[0018] In certain aspects, the invention provides bispecific molecules comprising a first polypeptide chain and a second polypeptide chain, covalently bonded to one another, wherein:

(I) the first polypeptide chain comprises in the N- to C-terminal direction:

(i) a domain (A) comprising SEQ ID NO: 553, 565, 567, 570, 574, 578, 582, 586, 590, or 594;

(ii) a domain (B) comprising SEQ ID NO: 500 or 504; and

(iii) a domain (C) comprising SEQ ID NO: 520, 521, 518, or 519;

(II) the second polypeptide chain comprises in the N- to C-terminal direction: (i) a domain (D) comprising SEQ ID NO: 502, or 506;

(ii) a domain (E) comprising SEQ ID NO: 551, 564, 566, 568, 572, 576, 580, 584, 588, 592; and

(iii) a domain (F) comprising SEQ ID NO: 518, 519, 520, or 521 ; and wherein:

the domains (B) and (D) associate to form a binding site that binds an epitope (2).

[0019] In certain aspects the invention provides such bispecific molecules wherein:

(i) domains (A) and (B) are separated by SEQ ID NO: 508;

(ii) domains (C) and (B) are separated by SEQ ID NO: 509 or 510;

(iii) domains (D) and (E) are separated by SEQ ID NO: 508; and

(iv) domains (F) and (E) are separated by SEQ ID NO: 509 or 510.

[0020] In certain aspects the invention provides bispecific molecules comprising a first polypeptide chain and a second polypeptide chain, covalently bonded to one another, wherein:

(I) the first polypeptide chain comprises in the N- to C-terminal direction:

(ii) a domain (B) comprising SEQ ID NO: 500 or 504, wherein domains (A) and (B) are separated from one another by SEQ ID NO: 508; and

(iii) a domain (C) comprising SEQ ID NO: 520, 521, 518, or 519; wherein domain (C) and domain (B) are separated by SEQ ID NO: 509 or 510;

(II) the second polypeptide chain comprises in the N- to C-terminal direction:

(i) a domain (D) comprising SEQ ID NO: 502, or 506;

(ii) a domain (E) comprising SEQ ID NO: 551, 564, 566, 568, 572, 576, 580, 584, 588, 592, wherein domains (D) and (E) are separated from one another by SEQ ID NO: 508; and (Hi) a domain (F) comprising SEQ ID NO: 518, 519, 520, or 521 ; wherein domain (F) and domain (E) are separated by SEQ ID NO: 509 or 510; and wherein:

wherein if the domain (A) comprises SEQ ID NO: 553 and domain (E) comprises SEQ ID NO: 551 they associate to form a binding site that binds the HIV-1 envelope like antibody DH557;

wherein if the domain (A) comprises SEQ ID NO: 565 and domain (E) comprises SEQ ID NO: 564 they associate to form a binding site that binds the HIV-1 envelope like antibody DH556;

wherein if the domain (A) comprises SEQ ID NO: 567 and domain (E) comprises SEQ ID NO: 566 they associate to form a binding site that binds the HIV-1 envelope like antibody DH555;

wherein if the domain (A) comprises SEQ ID NO: 570 and domain (E) comprises SEQ ID NO: 568 they associate to form a binding site that binds the HIV-1 envelope like antibody DH493;

wherein if the domain (A) comprises SEQ ID NO: 574 and domain (E) comprises SEQ ID NO: 572 they associate to form a binding site that binds the HIV-1 envelope like antibody DH492;

wherein if the domain (A) comprises SEQ ID NO: 578 and domain (E) comprises SEQ ID NO: 576 they associate to form a binding site that binds the HIV-1 envelope like antibody DH491 ;

wherein if the domain (A) comprises SEQ ID NO: 582 and domain (E) comprises SEQ ID NO: 580 they associate to form a binding site that binds the HIV-1 envelope like antibody DH490;

wherein if the domain (A) comprises SEQ ID NO: 586 and domain (E) comprises SEQ ID NO: 584 they associate to form a binding site that binds the HIV-1 envelope like antibody DH542;

wherein if the domain (A) comprises SEQ ID NO: 590 and domain (E) comprises SEQ ID NO: 588 they associate to form a binding site that binds the HIV-1 envelope like antibody DH542 QSA; or wherein if the domain (A) comprises SEQ ID NO: 594 and domain (E) comprises SEQ ID NO: 592 they associate to form a binding site that binds the HIV-1 envelope like antibody DH542 L4; and

wherein if the domain (B) comprises SEQ ID NO: 500 and domain (D) comprises SEQ ID NO: 502 they associate to form a binding site that binds CD3; or

wherein if the domain (B) comprises SEQ ID NO: 504 and domain (D) comprises SEQ ID NO: 506 they associate to form a binding site that binds CD16.

[0021] In certain aspects the invention provides a bispecific molecule comprising a first polypeptide chain, a second polypeptide chain, and a third polypeptide chain, wherein some of the polypeptides are covalently bonded, and wherein:

(I) the first polypeptide chain comprises in the N- to C-terminal direction:

(ii) a domain (B) comprising a binding region of a heavy chain variable domain of a second immunoglobulin (VH2) specific for an epitope (2);

(iii) a domain (C) comprising a heterodimer promoting domain; and

(iv) a CH2-CH3 domain;

(II) the second polypeptide chain comprises in the N- to C-terminal direction:

(i) a domain (D) comprising a binding region of a light chain variable domain of the second immunoglobulin (VL2) specific for the epitope (2);

(ii) a domain (E) comprising a binding region of a heavy chain variable domain of the first immunoglobulin (VH1) having the binding specificity of a CD4 binding site HIV-1 antibody or a V3 glycan binding antibody (1); and

(iii) a domain (F) comprising a heterodimer promoting domain;

(III) the third polypeptide chain comprises in the N- to C-terminal direction:

(i) a CH2-CH3 domain, and wherein:

the domains (D) and (E) are linked so that they do not associate with one another to form an epitope binding site;

the domains (A) and (E) associate to form a binding site that binds the HIV-1 envelope like a CD4 binding site HIV-1 antibody or like a V3 glycan binding antibody (1); the domains (B) and (D) associate to form a binding site that binds the epitope (2); and the CH2-CH3 domains of the first and third polypeptide form an Fc Domain.

[0022] In certain aspects the invention provides such bispecific molecules wherein:

(i) the third polypeptide chain further comprises a peptide linker 3 N-terminal to the

CH2-CH3 domain;

(ii) domains (A) and (B) are separated by a peptide linker 1 ;

(iii) domains (C) and (B) are separated by a peptide linker 2;

(iv) the CH2-CH3 domain and domain (C) are separated by a peptide linker 3 or a spacer linker 3;

(v) domains (D) and (E) are separated by a peptide linker 1 ; and

(vi) domains (F) and (E) are separated by a peptide linker 2.

[0023] In certain aspects the invention provides bispecific molecules comprising a first polypeptide chain, a second polypeptide chain, and a third polypeptide chain, wherein some of the polypeptides are covalently bonded (See Figure 3), and wherein:

(I) the first polypeptide chain comprises in the N- to C-terminal direction:

(i) a domain (A) comprising a binding region of the light chain variable domain of a first immunoglobulin (VLl) having the binding specificity of the V3 glycan binding antibody DH542, a variant of DH542 called DH542 QSA, DH542 L4 and/or other antibodies from the DH542 lineage, and/or the binding specificity of the CD4 binding site antibody CH557 (also referred to as CH235.12) or any of the antibodies of the CH235 lineage HIV- 1 antibody (1);

(ii) a domain (B) comprising a binding region of a heavy chain variable domain of a second immunoglobulin (VH2) specific for an epitope (2), wherein domains (A) and (B) are separated from one another by a peptide linker 1 ;

2;

(iv) a CH2-CH3 domain, wherein the CH2-CH3 domain and domain (C) are separated by a peptide linker 3 or a spacer-linker 3;

(II) the second polypeptide chain comprises in the N- to C-terminal direction:

(i) a domain (D) comprising a binding region of a light chain variable domain of the second immunoglobulin (VL2) specific for the epitope (2); (ii) a domain (E) comprising a binding region of a heavy chain variable domain of the first immunoglobulin (VH1) having the binding specificity of the V3 glycan binding antibody DH542, a variant of DH542 called DH542 QSA, DH542 L4 and/or other antibodies from the DH542 lineage, and/or the binding specificity of the CD4 binding site antibody CH557 (also referred to as CH235.12) or any of the antibodies of the CH235 lineage HIV- 1 antibody (1), wherein domains (D) and (E) are separated from one another by a peptide linker 1; and

(iii) a domain (F) comprising a heterodimer promoting domain, including but not limited to a K coil or E coil; wherein domain (F) and domain (E) are separated by a peptide linker

2;

(III) the third polypeptide chain comprises in the N- to C-terminal direction:

(i) a peptide linker 3,

(ii) a CH2-CH3 domain, and wherein:

the domains (A) and (B) do not associate with one another to form an epitope binding site; the domains (D) and (E) do not associate with one another to form an epitope binding site; the domains (A) and (E) associate to form a binding site that binds the HIV-1 envelope like the V3 glycan binding antibody DH542, a variant of DH542 called DH542 QSA, DH542 L4 and/or other antibodies from the DH542 lineage, and/or like the CD4 binding site antibody CH557 (also referred to as CH235.12) or any of the antibodies of the CH235 lineage HIV-1 antibody (1);

the domains (B) and (D) associate to form a binding site that binds the epitope (2); and the CH2-CH3 domains of the first and third polypeptide form an Fc Domain.

[0024] In certain aspects said first and second polypeptide chains are covalently bonded to one another; and said second and third polypeptide chains are covalently bonded to one another.

[0025] In certain aspects the invention provides a bispecific molecule comprising a first polypeptide chain, a second polypeptide chain, and a third polypeptide chain, wherein the first and second polypeptide chains are covalently bonded and the second and third polypeptide chains are covalently bonded, and wherein:

(I) the first polypeptide chain comprises in the N- to C-terminal direction:

(i) a domain (A) comprising SEQ ID NO: 553, 565, 567, 570, 574, 578, 582, 586, 590, or 594; (ii) a domain (B) comprising SEQ ID NO:500 or 504;

(iii) a domain (C) comprising SEQ ID NO: 520, 521 , 518, or 519; and

(iv) a CH2-CH3 domain comprising SEQ ID NO: 531, 532, 533, or 534;

(II) the second polypeptide chain comprises in the N- to C-terminal direction:

(i) a domain (D) comprising SEQ ID NO: 502, or 506;

(iii) a domain (F) comprising SEQ ID NO: 518, 519, 520, or 521;

(III) the third polypeptide chain comprises in the N- to C-terminal direction:

(i) a CH2-CH3 domain comprising SEQ ID NO: 531, 532, 533, or 534; and wherein: the domains (A) and (B) are linked so that they do not associate with one another to form an epitope binding site;

the domains (A) and (E) associate to form a binding site that binds the HIV-1 envelope like a

CD4 binding site HIV-1 antibody or like a V3 glycan binding antibody (1);

the domains (B) and (D) associate to form a binding site that binds an epitope (2); and the CH2-CH3 domains of the first and third polypeptide form an Fc Domain.

[0026] In certain aspects the invention provides such bispecific molecules wherein:

(i) the third polypeptide chain further comprises SEQ ID NO: 523 N-terminal to the

CH2-CH3 domain;

(ii) domains (A) and (B) are separated by SEQ ID NO: 508;

(iii) domains (C) and (B) are separated by SEQ ID NO: 509 or 510;

(iv) the CH2-CH3 domain and domain (C) are separated by SEQ ID NO: 523 or 522;

(v) domains (D) and (E) are separated by SEQ ID NO: 508; and

(vi) domains (F) and (E) are separated by SEQ ID NO: 509 or 510.

[0027] In certain aspects the invention provides bispecific molecules comprising a first polypeptide chain, a second polypeptide chain, and a third polypeptide chain, wherein the first and second polypeptide chains are covalently bonded and the second and third polypeptide chains are covalently bonded, and wherein:

(I) the first polypeptide chain comprises in the N- to C-terminal direction: (i) a domain (A) comprising SEQ ID NO: 553, 565, 567, 570, 574, 578, 582, 586, 590, or 594;

(ii) a domain (B) comprising SEQ ID NO: 500 or 504, wherein domains (A) and (B) are separated SEQ ID NO: 508;

(iv) a CH2-CH3 domain comprising SEQ ID NO: 531, 532, 533, or 534, wherein the CH2-CH3 domain and domain (C) are separated by SEQ ID NO: 523 or 522;

(II) the second polypeptide chain comprises in the N- to C-terminal direction:

(i) a domain (D) comprising SEQ ID NO: 502, or 506;

(ii) a domain (E) comprising SEQ ID NO: 551, 564, 566, 568, 572, 576, 580, 584, 588, 592, wherein domains (D) and (E) are separated from one another by SEQ ID NO: 508; and

(iii) a domain (F) comprising SEQ ID NO: 518, 519, 520, or 521; wherein domain (F) and domain (E) are separated by SEQ ID NO: 509 or 510;

(III) the third polypeptide chain comprises in the N- to C-terminal direction:

(i) SEQ ID NO: 523; and

(ii) a CH2-CH3 domain comprising SEQ ID NO: 531, 532, 533, or 534; and wherein: the domains (A) and (B) do not associate with one another to form an epitope binding site; the domains (D) and (E) do not associate with one another to form an epitope binding site; and

wherein if the domain (A) comprises SEQ ID NO: 570 and domain (E) comprises SEQ ID NO: 568 they associate to form a binding site that binds the HIV-1 envelope like antibody DH493; wherein if the domain (A) comprises SEQ ID NO: 574 and domain (E) comprises SEQ ID NO: 572 they associate to form a binding site that binds the HIV-1 envelope like antibody DH492;

wherein if the domain (A) comprises SEQ ID NO: 578 and domain (E) comprises SEQ ID NO: 576 they associate to form a binding site that binds the HIV-1 envelope like antibody DH491;

wherein if the domain (A) comprises SEQ ID NO: 590 and domain (E) comprises SEQ ID NO: 588 they associate to form a binding site that binds the HIV-1 envelope like antibody DH542 QSA; or

wherein if the domain (A) comprises SEQ ID NO: 594 and domain (E) comprises SEQ ID NO: 592 they associate to form a binding site that binds the HIV-1 envelope like antibody DH542 L4; and

wherein if the domain (B) comprises SEQ ID NO: 504 and domain (D) comprises SEQ ID NO: 506 they associate to form a binding site that binds CD16; and

the CH2-CH3 domains of the first and third polypeptide form an Fc Domain.

[0028] In certain aspects the invention provides bispecific molecules comprising a first polypeptide chain, a second polypeptide chain, and a third polypeptide chain, wherein some of the polypeptides are covalently bonded (See Figure 3), and wherein:

(I) the first polypeptide chain comprises in the N- to C-terminal direction:

(i) a CH2-CH3 domain;

(ii) a domain (A) comprising a binding region of the light chain variable domain of a first immunoglobulin (VLl) having the binding specificity of a CD4 binding site HIV- 1 antibody or a V3 glycan binding antibody (1); (iii) a domain (B) comprising a binding region of a heavy chain variable domain of a second immunoglobulin (VH2) specific for an epitope (2); and

(iv) a domain (C) comprising a heterodimer promoting domain;

(II) the second polypeptide chain comprises in the N- to C-terminal direction:

(ii) a domain (E) comprising a binding region of a heavy chain variable domain of the first immunoglobulin (VH1) having the binding specificity of a CD4 binding site HIV- 1 antibody or a V3 glycan binding antibody (1); and

(iii) a domain (F) comprising a heterodimer promoting domain;

(III) the third polypeptide chain comprises in the N- to C-terminal direction:

(i) a CH2-CH3 domain, and wherein:

the domains (A) and (E) associate to form a binding site that binds the HIV-1 envelope like a CD4 binding site HIV-1 antibody or like a V3 glycan binding antibody (1); the domains (B) and (D) associate to form a binding site that binds the epitope (2); and

the CH2-CH3 domains of the first and third polypeptide form an Fc Domain.

[0029] In certain aspects the invention provides such bispecific molecules wherein:

(i) the CH2-CH3 domain and domain (A) are separated by a peptide linker 4;

(ii) domains (A) and (B) are separated by a peptide linker 1;

(iii) domains (C) and (B) are separated by a peptide linker 2;

(iv) domains (D) and (E) are separated by a peptide linker 1;

(v) domains (F) and (E) are separated by a peptide linker 2;

(vi) the first polypeptide chain further comprises a peptide linker 3 N-terminal to the CH2- CH3 domain; and

(vii) the third polypeptide chain further comprises a peptide linker 3 N-terminal to the

CH2-CH3 domain. [0030] In other aspects the invention provides bispecific molecules comprising a first polypeptide chain, a second polypeptide chain, and a third polypeptide chain, wherein some of the polypeptides are covalently bonded (See Figure 3), and wherein:

(I) the first polypeptide chain comprises in the N- to C-terminal direction:

(i) a peptide linker 3 followed by a CH2-CH3 domain;

(ii) a domain (A) comprising a binding region of the light chain variable domain of a first immunoglobulin (VL1) having the binding specificity of the V3 glycan binding antibody DH542, a variant of DH542 called DH542 QSA, DH542 L4 and/or other antibodies from the DH542 lineage, and/or the binding specificity of the CD4 binding site antibody CH557 (also referred to as CH235.12) or any of the antibodies of the CH235 lineage (1), wherein the CH2-CH3 domain and domain (A) are separated by a peptide linker 4;

(iii) a domain (B) comprising a binding region of a heavy chain variable domain of a second immunoglobulin (VH2) specific for an epitope (2), wherein domains (A) and (B) are separated by a peptide linker 1;

(iv) a domain (C) comprising a heterodimer promoting domain, including but not limited to a K coil or E coil; wherein domain (C) and domain (B) are separated by a peptide linker 2;

(II) the second polypeptide chain comprises in the N- to C-terminal direction:

(ii) a domain (E) comprising a binding region of a heavy chain variable domain of the first immunoglobulin (VH1) having the binding specificity of the V3 glycan binding antibody DH542, a variant of DH542 called DH542 QSA, DH542-L4 and/or other antibodies from the DH542 lineage, and/or the binding specificity of the CD4 binding site antibody CH557 (also referred to as CH235.12) or any of the antibodies of the CH235 lineage (1), wherein domains (D) and (E) are separated by a peptide linker 1; and

(iii) a domain (F) comprising a heterodimer promoting domain, including but not limited to a K coil or E coil; wherein domain (F) and domain (E) are separated by a peptide linker 2;

(III) the third polypeptide chain comprises in the N- to C-terminal direction:

(i) a peptide linker 3, (ii) a CH2-CH3 domain, and wherein:

the domains (A) and (B) do not associate with one another to form an epitope binding site; the domains (D) and (E) do not associate with one another to form an epitope binding site; the domains (A) and (E) associate to form a binding site that binds the HIV-1 envelope like the V3 glycan binding antibody DH542, a variant of DH542 called DH542 QSA, DH542 L4 and/or other antibodies from the DH542 lineage, and/or like the CD4 binding site antibody CH557 (also referred to as CH235.12) or any of the antibodies of the CH235 lineage (1); the domains (B) and (D) associate to form a binding site that binds the epitope (2); and the CH2-CH3 domains of the first and third polypeptide form an Fc Domain.

[0031] In certain aspects said first and second polypeptide chains are covalently bonded to one another; and said first and third polypeptide chains are covalently bonded to one another.

[0032] In certain aspects the invention provides bispecific molecules comprising a first polypeptide chain, a second polypeptide chain, and a third polypeptide chain, wherein the first and second polypeptide chains are covalently bonded and the first and third polypeptide chains are covalently bonded (See Figure 3), and wherein:

(I) the first polypeptide chain comprises in the N- to C-terminal direction:

(i) a CH2-CH3 domain comprising SEQ ID NO: 531, 532, 533, or 534;

(ii) a domain (A) comprising SEQ ID NO: 553, 565, 567, 570, 574, 578, 582, 586, 590, or 594;

(iii) a domain (B) comprising SEQ ID NO:500 or 504; and

(iv) a domain (C) comprising SEQ ID NO: 520, 521, 518, or 519;

(II) the second polypeptide chain comprises in the N- to C-terminal direction:

(i) a domain (D) comprising SEQ ID NO: 502, or 506;

(iii) a domain (F) comprising SEQ ID NO: 518, 519, 520, or 521;

(III) the third polypeptide chain comprises in the N- to C-terminal direction:

(i) a CH2-CH3 domain comprising SEQ ID NO: 531, 532, 533, or 534, and wherein: the domains (A) and (B) are linked so that they do not associate with one another to form an epitope binding site;

the domains (D) and (E) are linked so that they do not associate with one another to form an epitope binding site; the domains (A) and (E) associate to form a binding site that binds the HIV-1 envelope like a CD4 binding site HIV-1 antibody or like a V3 glycan binding antibody (1); the domains (B) and (D) associate to form a binding site that binds an epitope (2); and

the CH2-CH3 domains of the first and third polypeptide form an Fc Domain.

[0033] In certain aspects the invention provides such bispecific molecules wherein:

(i) the CH2-CH3 domain and domain (A) are separated by SEQ ID NO: 524 or 525;

(ii) domains (A) and (B) are separated by SEQ ID NO: 508;

(iii) domains (C) and (B) are separated by SEQ ID NO: 509 or 510;

(iv) domains (D) and (E) are separated by SEQ ID NO: 508;

(v) domains (F) and (E) are separated by SEQ ID NO: 509 or 510;

(vi) the first polypeptide chain further comprises SEQ ID NO: 523 N-terminal to the CH2- CH3 domain; and

(vii) the third polypeptide chain further comprises SEQ ID NO: 523 N-terminal to the CH2-CH3 domain.

[0034] In other aspects the invention provides bispecific molecules comprising a first polypeptide chain, a second polypeptide chain, and a third polypeptide chain, wherein the first and second polypeptide chains are covalently bonded and the first and third polypeptide chains are covalently bonded (See Figure 3), and wherein:

(I) the first polypeptide chain comprises in the N- to C-terminal direction:

(i) SEQ ID NO: 523 followed by a CH2-CH3 domain comprising SEQ ID NO: 531, 532, 533, or 534;

(ii) a domain (A) comprising SEQ ID NO: 553, 565, 567, 570, 574, 578, 582, 586, 590, or 594, wherein the CH2-CH3 domain and domain (A) are separated SEQ ID NO: 524 or 525;

(iii) a domain (B) comprising SEQ ID NO: 500 or 504, wherein domains (A) and (B) are separated by SEQ ID NO: 508;

(iv) a domain (C) comprising SEQ ID NO: 520, 521, 518, or 519; wherein domain (C) and domain (B) are separated by SEQ ID NO: 509 or 510;

(II) the second polypeptide chain comprises in the N- to C-terminal direction:

(i) a domain (D) comprising SEQ ID NO: 502, or 506;

(ii) a domain (E) comprising SEQ ID NO: 551, 564, 566, 568, 572, 576, 580, 584, 588, 592, wherein domains (D) and (E) are separated by SEQ ID NO: 508; and (Hi) a domain (F) comprising SEQ ID NO: 518, 519, 520, or 521; wherein domain (F) and domain (E) are separated by SEQ ID NO: 509 or 510;

(III) the third polypeptide chain comprises in the N- to C-terminal direction:

(i) SEQ ID NO: 523,

(ii) a CH2-CH3 domain comprising SEQ ID NO: 531, 532, 533, or 534, and wherein: the domains (A) and (B) do not associate with one another to form an epitope binding site; the domains (D) and (E) do not associate with one another to form an epitope binding site; wherein if the domain (A) comprises SEQ ID NO: 553 and domain (E) comprises SEQ ID NO: 551 they associate to form a binding site that binds the HIV-1 envelope like antibody DH557;

wherein if the domain (A) comprises SEQ ID NO: 586 and domain (E) comprises SEQ ID NO: 584 they associate to form a binding site that binds the HIV-1 envelope like antibody DH542; wherein if the domain (A) comprises SEQ ID NO: 590 and domain (E) comprises SEQ ID NO: 588 they associate to form a binding site that binds the HIV-1 envelope like antibody DH542 QSA; or

the CH2-CH3 domains of the first and third polypeptide form an Fc Domain.

[0035] In certain aspects the invention provides a bispecific molecule comprising a first polypeptide chain, a second polypeptide chain, a third polypeptide chain, and a fourth polypeptide chain, wherein some of the polypeptides are covalently bonded, and wherein:

(I) the first and the third polypeptide chains each comprise in the N- to C-terminal direction:

(iii) a domain (C) comprising a heterodimer promoting domain; and

(iv) a CH2-CH3 domain;

(II) the second and fourth polypeptide chains each comprise in the N- to C-terminal direction:

(ii) a domain (E) comprising a binding region of a heavy chain variable domain of the first immunoglobulin (VH1) having the binding specificity of a CD4 binding site HIV-1 antibody or a V3 glycan binding antibody (1);

(iii) a domain (F) comprising a heterodimer promoting domain; and wherein

the domains (A) and (B) are linked so that they do not associate with one another to form an epitope binding site; the domains (D) and (E) are linked so that they do not associate with one another to form an epitope binding site;

CD4 binding site HIV-1 antibody or like a V3 glycan binding antibody (1);

[0036] In certain aspects the invention provides such bispecific molecules wherein:

(i) domains (A) and (B) are separated by a peptide linker 1 ;

(ii) domains (C) and (B) are separated by a peptide linker 2;

(iii) the CH2-CH3 domain and domain (C) are separated by a peptide linker 3 or a spacer linker 3;

(iv) domains (D) and (E) are separated by a peptide linker 1 ; and

(v) domains (F) and (E) are separated by a peptide linker 2.

[0037] In certain aspects the invention provides a bispecific molecule comprising a first polypeptide chain, a second polypeptide chain, a third polypeptide chain, and a fourth polypeptide chain, wherein some of the polypeptides are covalently bonded, and wherein:

(i) a domain (A) comprising a binding region of the light chain variable domain of a first immunoglobulin (VLl) having the binding specificity of the V3 glycan binding antibody DH542, a variant of DH542 called DH542 QSA, DH542 L4 and/or other antibodies from the DH542 lineage, and/or the binding specificity of the CD4 binding site antibody CH557 (also referred to as CH235.12) or any of the antibodies of the CH235 lineage (1);

(ii) a domain (B) comprising a binding region of a heavy chain variable domain of a second immunoglobulin (VH2) specific for an epitope (2), wherein domains (A) and (B) are separated by a peptide linker 1 ;

(iii) a domain (C) comprising a heterodimer promoting domain, wherein domains (C) and (B) are separated by a peptide linker 2; and

(iv) a CH2-CH3 domain, wherein the CH2-CH3 domain and domain (C) are separated by a peptide linker 3 or a spacer linker 3;

(i) a domain (D) comprising a binding region of a light chain variable domain of the second immunoglobulin (VL2) specific for the epitope (2); (ii) a domain (E) comprising a binding region of a heavy chain variable domain of the first immunoglobulin (VH1) having the binding specificity of the V3 glycan binding antibody DH542, a variant of DH542 called DH542 QSA, DH542 L4 and/or other antibodies from the DH542 lineage, and/or the binding specificity of the CD4 binding site antibody CH557 (also referred to as CH235.12) or any of the antibodies of the CH235 lineage (1), wherein domains (D) and (E) are separated by a peptide linker 1 ;

(iii) a domain (F) comprising a heterodimer promoting domain, wherein domains (F) and (E) are separated by a peptide linker 2; and wherein

[0038] In certain aspects said first and second polypeptide chains are covalently bonded to one another; said third and fourth polypeptide chains are covalently bonded to one another; and said first and third chains are covalently bonded to one another.

[0039] In certain aspects the invention provides a bispecific molecule comprising a first polypeptide chain, a second polypeptide chain, a third polypeptide chain, and a fourth polypeptide chain, wherein the first and second polypeptide chains are covalently bonded, the third and fourth polypeptide chains are covalently bonded, and the first and third chains are covalently bonded, and wherein:

(ii) a domain (B) comprising SEQ ID NO:500 or 504;

(iii) a domain (C) comprising SEQ ID NO: 520, 521, 518, or 519; and

(iv) a CH2-CH3 domain comprising SEQ ID NO: 527, 528, or 529;

(i) a domain (D) comprising SEQ ID NO: 502, or 506; (ii) a domain (E) comprising SEQ ID NO: 551, 564, 566, 568, 572, 576, 580, 584, 588, 592;

(iii) a domain (F) comprising SEQ ID NO: 518, 519, 520, or 521; and wherein the domains (A) and (B) are linked so that they do not associate with one another to form an epitope binding site;

the domains (A) and (E) associate to form a binding site that binds the HIV-1 envelope like a CD4 binding site HIV-1 antibody or like a V3 glycan binding antibody (1);

[0040] In certain aspects the invention provides such bispecific molecules wherein:

(i) domains (A) and (B) are separated by SEQ ID NO: 508;

(ii) domains (C) and (B) are separated by SEQ ID NO: 509 or 510;

(iii) the CH2-CH3 domain and domain (C) are separated by SEQ ID NO: 523 or 522;

(iv) domains (D) and (E) are separated by SEQ ID NO: 508; and

(v) domains (F) and (E) are separated by SEQ ID NO: 509 or 510.

[0041] In certain aspects the invention provides a bispecific molecule comprising a first polypeptide chain, a second polypeptide chain, a third polypeptide chain, and a fourth polypeptide chain, wherein the first and second polypeptide chains are covalently bonded, the third and fourth polypeptide chains are covalently bonded, and the first and third chains are covalently bonded, and wherein:

(ii) a domain (B) comprising SEQ ID NO: 500 or 504, wherein domains (A) and (B) are separated by SEQ ID NO: 508;

(iii) a domain (C) comprising SEQ ID NO: 520, 521, 518, or 519, wherein domains (C) and (B) are separated by SEQ ID NO: 509 or 510; and

(iv) a CH2-CH3 domain comprising SEQ ID NO: 527, 528, or 529, wherein the CH2- CH3 domain and domain (C) are separated by SEQ ID NO: 523 or 522; (II) the second and fourth polypeptide chains each comprise in the N- to C-terminal direction:

(i) a domain (D) comprising SEQ ID NO: 502, or 506;

(ii) a domain (E) comprising SEQ ID NO: 551, 564, 566, 568, 572, 576, 580, 584, 588, 592, wherein domains (D) and (E) are separated by SEQ ID NO: 508;

(iii) a domain (F) comprising SEQ ID NO: 518, 519, 520, or 521, wherein domains (F) and (E) are separated by SEQ ID NO: 509 or 510; and wherein

the domains (A) and (B) do not associate with one another to form an epitope binding site; the domains (D) and (E) do not associate with one another to form an epitope binding site; wherein if the domain (A) comprises SEQ ID NO: 553 and domain (E) comprises SEQ ID NO: 551 they associate to form a binding site that binds the HIV-1 envelope like antibody DH557;

the CH2-CH3 domains of the first and third polypeptide form an Fc Domain.

[0042] In certain aspects the invention provides trivalent binding molecules comprising a first, second, third and fourth polypeptide chain, wherein some of the polypeptides are covalently bonded and wherein:

(I) the first polypeptide chain comprises in the N-terminus to C-terminus direction:

(i) a domain (A) comprising a binding region of a light chain variable domain of a first immunoglobulin (VLl) specific for an epitope (1);

(iii) a domain (C) comprising a heterodimer promoting domain; and

(iv) a CH2-CH3 Domain;

(II) the second polypeptide chain comprises, in the N-terminus to C-terminus direction:

(i) a domain (D) comprising a binding region of a light chain variable domain of the first immunoglobulin (VLl) specific for the epitope (2);

(ii) a domain (E) comprising a binding region of a heavy chain variable domain of the second immunoglobulin (VH1) specific for the epitope (1); and

(iii) a domain (F) comprising a heterodimer promoting domain;

(III) the third polypeptide chain that comprises, in the N-terminus to C-terminus direction:

(i) a domain (G) comprising a binding region of a heavy chain variable domain of a third immunoglobulin (VH3) specific for an epitope (3); and

(ii) a CHI -Hinge Domain, and a CH2-CH3 Domain; and

(IV) the fourth polypeptide chain that comprises, in the N-terminus to C-terminus direction: (i) a domain (H) comprising a binding region of a light chain variable domain of the third immunoglobulin (VL3) specific for the epitope (3); and

(ii) CL Kappa Domain or a CL Lambda Domain; and wherein

the domains (A) and (E) associate to form a binding site that binds the epitope (1);

the domains (B) and (D) associate to form a binding site that binds the epitope (2);

the domains (G) and (H) associate to form a binding site that bind the epitope (3);

at least one of epitope (1), epitope (2), and epitope (3) is an epitope bound by a CD4 binding site HIV-1 antibody or a V3 glycan binding antibody, and at least one of epitope (1), epitope

(2), and epitope (3) is an epitope of, for example, but not limited to, CD3, CD8, or CD16, or an epitope on any suitable effector cell;

the CH2-CH3 domains of the first and third polypeptide form an Fc Domain;

said first and second polypeptide chains are covalently bonded to one another;

said first and third polypeptide chains are covalently bonded to one another; and

said third and fourth polypeptide chains are covalently bonded to one another.

[0043] In certain aspects the invention provides such trivalent molecules wherein:

(i) domains (A) and (B) are separated by a peptide linker 1;

(ii) domains (C) and (B) are separated by a peptide linker 2 or a peptide linker 2-C;

(iv) domains (D) and (E) are separated by a peptide linker 1; and

(v) domains (F) and (E) are separated by a peptide linker 2 or a peptide linker 2-C.

[0044] In certain aspects the invention provides trivalent binding molecules comprising a first, second, third and fourth polypeptide chain, wherein some of the polypeptides are covalently bonded and wherein:

(i) a domain (A) comprising a binding region of a light chain variable domain of a first immunoglobulin (VLl) specific for an epitope (1); (ii) a domain (B) comprising a binding region of a heavy chain variable domain of a second immunoglobulin (VH2) specific for an epitope (2), wherein domains (A) and (B) are separated by a Peptide Linker 1;

(iii) a domain (C) comprising:

(a) a heterodimer promoting domain; wherein domain (C) and domain (B) are separated by a Peptide Linker 2-C; or

(b) a heterodimer promoting domain; wherein domain (C) and domain (B) are separated by a Peptide Linker 2; and

(i) a domain (D) comprising a binding region of a light chain variable domain of the first immunoglobulin (VL1) specific for the epitope (2);

(ii) a domain (E) comprising a binding region of a heavy chain variable domain of the second immunoglobulin (VH1) specific for the epitope (1), wherein domains (D) and (E) are separated by a Peptide Linker 1;

(iii) a domain (F) comprising

(a) a heterodimer promoting domain; wherein domain (F) and domain (E) are separated by a peptide Peptide Linker 2-C; or

(b) a heterodimer promoting domain; wherein domain (F) and domain (E) are separated by a peptide Peptide Linker 2;

(ii) a CHI -Hinge Domain, and a CH2-CH3 Domain; and

(IV) the fourth polypeptide chain that comprises, in the N-terminus to C-terminus direction:

(i) a domain (H) comprising a binding region of a light chain variable domain of the third immunoglobulin (VL3) specific for the epitope (3); and

(ii) CL Kappa Domain or a CL Lambda Domain; and wherein

the domains (A) and (B) do not associate with one another to form an epitope binding site; the domains (D) and (E) do not associate with one another to form an epitope binding site; the domains (A) and (E) associate to form a binding site that binds the epitope (1);

the domains (B) and (D) associate to form a binding site that binds the epitope (2); the domains (G) and (H) associate to form a binding site that bind the epitope (3); at least one of epitope (1), epitope (2), and epitope (3) is an epitope bound by the V3 glycan binding antibody DH542, a variant of DH542 called DH542 QSA, DH542 L4 and/or other antibodies from the DH542 lineage, and/or the CD4 binding site antibody CH557 (also referred to as CH235.12) or any of the antibodies of the CH235 lineage, and at least one of epitope (1), epitope (2), and epitope (3) is an epitope of for example, but not limited to, CD3,

CD8, or CD 16, or an epitope on any suitable effector cell;

the CH2-CH3 domains of the first and third polypeptide form an Fc Domain;

said first and second polypeptide chains are covalently bonded to one another;

said third and fourth polypeptide chains are covalently bonded to one another.

[0045] In certain aspects the invention provides trivalent binding molecules comprising a first, second, third and fourth polypeptide chain, wherein some of the polypeptides are covalently bonded and wherein:

(ii) a domain (B) comprising SEQ ID NO:500;

(iii) a domain (C) comprising SEQ ID NO: 520, 521, 518, or 519; and

(iv) a CH2-CH3 Domain comprising SEQ ID NO: 531, 532, 533, or 534;

(i) a domain (D) comprising SEQ ID NO: 502;

(iii) a domain (F) comprising SEQ ID NO: 518, 519, 520, or 521;

(i) a domain (G) comprising SEQ ID NO: 551, 564, 566, 568, 572, 576, 580, 584, 588, 592, 543, or 547; and

(ii) a CHI -Hinge Domain comprising SEQ ID NO: 515, and a CH2-CH3 Domain comprising SEQ ID NO: 531, 532, 533, or 534; and

(IV) the fourth polypeptide chain that comprises, in the N-terminus to C-terminus direction: (i) a domain (H) comprising SEQ ID NO: 553, 565, 567, 570, 574, 578, 582, 586, 590, 594, 545, or 549; and

(ii) CL Kappa Domain comprising SEQ ID NO: 516 or a CL Lambda Domain comprising SEQ ID NO: 517; and wherein

the domains (A) and (E) associate to form a binding site that binds an epitope (1);

the domains (B) and (D) associate to form a binding site that binds an epitope (2);

the domains (G) and (H) associate to form a binding site that bind an epitope (3);

wherein epitope (1) is an epitope bound by a CD4 binding site HIV-1 antibody or a V3 glycan binding antibody, epitope (2) is an epitope of CD3, and epitope (3) is an epitiope bound by a

CD4 binding site HIV-1 antibody or a V3 glycan binding antibody or is an epitope of CD8; the CH2-CH3 domains of the first and third polypeptide form an Fc Domain;

said first and second polypeptide chains are covalently bonded to one another;

said third and fourth polypeptide chains are covalently bonded to one another.

[0046] In certain aspects the invention provides such trivalent molecules wherein:

(i) domains (A) and (B) are separated by SEQ ID NO: 508;

(ii) domains (C) and (B) are separated by SEQ ID NO: 509 or 510;

(iii) the CH2-CH3 domain and domain (C) are separated by SEQ ID NO: 522, or 523;

(iv) domains (D) and (E) are separated by SEQ ID NO: 508; and

(v) domains (F) and (E) are separated by SEQ ID NO: 509 or 510.

[0047] In certain aspects the invention provides trivalent binding molecules comprising a first, second, third and fourth polypeptide chain, wherein some of the polypeptides are covalently bonded and wherein:

(i) a domain (A) comprising SEQ ID NO: 553, 565, 567, 570, 574, 578, 582, 586, 590, or 594; (ii) a domain (B) comprising SEQ ID NO:500, wherein domains (A) and (B) are separated by SEQ ID NO: 508;

(iii) a domain (C) comprising SEQ ID NO: 520, 521, 518, or 519, wherein domains (C) and (B) are separated by SEQ ID NO: 509 or 510;

(iv) a CH2-CH3 Domain comprising SEQ ID NO: 531, 532, 533, or 534, wherein the CH2-CH3 domain and domain (C) are separated by SEQ ID NO: 522, or 523;

(i) a domain (D) SEQ ID NO: 502;

(ii) a domain (E) comprising SEQ ID NO: 551, 564, 566, 568, 572, 576, 580, 584, 588, 592, wherein domains (D) and (E) are separated from one another by SEQ ID NO: 508;

(iii) a domain (F) comprising SEQ ID NO: 518, 519, 520, or 521, wherein domain (F) and domain (E) are separated by SEQ ID NO: 509 or 510;

(ii) a CHI -Hinge Domain SEQ ID NO: 515, and a CH2-CH3 Domain comprising SEQ ID NO: 531, 532, 533, or 534; and

(i) a domain (H) comprising SEQ ID NO: 553, 565, 567, 570, 574, 578, 582, 586, 590, 594, 545, or 549; and

the domains (A) and (B) do not associate with one another to form an epitope binding site; the domains (D) and (E) do not associate with one another to form an epitope binding site; the domains (A) and (E) associate to form a binding site that binds an epitope (1);

wherein if the domain (A) comprises SEQ ID NO: 553 and domain (E) comprises SEQ ID NO: 551 they associate to form a binding site that binds the HIV-1 envelope like antibody DH557; wherein if the domain (A) comprises SEQ ID NO: 565 and domain (E) comprises SEQ ID NO: 564 they associate to form a binding site that binds the HIV-1 envelope like antibody DH556;

wherein if the domain (H) comprises SEQ ID NO: 553 and domain (G) comprises SEQ ID NO: 551 they associate to form a binding site that binds the HIV-1 envelope like antibody DH557; wherein if the domain (H) comprises SEQ ID NO: 565 and domain (G) comprises SEQ ID NO: 564 they associate to form a binding site that binds the HIV-1 envelope like antibody DH556;

wherein if the domain (H) comprises SEQ ID NO: 567 and domain (G) comprises SEQ ID NO: 566 they associate to form a binding site that binds the HIV-1 envelope like antibody DH555;

wherein if the domain (H) comprises SEQ ID NO: 570 and domain (G) comprises SEQ ID NO: 568 they associate to form a binding site that binds the HIV-1 envelope like antibody DH493;

wherein if the domain (H) comprises SEQ ID NO: 574 and domain (G) comprises SEQ ID NO: 572 they associate to form a binding site that binds the HIV-1 envelope like antibody DH492;

wherein if the domain (H) comprises SEQ ID NO: 578 and domain (G) comprises SEQ ID NO: 576 they associate to form a binding site that binds the HIV-1 envelope like antibody DH491;

wherein if the domain (H) comprises SEQ ID NO: 582 and domain (G) comprises SEQ ID NO: 580 they associate to form a binding site that binds the HIV-1 envelope like antibody DH490;

wherein if the domain (H) comprises SEQ ID NO: 586 and domain (G) comprises SEQ ID NO: 584 they associate to form a binding site that binds the HIV-1 envelope like antibody DH542;

wherein if the domain (H) comprises SEQ ID NO: 590 and domain (G) comprises SEQ ID NO: 588 they associate to form a binding site that binds the HIV-1 envelope like antibody DH542 QSA; or

wherein if the domain (H) comprises SEQ ID NO: 594 and domain (G) comprises SEQ ID NO: 592 they associate to form a binding site that binds the HIV-1 envelope like antibody DH542 L4;

wherein if the domain (H) comprises SEQ ID NO: 545 and domain (G) comprises SEQ ID NO: 543 they associate to form a binding site that binds CD8; or

wherein if the domain (H) comprises SEQ ID NO: 549 and domain (G) comprises SEQ ID NO: 547 they associate to form a binding site that binds CD8; and wherein

epitope (1) is an epitope bound by the antibody CH557, CH556, CH555, CH493, CH492, CH491, CH490, DH542, DH542 QSA, or DH542 L4, epitope (2) is an epitope of CD3, and epitope (3) is is an epitope bound by the antibody CH557, CH556, CH555, CH493, CH492, CH491, CH490, DH542, DH542 QSA, or DH542 L4 or is an epitope of CD8; and wherein

the CH2-CH3 domains of the first and third polypeptide form an Fc Domain;

said first and second polypeptide chains are covalently bonded to one another;

said third and fourth polypeptide chains are covalently bonded to one another.

[0048] In certain aspects the invention also provides trivalent binding molecules comprising a first, second, and third polypeptide chain, wherein some of the polypeptides are covalently bonded and wherein:

(i) a domain (A) comprising a binding region of a light chain variable domain of a first immunoglobulin (VL1) specific for an epitope (1);

(iii) a domain (C) comprising a heterodimer promoting domain; and

(iv) a CH2-CH3 Domain;

(ii) a domain (E) comprising a binding region of a heavy chain variable domain of the second immunoglobulin (VH1) specific for the epitope (1);

(iii) a domain (F) comprising a heterodimer promoting domain; and

(i) a domain (H) comprising a binding region of a light chain variable domain of a third immunoglobulin (VL3) specific for an epitope (3)

(ii) a domain (G) comprising a binding region of a heavy chain variable domain of the third immunoglobulin (VH3) specific for the epitope (3);

(iii) a CH2-CH3 Domain; and wherein

at least one of epitope (1), epitope (2), and epitope (3) is an epitope bound by a CD4 binding site HIV-1 antibody or a V3 glycan antibody, and at least one of epitope (1), epitope (2), and epitope (3) is and epitope of for example, but not limited to, CD3, CD8, or CD16, or any other suitable epitope on an effector cell ;

the CH2-CH3 domains of the first and third polypeptide form an Fc Domain;

said first and second polypeptide chains are covalently bonded to one another;

said third and fourth polypeptide chains are covalently bonded to one another.

[0049] In certain aspects the invention provides such trivalent molecules wherein:

(i) domains (A) and (B) are separated by a peptide linker 1;

(iv) domains (D) and (E) are separated by a peptide linker 1;

(v) domains (F) and (E) are separated by a peptide linker 2 or a peptide linker 2-C;

(vi) domains (H) and (G) are separated by a peptide linker 5; and

(vii) the CH2-CH3 domain and domain (G) are separated by a peptide linker 3.

[0050] In certain aspects the invention also provides trivalent binding molecules comprising a first, second, and third polypeptide chain, wherein some of the polypeptides are covalently bonded and wherein:

(ii) a domain (B) comprising a binding region of a heavy chain variable domain of a second immunoglobulin (VH2) specific for an epitope (2), wherein domains (A) and (B) are separated by a Peptide Linker 1;

(iii) a domain (C) comprising: (a) a heterodimer promoting domain; wherein domain (C) and domain (B) are separated by a Peptide Linker 2-C; or

(iii) a domain (F) comprising:

(ii) a domain (G) comprising a binding region of a heavy chain variable domain of the third immunoglobulin (VH3) specific for the epitope (3), wherein domains (H) and (G) are separated by a Peptide Linker 5;

(iii) a Peptide Linker 3; and

(iv) a CH2-CH3 Domain; and wherein

at least one of epitope (1), epitope (2), and epitope (3) is an epitope bound by the V3 glycan binding antibody DH542, a variant of DH542 called DH542 QSA, DH542 L4 and/or other antibodies from the DH542 lineage, and/or the CD4 binding site antibody CH557 (also referred to as CH235.12) or any of the antibodies of the CH235 lineage, and at least one of epitope (1), epitope (2), and epitope (3) is and epitope of for example, but not limited to, CD3,

CD8 or CD 16, or an epitope on any suitable effector cell;

the CH2-CH3 domains of the first and third polypeptide form an Fc Domain;

said first and second polypeptide chains are covalently bonded to one another; and

said first and third polypeptide chains are covalently bonded to one another.

[0051] In certain aspects the invention also provides trivalent binding molecules comprising a first, second, and third polypeptide chain, wherein some of the polypeptides are covalently bonded and wherein:

(ii) a domain (B) comprising SEQ ID NO:500;

(iii) a domain (C) comprising SEQ ID NO: 520, 521, 518, or 519; and

(iv) a CH2-CH3 Domain comprising SEQ ID NO: 531, 532, 533, or 534;

(i) a domain (D) comprising SEQ ID NO: 502;

(ii) a domain (E) comprising SEQ ID NO: 551, 564, 566, 568, 572, 576, 580, 584, 588, or 592;

(iii) a domain (F) comprising SEQ ID NO: 518, 519, 520, or 521; and

(i) a domain (H) comprising SEQ ID NO: 545 or 549;

(ii) a domain (G) comprising SEQ ID NO: 543 or 547;

(iii) a CH2-CH3 Domain comprising SEQ ID NO: 531, 532, 533, or 534; and wherein

the domains (A) and (E) associate to form a binding site that binds an epitope bound by a CD4 binding site HIV-1 antibody or a V3 glycan antibody; the domains (B) and (D) associate to form a binding site that binds CD3;

the domains (G) and (H) associate to form a binding site that binds CD8; wherein

the CH2-CH3 domains of the first and third polypeptide form an Fc Domain;

said first and second polypeptide chains are covalently bonded to one another;

said third and fourth polypeptide chains are covalently bonded to one another.

[0052] In certain aspects the invention provides such trivalent molecules wherein:

(i) domains (A) and (B) are separated by SEQ ID NO: 508;

(ii) domains (C) and (B) are separated by SEQ ID NO: 509 or 510;

(iv) domains (D) and (E) are separated by SEQ ID NO: 508;

(v) domains (F) and (E) are separated by SEQ ID NO: 509 or 510;

(vi) domains (H) and (G) are separated by SEQ ID NO: 526; and

(vii) the CH2-CH3 domain and domain (G) are separated by SEQ ID NO: 523 or a CHI- Hinge Domain comprising SEQ ID NO: 515.

[0053] In certain aspects the invention also provides trivalent binding molecules comprising a first, second, and third polypeptide chain, wherein some of the polypeptides are covalently bonded and wherein:

(ii) a domain (B) comprising SEQ ID NO:500, wherein domains (A) and (B) are separated by SEQ ID NO: 508;

(iii) a domain (C) comprising SEQ ID NO: 520, 521, 518, and 519, wherein domains (C) and (B) are separated by SEQ ID NO: 509 or 510; and

(i) a domain (D) comprising SEQ ID NO: 502;

(ii) a domain (E) comprising SEQ ID NO: 551, 564, 566, 568, 572, 576, 580, 584, 588, or 592, wherein domains (D) and (E) are separated by SEQ ID NO: 508; (Hi) a domain (F) comprising comprising SEQ ID NO: 518, 519, 520, or 521, wherein domains (F) and (E) are separated by SEQ ID NO: 509 or 510;

(i) a domain (H) comprising SEQ ID NO: 545 or 549;

(ii) a domain (G) comprising SEQ ID NO: 543 or 547, wherein domains (H) and (G) are separated by SEQ ID NO: 526;

(iii) SEQ ID NO: 523, or a CHI-Hinge Domain comprising SEQ ID NO: 515; and

(iv) a CH2-CH3 Domain comprising SEQ ID NO: 531, 532, 533, or 534; and wherein

the domains (B) and (D) associate to form a binding site that binds CD3;

the domains (G) and (H) associate to form a binding site that binds CD8; and

wherein if the domain (A) comprises SEQ ID NO: 578 and domain (E) comprises SEQ ID NO: 576 they associate to form a binding site that binds the HIV-1 envelope like antibody DH491; wherein if the domain (A) comprises SEQ ID NO: 582 and domain (E) comprises SEQ ID NO: 580 they associate to form a binding site that binds the HIV-1 envelope like antibody DH490;

wherein if the domain (H) comprises SEQ ID NO: 545 and domain (G) comprises SEQ ID NO: 543 they associate to form a binding site that binds CDS; or

wherein if the domain (H) comprises SEQ ID NO: 549 and domain (G) comprises SEQ ID NO: 547 they associate to form a binding site that binds CDS;

wherein the CH2-CH3 domains of the first and third polypeptide form an Fc Domain;

said first and second polypeptide chains are covalently bonded to one another; and said first and third polypeptide chains are covalently bonded to one another.

[0054] In certain aspects the invention provides such trivalent binding molecules wherein one of epitope (1), epitope (2), and epitope (3) is an epitope of HIV-1 Envelope, one of epitope (1), epitope (2), and epitope (3) is an epitope of CD3, and one of epitope (1), epitope (2), and epitope (3) is an epitope of CD8. In particular aspects of such trivalent binding molecules:

(a) epitope (1) is an epitope of HIV-1 Envelope V3 glycan or an epitope of HIV-1 Envelope CD4 binding site, epitope (2) is an epitope of CD3, and epitope (3) is an epitope of CD8;

(b) epitope (1) is an epitope of HIV-1 Envelope V3 glycan or an epitope of HIV-1 Envelope CD4 binding site, epitope (2) is an epitope of CD8, and epitope (3) is an epitope of CD3;

(c) epitope (1) is an epitope of CD3, epitope (2) is an epitope of HIV-1 Envelope V3 glycan or an epitope of HIV-1 Envelope CD4 binding site, and epitope (3) is an epitope of CD8; (d) epitope (1) is an epitope of CD3, epitope (2) is an epitope of CD8, and epitope (3) is an epitope of HIV- 1 Envelope V3 glycan or an epitope of HIV- 1 Envelope CD4 binding site;

(e) epitope (1) is an epitope of CD8, epitope (2) is an epitope of CD3, and epitope (3) is an epitope of HIV- 1 Envelope V3 glycan or an epitope of HIV- 1 Envelope CD4 binding site; or

(f) epitope (1) is an epitope of CD8, epitope (2) is an epitope of HIV-1 Envelope V3 glycan or an epitope of HIV-1 Envelope CD4 binding site, and epitope (3) is an epitope of CD3.

[0055] In certain aspects the invention provides such trivalent binding molecules wherein two of epitope (1), epitope (2), and epitope (3) are an epitopes of HIV-1 Envelope, and one of epitope (1), epitope (2), and epitope (3) is an epitope of CD3, where said epitopes of HIV-1 Envelope may be the same epitope or different epitopes. In particular aspects of such trivalent binding molecules:

(a) epitope (1) is an epitope of HIV-1 Envelope V3 glycan or an epitope of HIV-1 Envelope CD4 binding site, epitope (2) is an epitope of HIV-1 Envelope V3 glycan or an epitope of HIV-1 Envelope CD4 binding site, and epitope (3) is an epitope of CD3;

(b) epitope (1) is an epitope of HIV-1 Envelope V3 glycan or an epitope of HIV-1 Envelope CD4 binding site, epitope (2) is an epitope of CD3, and epitope (3) is an epitope of HIV-1 Envelope V3 glycan or an epitope of HIV-1 Envelope CD4 binding site; or

(c) epitope (1) is an epitope of CD3, epitope (2) is an epitope of HIV-1 Envelope V3 glycan or an epitope of HIV-1 Envelope CD4 binding site, and epitope (3) is an epitope of HIV-1 Envelope V3 glycan or an epitope of HIV-1 Envelope CD4 binding site.

[0056] In certain aspects, domain (H) comprises a binding region of a light chain variable domain of an anti-CD8 antibody, an anti-CD3 antibody, an anti-CD 16 antibody, an HIV-1 envelope CD4 binding site antibody, or a HIV-1 envelope V3 glycan antibody. In certain aspects, domain (G) comprises a binding region of a heavy chain variable domain of an anti- CBS antibody, an anti-CD3 antibody, an anti-CD 16 antibody, an HIV-1 envelope CD4 binding site antibody, or a HIV-1 envelope V3 glycan antibody. [0057] In certain aspects, domain (B) comprises the heavy chain variable domain of an anti- CD3 antibody, an anti-CD8 antibody, or an anti-CD 16 antibody. In certain embodiment, domain (D) comprises the light chain variable domain of an anti-CD3 antibody, an anti-CD8 antibody, or an anti-CD 16 antibody.

[0058] In certain aspects, the CH2-CH3 domain of the first polypeptide chain of any of the multispecific molecules of the invention is the of the "knob" design and the CH2-CH3 domain of the third polypeptide chain of any of the multivalent molecules of the invention is of the "hole" design.

[0059] In certain aspects, the CH2-CH3 domain of the third polypeptide chain of any of the multispecific molecules of the invention is the of the "knob" design and the CH2-CH3 domain of the first polypeptide chain of any of the multivalent molecules of the invention is of the "hole" design.

[0060] In certain aspects, the CH2-CH3 domain of the first polypeptide chain is the of the "knob" design (SEQ ID NOs: 531 or 532) and the CH2-CH3 domain of the third polypeptide chain is of the "hole" design (SEQ ID NOs: 533 or 534). In certain aspects, the CH2-CH3 domain of the first polypeptide comprises SEQ ID NO: 531 and the CH2-CH3 domain of the third polypeptide chain comprises SEQ ID NO: 533. In certain aspects, the CH2-CH3 domain of the third polypeptide chain is the of the "knob" design (SEQ ID NOs: 531 or 532) and the CH2-CH3 domain of the first polypeptide chain is of the "hole" design (SEQ ID NOs: 533 or 534). In certain aspects, the CH2-CH3 domain of the third polypeptide comprises SEQ ID NO: 531 and the CH2-CH3 domain of the first polypeptide chain comprises SEQ ID NO: 533.

[0061] In certain aspects, the epitope (2) is a CD3 epitope, CD8 epitope or a CD 16 epitope. In certain embodiments, the bispecific or trivalent molecule binds HIV-1 envelope with the specificity of the V3 glycan binding antibody DH542, a variant of DH542 called

DH542 QSA, DH542 L4 and/or other antibodies from the DH542 lineage, and/or the binding specificity of CD4 binding site antibody CH557 (also referred to as CH235.12) or any of the antibodies from the CH235 HIV-1 antibody lineage and also binds CD3. In certain embodiments, the bispecific or trivalent molecule binds HIV-1 envelope with the specificity of the V3 glycan binding antibody DH542, a variant of DH542 called DH542 QSA,

DH542 L4 and/or other antibodies from the DH542 lineage, and/or the binding specificity of CD4 binding site antibody CH557 (also referred to as CH235.12) or any of the antibodies of the CH235 HIV-1 antibody lineage and also binds CD3, CD8, or CD16. [0062] In certain aspects, domain (A) comprises the CDR1, CDR2, and CDR3 of the light chain variable domain of immunoglobulin CH557, CH556, CH555, CH493, CH492, CH491, or CH490. In certain aspects, the domain (E) comprises the CDR1, CDR2, and CDR3 of the heavy chain variable domain of immunoglobulin CH557, CH556, CH555, CH493, CH492, CH491, or CH490. In certain aspects, domain (A) comprises the light chain variable domain of immunoglobulin CH557, CH556, CH555, CH493, CH492, CH491, or CH490. In certain aspects, domain (E) comprises the heavy chain variable domain of immunoglobulin CH557, CH556, CH555, CH493, CH492, CH491, or CH490.

[0063] In certain aspects, domain (A) comprises the CDR1, CDR2, and CDR3 of the light chain variable domain of immunoglobulin DH542, DH542 QSA, DH542 L4, DH429. In certain aspects, the domain (E) comprises the CDR1, CDR2, and CDR3 of the heavy chain variable domain of immunoglobulin DH542, DH542 QSA, DH542 L4, DH429. In certain aspects, domain (A) comprises the light chain variable domain of immunoglobulin DH542, DH542 QSA, DH542 L4, DH429. In certain aspects, domain (E) comprises the heavy chain variable domain of immunoglobulin DH542, DH542 QSA, DH542 L4, DH429.

[0064] In certain aspects, the first polypeptide comprises SEQ ID NO: 555. In certain aspects, the second polypeptide comprises SEQ ID NO: 557. In certain aspects, the third polypeptide comprises SEQ ID NO: 559.

[0065] In certain aspects, the bispecific molecule comprises the first polypeptide of SEQ ID NO: 555, the second polypeptide of SEQ ID NO: 557, and the third polypeptide of SEQ ID NO: 559.

[0066] In certain aspects, the bispecific molecule consists essentially of the first polypeptide of SEQ ID NO: 555, the second polypeptide of SEQ ID NO: 557, and the third polypeptide of SEQ ID NO: 559. In certain aspects, the bispecific molecule consists of the first polypeptide of SEQ ID NO: 555, the second polypeptide of SEQ ID NO: 557, and the third polypeptide of SEQ ID NO: 559.

[0067] In certain aspects, a four chain trivalent binding molecule is a trispecific molecule and comprises the first polypeptide of SEQ ID NO: 555, the second polypeptide of SEQ ID NO: 557, the third polypeptide of SEQ ID NO: 561, the fourth polypeptide of SEQ ID NO: 562 (See Figure 4A).

[0068] In certain aspects, a three chain trivalent binding molecule is a trispecific molecule and comprises the first polypeptide of SEQ ID NO: 555, the second polypeptide of SEQ ID NO: 557, the third polypeptide of SEQ ID NO: 563 (See Figure 4D).

[0069] In certain aspects, the first polypeptide comprises SEQ ID NO: 596. In certain embodiments, the second polypeptide comprises SEQ ID NO: 597. In certain embodiments, the third polypeptide comprises SEQ ID NO: 559. An exemplary DH542 bispecific Fc bearing diabody is shown in Figure 3 A.

[0070] In certain aspects, the invention provides a composition comprising any one of the multispecific molecules or any combination thereof. In certain aspects, the composition comprises a composition comprising a bispecific molecule comprising a first arm with the binding specificity of a HIV-1 envelope CD4 binding site antibody or HIV-1 V3 glycan binding site antibody and a second arm targeting CD3, CD8, or CD16. In certain aspects, the bispecific molecule comprises an Fc portion or any other modification which extends its serum half-life. In certain aspects, the composition further comprises a second bispecific molecule or trivalent binding molecule comprising a first arm with an HIV-1 envelope binding specificity different from the HIV-1 binding specificity of the first multispecific molecule, and a second arm targeting CD3, CD8, or CD 16, wherein the first and second multipecific molecules are different in either the HIV-1 binding specificity and/or the specificity of the second arm.

[0071] In certain aspects, the invention provides a method to treat or prevent HIV-1 infection in a subject in need thereof comprising administering to the subject a composition comprising any one of the multispecific molecules of the invention or a combination of any one of the multispecific molecules in a therapeutically effective amount. In certain embodiments, the methods further comprise administering a latency activating agent. In some embodiments, the latency activating agent is vorinostat, romidepsin , panobinostat, disulfiram, JQl, bryostatin, PMA, inonomycin, or any combination thereof. [0072] In certain aspects, the invention provides nucleic acids comprising nucleotides encoding the multispecific molecules of the invention. In certain aspects, the invention provides a vector comprising nucleic acids comprising nucleotides encoding the multispecific molecules of the invention. Provided are also compositions comprising a vector comprising a nucleic acid encoding the multispecific molecules. In certain aspects the invention provide a cell line comprising vectors or nucleic acids encoding the multispecific molecules of the invention, wherein the vectors encode polypeptide chains for expression of the multispecific molecules of the invention, e.g. but not limited to, polypeptide chain 1 and polypeptide chain 2, or polypeptide chain 1, polypeptide chain 2 and polypeptide chain 3. In certain

embodiments, the vector is suitable for gene delivery and expression. In certain embodiment, the vector is an adenoviral vector, an adeno associated virus based vector, or a combination thereof.

[0073] In certain embodiments, the multispecific molecule binds to the HIV-1 envelope like the HIV-1 antibody from which it is derived. In certain embodiments, the multispecific molecule binds to the CH557-HIV-1 envelope epitope, i.e. the multispecific molecule binds to the HIV-1 envelope like the CH557 antibody, and also binds CD3, CD8, or CD16.

[0074] In certain embodiments, the multispecific molecule binds to the DH542-HIV-1 envelope epitope, i.e. the multispecific molecule binds to the HIV-1 envelope like the DH542 antibody, and also binds CD3, CD8, or CD 16.

[0075] In certain embodiments a multispecific molecule of the invention comprises, consists essentially of or consists of sequences as described herein, (e.g., Table 4).

[0076] In certain aspects the invention provides compositions comprising any of the multispecific molecules described herein, or a combination thereof. In certain embodiments, these compositions are formulated as pharmaceutical composition for therapeutic use.

[0077] In certain aspects the invention is directed to nucleic acids which encode the multispecific molecule of the invention. In certain embodiments, these nucleic acids are comprised in a vector, and are operably linked to a promoter. In certain aspects the invention provides cell lines, or isolated cells, which comprise nucleic acids for the expression of the multispecific molecule of the invention.

[0078] In certain aspects, the invention provides compositions comprising the multispecific molecule of the invention or nucleic acids encoding the same for use in methods of treating or preventing HIV-1 infection. In some embodiments, these methods further comprise administering a Latency Activating Reagent. Non-limiting examples of these include HDAC inhibitors, e,g, vorinostat, romidepsin, panobinostat, disulfiram, JQ1, bryostatin, PMA, inonomycin, or any combination thereof. In some embodiments, this combination therapy targets the pool of latently infected HIV-1 cells.

[0079] In certain aspects, the invention provides methods of treating or preventing an HIV-1 infection in a subject, the method comprising administering to the subject a composition comprising any one of the multispecific molecules the invention, or a combination thereof in a therapeutically sufficient amount. In certain embodiments, the methods further comprise administering a latency activating agent.

BRIEF DESCRIPTION OF THE DRAWINGS

[0080] To conform to the requirements for PCT patent applications, many of the figures presented herein are black and white representations of images originally created in color.

[0081] Figure 1 provides a schematic of a representative covalently bonded diabody having two epitope-binding sites composed of two polypeptide chains, each having an E-coil or K- coil Heterodimer-Promoting Domain (alternative Heterodimer-Promoting Domains are provided below). A cysteine residue may be present in a linker and/or in the Heterodimer- Promoting Domain as shown in Figure 2B. VL and VH Domains that recognize the same epitope are shown using the same shading or fill pattern.

[0082] Figures 2A-2C provide schematics showing representative covalently bonded tetravalent diabodies having four epitope-binding sites composed of two heterodimer pairs of polypeptide chains (i.e., four polypeptide chains in all). One polypeptide of each heterodimer pair possesses a CH2 and CH3 Domain, such that the associated chains form all or part of an Fc Domain. VL and VH Domains that recognize the same epitope are shown using the same shading or fill pattern. The two pairs of polypeptide chains may be the same. In such embodiments wherein the two pairs of polypeptide chains are the same and the VL and VH Domains recognize different epitopes (as shown in Figures 2A-2B), the resulting molecule possesses four epitope-binding sites and is bispecific and bivalent with respect to each bound epitope. In such embodiments wherein the VL and VH Domains recognize the same epitope (e.g., the same VL Domain CDRs and the same VH Domain CDRs are used on all chains) the resulting molecule possesses four epitope-binding sites and is monospecific and tetravalent with respect to a single epitope. Alternatively, the two pairs of polypeptides may be different. In such embodiments wherein the two pairs of polypeptide chains are different and the VL and VH Domains of each pair of polypeptides recognize different epitopes (as shown by the different shading and patterns in Figure 2C), the resulting molecule possesses four epitope- binding sites and is tetraspecific and monovalent with respect to each bound epitope. Figure 2A shows an Fc Domain-containing diabody which contains a peptide Heterodimer- Promoting Domain comprising a cysteine residue. Figure 2B shows an Fc Domain- containing diabody, which contains E-coil and K-coil Heterodimer-Promoting Domains comprising a cysteine residue and a linker (with an optional cysteine residue). Figure 2C, shows an Fc-Region-Containing diabody, which contains antibody CHI and CL domains which could serve as Heterodimer Promoting Domains.

[0083] Figures 3A and 3B provide schematics of a representative covalently bonded Fc bearing diabody molecule having two epitope-binding sites composed of three polypeptide chains. Two of the polypeptide chains comprise a CH2 and CH3 Domain, such that the associated chains form all or part of an Fc Domain. The polypeptide chains comprising the VL and VH Domain further comprise a Heterodimer-Promoting Domain. VL and VH

Domains that recognize the same epitope are shown using the same shading or fill pattern.

[0084] Figures 4A-4F provide schematics of representative Fc Domain-containing trivalent binding molecules having three epitope-binding sites. Figures 4A and 4B, respectively, illustrate schematically the domains of trivalent binding molecules comprising two diabody- type binding domains and a Fab-type binding domain having different domain orientations in which the diabody-type binding domains are N-terminal or C-terminal to an Fc Domain. The molecules in Figures 4A and 4B comprise four chains. Figures 4C and 4D, respectively, illustrate schematically the domains of trivalent binding molecules comprising two diabody- type binding domains N-terminal to an Fc Domain, and a Fab-type binding domain in which the light chain and heavy chain are linked via a polypeptide spacer, or an scFv-type binding domain. The trivalent binding molecules in Figures 4E and 4F, respectively, illustrate schematically the domains of trivalent binding molecules comprising two diabody-type binding domains C-terminal to an Fc Domain, and a Fab-type binding domain in which the light chain and heavy chain are linked via a polypeptide spacer, or an scFv-type binding domain. The trivalent binding molecules in Figures 4C-4F comprise three chains. VL and VH Domains that recognize the same epitope are shown using the same shading or fill pattern.

[0085] Figures 5A-5E provide schematics of a representative HIVxCD3 bispecific monovalent diabody comprising three polypeptide chains. Figure 5A shows the domains of each of the three polypeptide chains, dashed lines represent disulfide bonds which form between the chains, and the arrows indicate the interactions of the Variable Domains. Figure 5B provides a schematic of the assembled chains. Such diabodies contain an anti -HIV-1 binding arm (e.g., DH491 or CH493, DH542, CH557, CH558, or any of the CH235 lineage antibodies) combined with an anti-CD3 binding arm (e.g., hXR32). They are composed of two polypeptide chains: one with the VL of an anti-CD3 antibody linked to the VH of an anti- HIV-1 antibody; the second with the VL of an anti -HIV-1 antibody linked to the VH of an anti-CD3 antibody. The first and the second polypeptide chains are linked by interchain disulfide bond and paired via oppositely charged E-coil/K-coil Heterodimer-Promoting Domains. The amino acid and nucleotide sequences of Chain 1, 2 and 3 which form the bispecific monovalent diabody designated CH557xCD3 Fc are provided in Table 4 (SEQ ID NOs: 555, 556, 557, 558, 559, 560). Control molecules have one of the arms replaced by a non-HIV-1 envelope binding arm derived, for example, from an anti-FITC antibody (4420) or from an anti-RSV antibody (palivizumab).

[0086] Figures 6A-6C show antigen binding by ELISA. Binding of bispecific monovalent diabodies (CH557xCD3 Fc, the comparator A32xCD3 Fc, and the controls RSVxCD3 Fc and A32xRSV Fc) to human CD3 protein (Figure 6A), to M.CONS gpl40 protein (Figure 6B) or simultaneously binding to both human CD3 and M.CONS gpl40 proteins (Figure 6C).

[0087] Figures 7A-7B show cell surface binding by FACS. Binding of the bispecific monovalent diabodies (CH557xCD3 Fc, the comparator A32xCD3 Fc, and the controls RSVxCD3 Fc and A32xRSV Fc) to HEK293-D375 cells expressing HIV-1 Env, (Figure 7A) or to primary human T cells expressing CD3 (Figure 7B), Data are reported as mean fluorescence intensity (MFI).

[0088] Figures 8A-8B show redirected T-cell killing of HIV-1 Env+ target cells. The results from two different donors are shown (Figures 8A and 8B). Both HIVxCD3 Fc bispecific monovalent diabodies mediate concentration dependent killing of HIV-1 Env+ HEK293- D375 cells in the presence of human PBMC effector cells at an E:T ratio of 30: 1 for 28 hours with cytolysis measured by LDH release assay. The control molecule (RSVxCD3 Fc) was inactive. The average EC50 values were 13.8 ng/mL and 12.3 ng/mL for CH557xCD3 Fc and A32xCD3 Fc, respectively.

[0089] Figures 9A-9E provide the amino acid sequences of exemplary

(CH557HIVxCD3xCD8 trivalent binding molecules. Figures 9A-9D represent the amino acid sequences of polypeptide chains 1-4, respectively, of an exemplary HIVxCD3xCD8 trivalent binding molecule comprising four polypeptide chain (SEQ ID NOs: 555, 557, 561, 561). Figures 9A, 9B and 9E represent the amino acid sequences of polypeptide chains 1, 2 and 3, respectively, of an exemplary HIVxCD3xCD8 trivalent binding molecule comprising three polypeptide chains (SEQ ID NOs: 555, 557, 563).

[0090] FIG. 10A shows DH542 sequences (CDRs are bolded and underlined) (SEQ ID NOs: 1-4).

[0091] FIG. 10B shows the amino acids sequences of VH and VL chains of antibodies of the DH270 lineage, and nucleic acid sequences encoding these amino acids. CDRs are highlighted in each antibody. DH270IA1 as listed in Figure 3 is the same antibody referred to as II in the sequence of Figure 2. The figure shows SEQ ID NOs: 5-16 (Heavy chain nucleotide sequences in order of appearance from UCA-DH270H), SEQ ID NOs: 17-28 (Heavy chain amino acid sequences in order of appearance from UCA-DH270H), SEQ ID NOs: 29-40 (Light chain nucleotide sequences in order of appearance from UCA-DH270H), SEQ ID NOs: 41-52 (Light chain amino acid sequences in order of appearance from UCA- DH270H).

[0092] FIG. 11 shows neutralization by antibodies CH555, CH556, CH557, CH558, CH560, CH561, CH562, DH210 and DH211 identified from subject CH505 measured in TZM-bl cells. Pseudoviruses were produced by transfection in 293T cells. Values are the antibody concentration ^g/ml) at which relative luminescence units (RLUs) were reduced 50% compared to virus control wells (no test sample). Values in bold are considered positive for neutralizing antibody activity. CH557 IC50 neutralization summary: mean IC50 = 3.66 μg/ml; geometric mean IC50 = 0.66 μg/ml; median = 0.62 μg/ml.

[0093] FIG. 12 shows sequences of CH557 (SEQ ID NOs: 124-127). CDRs are bolded and underlined.

[0094] FIG. 13 shows CH557 gene information. (SEQ ID NO: 215).

[0095]

[0096] FIG. 14 shows gene information for antibodies in the CH235 lineage.

[0097] FIGS. 15A and 15B show amino acid alignment of CH235 lineage antibody heavy chain (Figure 15 A) and light chain (Figure 15B). Antibodies are listed in ascending order of somatic mutations and compared to the inferred unmutated common ancestor previously published (Gao, Bonsignori, Liao et al. Cell 2014). FIG. 15A shows SEQ ID NOs: 128-139 (in order of appearance from UCA-CH557). FIG. 15B shows SEQ ID NOs: 140-149 (in order of appearance from UCA-CH556).

[0098] FIG. 16 shows alignment of CH557 heavy chain amino acid sequence compared to CH235 lineage antibodies with increasing levels of somatic mutations and neutralization breadth. Contact sites with gpl20 identified from CH235/gpl20 co-crystal structure are indicated with asterisks. Amino acid mutations within the contact sites are bolded and underlined. Figure 16 shows SEQ ID NOs: 150-154 (in order of appearance from UCA- CH557).

[0099] FIG. 17 shows alignment of CH557 light chain amino acid sequence compared to CH235 lineage antibodies with increasing levels of somatic mutations and neutralization breadth. Figure 17 shows SEQ ID NOs: 155-158 (in order of appearance from UCA-CH557).

[0100]

[0101] FIG. 18 shows amino acid sequences of VH and VL chains of antibodies CH490, CH491, CH492 and CH493 (CH235_129w66 = CH490; CH235_68wl00 = CH491;

CH235_115wl00 = CH492; CH235_75wl52 = CH493) (SEQ ID NOs: 159-164).

[0102] FIG. 19 shows nucleic acid sequences of VH and VL chains of antibodies CH490, CH491, CH492 and CH493 (CH235_129w66 = CH490; CH235_68wl00 = CH491;

CH235_115wl00 = CH492; CH235_75wl52 = CH493) (SEQ ID NOs: 165-170).

[0103] FIG. 20 shows phylogenetic tree of the heavy chains of antibodies CH490, CH491, CH492 and CH493 (see table in Figure 19). See also Example 6: CH240, CH239, CH235, CH236, CH241 VH chains were identified from cultured memory B cells. The rest of the VH chains were retrieved with deep sequencing.

[0104] FIG. 21 shows ELISA binding of CH490, CH491, CH492 and CH493 antibodies to various antigens as listed in the figure.

[0105] FIG. 22 shows a summary of neutralization data of CH490, CH491, CH492 and CH493 antibodies for various HIV-1 strains in TZMbl assay. Intermediate antibodies are described in Gao, Bonsignori, Liao et al. Cell 158, 481-491, July 31, 2014.

[0106] FIG. 23 shows a summary of neutralization data for antibodies CH235, CH490, CH491, and CH493. The viruses are CH505 TF (autologous virus) in which either point or multiple single mutations were introduced in the loop D region. The loop D mutations are described in Gao et al Cell 158, 481-491, July 31, 2014 (incorporated by reference). These mutations reflect mutations in the loop D region that naturally occurred in the autologous virus of this subject over time. They were artificially introduced into the CH505 TF to study their effect in absence of other mutations in other parts of the autologous virus that also occurred during virus evolution. These mutations were induced by CH235 lineage antibodies identified early during the course of infection. These data show that the more mutated antibody CH493 which came from a later time point and that is broadly neutralizing also acquired the ability of recognizing virus escape mutants that escaped earlier antibodies from the same lineage, before they acquired substantial breadth.

[0107] FIG. 24 shows a summary of neutralization (TZMbl assay) data of CH505 D loop mutants by various antibodies. The summary shows that CH493 neutralizes all Loop D CH505 mutants.

[0108] FIG. 25 shows results of HEp-2 cell IF staining for CH557.

[0109] FIGS. 26A and 26B shows summary of data from CH557 microarray polyreactivity.

[0110] FIG. 27A shows summary results of neutralization data of CH557, CH235, VRCOl, VRC07-523-LS, N6, 3BNC117, 8ANC131, CH103, F105, and DH522 against a panel of HIV-1 isolates in the Luc/TZM-bl neutralization assay. Values represent IC50 in μg/ml.

[0111] FIG. 27B shows the mean IC50 and percent of isolates neutralized at different IC50 values.

[0112] FIG. 28A shows summary results of neutralization data of CH557, CH235, VRCOl, VRC07-523-LS, N6, 3BNC117, 8ANC131, CH103, F105, and DH522 against a panel of HIV-1 isolates in the Luc/TZM-bl neutralization assay. Values represent IC80 in μg/ml. [0113] FIG. 28B shows the mean IC80 and percent of isolates neutralized at different IC80<50ug/ml values.

[0114] FIGS. 29A-29B show CH235 Lineage, with Time of Appearance and Neutralization by Select Members. (A) Phylogram of CH235 lineage. Phylogenetic tree is colored by first time (wk post-infection) from which sequences were obtained. Key members of the CH235 lineage are labeled. CH235.6, CH235.7, CH235.8 and CH235.9 VH were complemented with full heavy chain gene regions and paired with the VL from the closest natural antibody. (B) Neutralization dendrograms display single mAb neutralization of a genetically diverse panel of 199 HIV-1 isolates. Coloration is by IC50. See also Figures 36A-B, 40C, and 41.

[0115] FIGS. 30A-30E show structures of CH235 -Lineage Members in Complex with HIV-1 Env. (A) Co-crystal structures of the antigen-binding fragments (Fabs) of CH235-lineage members with core gpl20. Structures are shown in ribbon diagram, with gpl20 in gray and residues altered by SHM in stick representation colored by time-of-appearance. (B) Negative stain EM of Fabs of CH235-lineage members and trimeric HIV-1 Env from BG505 (top row) and B41 (bottom row). Structures in surface representation, with Env portions colored gray and Fabs by time-of-appearance. (C) Epitope displayed on the gpl20 surface and colored by antibody time-of-appearance, with the vulnerable portion of the CD4bs highlighted in yellow and select regions labeled. (D) Targeting precision of CD4bs-directed antibodies vs neutralization breadth. (E) VH-gene SHM of CD4bs-directed antibodies vs neutralization breadth. See also Figures 37A-G, and 42.

[0116] FIGS. 31A-C show sequence Evolution of CH235 Lineage: SHM, Timing, and Conformity of CH235-Lineage Development from UCA to Antibody with 90% Breadth. (A) Heavy chain SHM over time for the CH235 lineage (left panel). SHM levels of other VH1- 46-derived CD4bs mAbs and selected VHl-2-derived VRCOl -class mAbs are shown (middle and right panels, respectively); the time since infection is unknown for these mAbs. (B) Maturation conformity vs overall heavy chain SHM. Positional conformity (top row) is defined as the number of aa positions differing from the germline sequence in both the conforming and reference sequences, divided by the total number of aa changes in the conforming antibody. Identity conformity (bottom row) is defined as the number of such positions which are additionally mutated to the same residue, divided by the total number of mutations in the conforming antibody. Conformity to 1B2530 (left) and to 8ANC131 (right) is shown for both position and identity. (C) VH-gene mutability accounts for the majority of positional conformity of CH235 lineage. The mutability of the VH-gene for VH1-46 (top; SEQ ID NOS 224-229, respectively, in order of appearance) and VH1-2 (bottom; SEQ ID NOS 230-233, respectively, in order of appearance) is shown. Sequence logos are shown at each position; the height of each logo corresponds to the percent of mutated reads. Green bars are shown for SHM in antibody CH235, which are altered in over a quarter of VH1-46- derived antibodies. See also Figures 38A-E, and 43A-C.

[0117] FIGS. 32A-32D show binding Kinetics of CH103 and CH235 Lineage Antibodies. Binding association (ka) and dissociation (kd) rates of the CHI 03 (A-B, squares) and CH235 (C-D, circles) lineage mAbs to CH505.TF gpl20 Env were measured with SPR and used to calculate the dissociation rate constants (Kd). Kds are shown in A and C, ka (solid lines, plotted on the left y-axis) and kd (dashed lines, plotted on the right y-axis) are shown in B and D. See also Figures 44 A-B.

[0118] FIGS. 33A-33C show CH235 Lineage Antibodies Neutralization of Autologous Virus and CH505.TF Loop D Mutants. (A) Heatmap analysis of neutralization of 76 pseudoviruses (row) by 16 CH235 lineage mAbs (column). Coloration is by IC50. This analysis extends previous observations on early CH235 lineage antibodies (Gao et al., 2014) by including late mAbs CH235.7, CH235.8, CH235.10, CH235.11, CH235.12 and CH235.13 and by adding pseudoviruses isolated from wk 136 to 323 post-transmission. (B) CH505 TF and loop D mutants M5, M6, M10, M19, Mi l, M7, M8, M9, M20 and M21 neutralization by CH236 mAb, late mAbs CH235.7, CH235.9 CH235.10, CH235.11, CH235.12, CH235.13 (left panel) and CH235.9 mAb mutants (right panel). Neutralization is expressed as IC50 μg/ml. CH505 TF sequence mutations are shown on the right. (C) The CDR HI N30 (sticks, dark red) in CH235.9, which interacts with the D20- D21 loop in the bridging sheet of gpl20 (cyan), is over 19A away from the N280S mutation site in loop D (orange). See also Figures 39A-B, 45, and 46.

[0119] FIGS. 34A-34B show binding of CH235 and CH103 Lineage mAbs to Autologous CH505. (A) and CH235 UCA Binding to Heterologous HIV-1 Env Glycoproteins (B). (A) Heatmap analysis of UCA, intermediate (IA) and mature CH235 and CHI 03 lineage mAbs binding to 113 CH505 autologous Env isolated from time of infection (TF) to 160 wks postinfection and to the CH505.TF mutants (Gao et al., Cell 2014). Mabs were tested in ELISA at concentrations ranging from 100 μg/ml to 0.6 ng/ml . Binding is expressed as a LogAUC. (B) Affinity of CH235 UCA, CH235 wild-type and select SHM variants to a panel of 15 heterologous gpl20 Envs. See also Figures 40A-40B and 43. [0120] FIGS. 35A-35D show CH235 Antibody Lineage Auto- and Polyreactivity. (A) CH235 lineage antibody binding to ANA measured in ELISA. LogAUC was calculated from duplicate samples. Results representative of duplicate experiments. (B) Binding to cardiolipin was determined using Quanta Lite ACA IgG III ELISA Assay. (C) Hep2 cell IF staining. Size bars = 50 μιη. (D) Measurement of polyreactivity against 9,400 human antigens using ProtoArray 5 microchip: CH235 lineage mAbs binding (x-axis) was compared to non- polyreactive control mAb 15 IK (y-axis). Polyreactivity is defined as 1 log stronger binding than 151k mAb to more than 90% of the test proteins. High affinity binding was measured as a >2 log increase in binding (dotted line) (Liu et al., 2015).

[0121] FIGS. 36A-36B show CH235 Lineage: Sequences and Neutralization Fingerprint Dendrogram, Related to Figure 29. (A) Alignment of NGS sequences (SEQ ID NOS 234 and 234-294, respectively, in order of appearance) and antibodies identified from 17 time points from 6 to 323 weeks post-transmission and comparison of mutation patterns to other IGHVl- 46 (1B2530 and 8ANC131) and IGHV1-2 (VRCOl, VRC-CH31 and VRC-PG04) derived broadly neutralizing antibodies. Antibodies identified from single B cells are shown in bold. The positions mutated in CH235 were color coded based on the time points at which these mutations were firstly observed in the NGS reads. Mutated positions not seen in the NGS data are colored based on the time of isolation of CH235 (41 weeks). IGHV1-46*01 is used as reference for IGHVl-46 derived antibodies and IGHVl-2*02 is used as reference for the three VRCOl -class antibodies. (B) The neutralization fingerprints for three antibodies from the CH235 lineage were compared to the fingerprints for other VH1-46 class antibodies and non- VH1-46 class CD4-binding-site antibodies; coloring same as in Figure 31A-C. Antibodies targeting other sites of vulnerability on HIV-1 Env are shown as control (black).

[0122] FIGS. 37A-37G show CH235 Lineage Versus Other CD4-Binding Site Antibodies and Negative-stain EM Reconstructions of gpl40 SOSIP Trimers with CH235-lineage Fabs, Related to Figure 30A-E. (A) CD4-mimicry by CH235. Recognition of gpl20 by the N- terminal domain of the CD4 receptor (far left) is compared to VH genes from CH235 and prototypic antibodies VRCOl (from VH1-2) and 8ANC131 (from VH1-46). (B) Conserved molecular interactions between antibody CH235, receptor CD4 and antibody VRCOl . Top row shows intermolecular antiparallel strand interactions and bottom row Asp368 electrostatic interaction. (C) Binding orientation of VH-gene derived antibodies relative to CD4. (D) Negative-stain EM 3D models with BG505 SOSIP.664. (left) Top and side views of

CH235.12 in complex with BG505 SOSIP (purple) aligned to the EM volume of VRCOl in complex with BG505 (gold mesh; EMD-6252). (middle) Top and side views of the

CH235.12-BG505 complex (purple mesh) aligned to the EM volume of CH103 in complex with BG505 SOSIP (gray; EMD-6250). (right) Top and side views of the CH103-BG505 complex (gray mesh) aligned to the EM reconstruction of BG505 SOSIP in complex with soluble CD4 and 17b Fab (blue; EMDB ID 5723). (E) Negative-stain EM of gpl40 SOSIP trimers with CH235-lineage Fabs. (F) Top and side views of 3D reconstructions of each complex. (G) Fourier shell correlation curves for each dataset with a resolution estimate using an FSC cutoff of 0.5.

[0123] FIGS. 38A-38E show Sequence Similarity Between VHl-2 and VHl-46 Broadly Neutralizing Antibodies and Mutability of Germline Genes, Related to Figure 31A-C. (A) Amino acid alignment of 8ANC131 (SEQ ID NO: 290) and CH235 (SEQ ID NO: 271) to the IGHV1-46 germline gene (SEQ ID NO: 234) showing the definition of conformity. (B) Probability distribution of the number of sharing mutation positions for each pair of antibodies. (C) Probability distribution of the number of identical mutations for each pair of antibodies. (D) SHM frequency is shown versus VH-gene position for VHl-46, VHl-2 and three others. Sequences were aligned to VHl-46 and positions not aligned to VHl-46 (indels) were removed. (E) Dendrogram showing sequence segregation of VHl-2 and VHl-46 derived broadly neutralizing antibodies, despite similarity of VHl-2 and VHl-46 germline genes shown with underline.

[0124] FIGS. 39A-39B show Generation of CH235.9 Mutants to Evaluate the Effect of Mutations in the V-heavy Chain on the Ability of CH235.9 to Neutralize loop D Mutant CH505 Autologous Viruses, Related to Figure 33A-C. (A) The interaction between CH235 CDR L3 (purple) and N280 in the HIV-1 gpl20 Env loop D (orange) from the crystal structure of the CH235-gpl20 complex (left panel). Asparagine in position 280 in gpl20 forms three hydrogen bonds (yellow dotted lines) with residues in the CDR L3 (left panel). Structural modeling predicted these hydrogen bonds to be disrupted in the N280S (right panel) and N280T (not shown) mutations which occur in autologous CH505 escape mutants. (B) Alignment of CH235.7 and CH235.9 through CH235.13 VH amino acid sequences to CH236 VH (SEQ ID NOs: 171-177 (in order of appearance from CH236-CH235.7)).

CH235.9 aa mutations expressed as recombinant IgG and tested for neutralization of CH505 TF loop D mutants are shown in red. Asterisks indicate points of contact with gpl20 derived from the CH235 crystal structure in complex with gpl20 Env. [0125] FIGS. 40A-40B show CH505 gpl20 Env Quasi-species Selected as Optimized Immunogens to Induce Both CH235 and CH103-like bnAbs, Related to Figure 34A-B. (A) Heatmap of the binding data of selected CH235 and CHI 03 lineage members to the CH505 Env glycoproteins selected to be used as immunogens. Individual Env clone names and weeks of isolation are shown on the left. (B) Affinity of gHgL of 1B2530, 8ANC131, VRCOl, VRC- PG04 and VRC-CH31 to a panel of 15 heterologous gpl20 envelope glycoproteins.

[0126] FIG. 40C shows a Table with characteristics of the V(D)J rearrangements of key CH235 lineage antibodies. Related to Figure 29.

[0127] FIG. 41 shows a Table with a summary of the Breadth and Potency of Antibody Neutralization Against 199 HIV-1 Env-Pseudoviruses. Related to Figure 29.

[0128] FIG. 42 shows a Table with crystallographic Data Collection and Refinement

Statistics. Related to Figures 30A-30E.

[0129] FIGS. 43A-43C show Sequence Similarity Between VH1-2 and VH1-46 Broadly Neutralizing Antibodies and Mutability of Germline Genes. Related to Figures 31A-31C. (A) The probability of a conforming VH1-46 antibody with x VH mutations, having c common mutation positions with a reference antibody were estimated based on 100,000 simulated events, with the likelihood of each residue being mutated based on uniform distribution (position) (Puniform), or the mutation frequency at each residue position derived from the VH1-46 antibodies (PVH1-46). (B) The probability of a conforming VH1-46 antibody with x VH mutations, having i identical mutations with a reference antibody were estimated based on 100,000 simulated events, with the likelihood of each residue being mutated based on uniform distribution (position and mutation type) (Puniform), or the mutation frequency at each residue position derived from the VH1-46 antibodies (PVH1-46). (C) Pearson correlation coefficients of positional somatic mutation frequency between VH1-46, VH1-2 and three others.

[0130] FIGS. 44A-44B show CH235 Lineage and CH106 Monoclonal Antibodies Cross- Blocking. Related to Figure 32. (A) CH235 lineage antibodies blocking of sCD4 and CH106 binding to CH505 TF gpl20 and B.63521 gpl20 Envs. Results expressed as IC50 ug/ml. nb= no blocking. (B) Monoclonal antibody CHI 06 blocking of CH235 lineage antibodies to CH505 TF gpl20. Results expressed as IC50 ug/ml. nb= no blocking.

[0131] FIG. 45 shows CH235 lineage autologous neutralization. Related to Figure 33A-C.

[0132] FIG. 46 shows CH235 lineage antibodies and CH235.9 mutants neutralization of CH505 TF loop D mutant viruses. Related to Figure 33A-C. [0133] FIG. 47 shows binding of antibodies in the CH235 and CH103 lineages to CH505 autologous Env glycoproteins, Related to Figure 34A-B.

[0134] Figure 48 shows levels of GFP expression detected with each Doxycycline

concentration in each staining condition of the examined bispecific molecules (Example IOC).

[0135] Figure 49 shows levels of GFP expression detected with each Doxycycline

concentration in each staining condition of the control molecules (Example IOC).

[0136] Figure 50 shows analysis of A32x4420 and CH557xhXR32 staining to the 491 control and 598 cell lines (Example IOC).

[0137] Figure 51 shows analysis of control antibodies staining to the 491 control and 598 cell lines (Example IOC).

DETAILED DESCRIPTION

[0138] Highly active anti-retroviral therapy (HAART) alone or in combination with latency reversing agents fails to reduce the pool of latently infected cells. This is due to limited ability of the CD8+ T cells to eliminate HIV-1 latently infected cells. Dual Affinity Re-Targeting proteins (DARTs) are multispecific, antibody-based diabody molecules that can bind at least two distinct antigens simultaneously. HIV-1 diabodies contain an HIV-1 binding arm combined with an effector cell binding arm (e.g., but not limited to CD3 effector cells) are designed to redirect effector cells (e.g. but not limited to cytotoxic CD3+ T cells) to engage and kill HIV-1 -infected cells. Additionally, multispecific molecules such as trivalent binding molecules containing one or more HIV-1 binding arms combined with one or more effector cell binding arms, may also be designed to redirected effector cells to engage and kill HIV-1- infected cells.

[0139] The provision of multispecific/non-mono-specific molecules provides a significant advantage over typical mono-specific antibodies: the capacity to co-ligate and co-localize cells that express different epitopes. Bivalent diabodies have wide-ranging applications including therapy and immunodiagnosis. Bi-valency allows for great flexibility in the design and engineering of the diabody in various applications, providing enhanced avidity to multimeric antigens, the cross-linking of differing antigens, and directed targeting to specific cell types relying on the presence of both target antigens. Due to their increased valency, low dissociation rates and rapid clearance from the circulation (for diabodies of small size, at or below -50 kDa), diabody molecules known in the art have also shown particular use in the field of tumor imaging (Fitzgerald et al. (1997) "Improved Tumour Targeting By Disulphide Stabilized Diabodies Expressed In Pichia pastoris, " Protein Eng. 10: 1221). Of particular importance is the co-ligating of differing cells, for example, the cross-linking of cytotoxic T cells to tumor cells (Staerz et al. (1985) "Hybrid Antibodies Can Target Sites For Attack By T Cells, " Nature 314:628-631, and Holliger et al. (1996) "Specific Killing Of Lymphoma Cells By Cytotoxic T-Cells Mediated By A Bispecific Diabody, " Protein Eng. 9:299-305).

[0140] Diabody epitope binding domains may also be directed to a surface determinant of a B cell, such as CD19, CD20, CD22, CD30, CD37, CD40, and CD74 (Moore, P. A. et al. (2011) "Application Of Dual Affinity Retargeting Molecules To Achieve Optimal Redirected T-Cell Killing Of B-Cell Lymphoma," Blood 117(17):4542-4551; Cheson, B.D. et al. (2008) "Monoclonal Antibody Therapy For B-Cell Non-Hodgkin 's Lymphoma " N. Engl. J. Med. 359(6):613-626; Castillo, J. et al. (2008) "Newer monoclonal antibodies for hematological malignancies," Exp. Hematol. 36(7):755-768. In many studies, diabody binding to effector cell determinants, e.g., Fey receptors (FcyR), was also found to activate the effector cell (Holliger et al. (1996) "Specific Killing Of Lymphoma Cells By Cytotoxic T-Ce lis Mediated By A Bispecific Diabody, " Protein Eng. 9:299-305; Holliger et al. (1999) "Carcinoembryonic Antigen (CEA)-Specific T-Cell Activation In Colon Carcinoma Induced By Anti-CD3 x Anti- CEA Bispecific Diabodies AndBl x Anti-CEA Bispecific Fusion Proteins, " Cancer Res. 59:2909-2916; WO 2006/113665; WO 2008/157379; WO 2010/080538; WO 2012/018687; WO 2012/162068). Normally, effector cell activation is triggered by the binding of an antigen bound antibody to an effector cell via Fc-FcyR interaction; thus, in this regard, diabody molecules may exhibit Ig-like functionality independent of whether they comprise an Fc Domain {e.g., as assayed in any effector function assay known in the art or exemplified herein {e.g., ADCC assay)). By cross-linking tumor and effector cells, the diabody not only brings the effector cell within the proximity of the tumor cells but leads to effective tumor killing (see e.g., Cao et al. (2003) "Bispecific Antibody Conjugates In Therapeutics, " Adv. Drug. Deliv. Rev. 55: 171-197).

[0141] The formation of such non-mono-specific diabodies requires the successful assembly of two or more distinct and different polypeptides {i.e., such formation requires that the diabodies be formed through the heterodimerization of different polypeptide chain species). This fact is in contrast to mono-specific diabodies, which are formed through the

homodimerization of identical polypeptide chains. Because at least two dissimilar polypeptides {i.e., two polypeptide species) must be provided in order to form a non-mono- specific diabody, and because homodimerization of such polypeptides leads to inactive molecules (Takemura, S. et al. (2000) "Construction Of A Diabody (Small Recombinant Bispecific Antibody) Using A Refolding System," Protein Eng. 13(8):583-588), the production of such polypeptides must be accomplished in such a way as to prevent covalent bonding between polypeptides of the same species {i.e., so as to prevent homodimerization)

(Takemura, S. et al. (2000) "Construction Of A Diabody (Small Recombinant Bispecific Antibody) Using A Refolding System," Protein Eng. 13(8):583-588). The art has therefore taught the non-covalent association of such polypeptides (see, e.g., Olafsen et al. (2004) "Covalent Disulfide -Linked Anti-CEA Diabody Allows Site-Specific Conjugation And Radiolabe ling For Tumor Targeting Applications, " Prot. Engr. Des. Sel. 17:21-27; Asano et al. (2004) "A Diabody For Cancer Immunotherapy And Its Functional Enhancement By Fusion Of Human Fc Domain," Abstract 3P-683, J. Biochem. 76(8):992; Takemura, S. et al. (2000) "Construction Of A Diabody (Small Recombinant Bispecific Antibody) Using A Refolding System " Protein Eng. 13(8):583-588; Lu, D. et al. (2005) "A Fully Human

Recombinant IgG-Like Bispecific Antibody To Both The Epidermal Growth Factor Receptor And The Insulin-Like Growth Factor Receptor For Enhanced Antitumor Activity," J. Biol. Chem. 280(20): 19665-19672).

[0142] The art has recognized that bispecific diabodies composed of non-covalently associated polypeptides are unstable and readily dissociate into non-functional monomers (see, e.g., Lu, D. et al. (2005) "A Fully Human Recombinant IgG-Like Bispecific Antibody To Both The Epidermal Growth Factor Receptor And The Insulin-Like Growth Factor Receptor For Enhanced Antitumor Activity," J. Biol. Chem. 280(20): 19665-19672).

[0143] In the face of this challenge, the invention provides stable, covalently bonded heterodimeric multispecific diabodies, termed DARTs™ (see, e.g., United States Patent Publications No. 2014-0099318; 2013-0295121; 2010-0174053 and 2009-0060910; European Patent Publication No. EP 2714079; EP 2601216; EP 2376109; EP 2158221 and PCT Publications No. WO 2015/026894; WO2015/026892; WO 2015/021089; WO 2014/159940; WO 2012/162068; WO 2012/018687; WO 2010/080538; Moore, P.A. et al. (2011)

"Application Of Dual Affinity Retargeting Molecules To Achieve Optimal Redirected T-Cell Killing Of B-Cell Lymphoma," Blood 117(17):4542-4551; Veri, M.C. et al. (2010)

"Therapeutic Control OfB Cell Activation Via Recruitment Of Fcgamma Receptor lib (CD32B) Inhibitory Function With A Novel Bispecific Antibody Scaffold," Arthritis Rheum. 62(7): 1933-1943; Johnson, S. et al. (2010) "Effector Cell Recruitment With Novel Fv-Based Dual-Affinity Re-Targeting Protein Leads To Potent Tumor Cytolysis And in vivo B-Cell Depletion " J. Mol. Biol. 399(3):436-449), the contents of which publications are herein incorporated by reference in their entirety). Such multispecific molecules comprise two or more covalently complexed polypeptides and involve engineering one or more cysteine residues into each of the employed polypeptide species. For example, the addition of a cysteine residue to the C-terminus of such constructs has been shown to allow disulfide bonding between the polypeptide chains, stabilizing the resulting heterodimer without interfering with the binding characteristics of the bivalent molecule.

[0144] The invention provides multispecific, antibody-based molecules that can bind at least two distinct antigens simultaneously, wherein at least one of the antigens is comprised in an HIV-1 envelope. In certain aspects, the present invention is directed to HIV-1 multispecific molecules that are capable of simultaneous binding to an epitope of HIV-1 envelope and an epitope of an antigen on a number of effector cells, e.g. but not limited to an effector cell which expresses CD2, CD3, CD8, CD16, T-Cell Receptor (TCR), KG2D, etc. epitope) and to the uses of such molecules in the treatment of HIV-1 infection.

[0145] In certain embodiments the invention provides molecules with dual targeting specificity (including but not limited to bispecific antibodies, bispecific diabodies, and trivalent binding molecules). In certain aspects the invention provides bispecific molecules that are capable of localizing an immune effector cell to an HIV-1 envelope expressing cell, so as facilitate the killing of the HIV-1 envelope expressing cell. In this regard, bispecific molecules bind with one "arm" to an epitope of a surface antigen on target cells, and with the second "arm" to an epitope of an activating, invariant component of the T cell receptor (TCR) complex. The simultaneous binding of such a bispecific molecule to both of its targets will force a temporary interaction between target cell and T cell, causing activation of any cytotoxic T cell and subsequent lysis of the target cell. Hence, the immune response is redirected to the target cells and is independent of peptide antigen presentation by the target cell or the specificity of the T cell as would be relevant for normal MHC-restricted activation of CTLs. In this context it is crucial that CTLs are only activated when a target cell is presenting the bispecific molecule to them, i.e. the immunological synapse is mimicked. Particularly desirable are bispecific molecules that do not require lymphocyte preconditioning or co- stimulation in order to elicit efficient lysis of target cells. In certain embodiments, such molecule may further comprise a third binding "arm" and be trivalent. In some embodiments, the third arm binds to an epitope of a surface antigen on target cells, which may be the same epitope or a different epitope as bound by the first arm. In some embodiments, the third arm binds to an epitope of an activating, invariant component of the TCR complex, which may be the same epitope or a different epitope as bound by the second arm. In alternative

embodiments, the third arm binds to a different epitope to that bound by the first arm or second arm, such as an epitope of a surface antigen on target cells or an epitope expressed on the surface of an effector cell (e.g., but not limited to an epitope of CD2, CD3, CD8, CD 16, T-Cell Receptor (TCR), NKG2D, etc.).

[0146] In certain embodiments, such bispecific molecules comprise one portion which targets HIV-1 envelope and a second portion which binds a second target. In certain embodiments, the first portion comprises VH and VL sequences, or CDRs from the antibodies described herein. In certain embodiments, the second target could be, for example but not limited to an effector cell. In certain embodiments the second portion is a T-cell engager. In certain embodiments, the second portion comprises a sequence/paratope which targets CD3. In certain embodiments, the second portion is an antigen-binding region derived from a CD3 antibody, optionally a known CD3 antibody. In certain embodiments, the anti-CD antibody induces T cell-mediated killing. In certain embodiments, the bispecific molecules comprise whole antibodies. In other embodiments, the dual targeting bispecific molecules consist essentially of Fab fragments. In other embodiments, the dual targeting bispecific molecules comprise a heavy chain constant region (CHI). In certain embodiments, the bispecific molecule does not comprise Fc Domain. In certain embodiments, the bispecific molecules have improved effector function. In certain embodiments, the bispecific molecules have improved cell killing activity. Various methods and platforms for design of bispecific molecules (including but not limited to bispecific antibodies, bispecific diabodies, etc.) are known in the art. See for example US Pub. 20140206846, US Pub. 20140170149,

20100174053, US Pub. 20090060910, US Pub 20130295121, US Pub. 20140099318, US Pub. 20140088295 which contents are herein incorporated by reference in their entirety.

[0147] A bispecific or bifunctional molecule is an artificial hybrid antibody that can comprise two different heavy /light chain pairs and two different binding sites (see, e.g., Romain Rouet & Daniel Christ "Bispecific antibodies with native chain structure" Nature Biotechnology 32, 136-137 (2014); Byrne et al. "A tale of two specificities: bispecific antibodies for therapeutic and diagnostic applications" Trends in Biotechnology, Volume 31, Issue 11, November 2013, Pages 621-632 Songsivilai and Lachmann, Clin. Exp. Immunol., 79:315-321 (1990);

Kostelny et al., J. Immunol. 148: 1547-53 (1992) (and references therein)). In certain embodiments the bispecific molecule is a whole antibody of any isotype. In other embodiments the bispecific molecule is a bispecific fragment, for example but not limited to F(ab)₂ fragment. In some embodiments, the bispecific molecules do not include Fc portion, which makes these bispecific molecules relatively small in size and easy to penetrate tissues.

[0148] In some embodiments, the invention encompasses polypeptide chains, each of which polypeptide chains comprise a VH and VL domain, comprising CDRs as described herein. In certain embodiments, the VL and VH domains comprising each polypeptide chain have the same specificity, and the multimer molecule is bivalent and monospecific. In other embodiments, the VL and VH domains comprising each polypeptide chain have differing specificity and the multimer is bivalent and bispecific. In still other embodiments, the VH and VL domains of each polypeptide chain have differing specificity and the multimer is trivalent and bispecific or trivalent and trispecific.

[0149] The multispecific molecules of the invention can simultaneously bind two separate and distinct epitopes. In certain embodiments the epitopes are from the same antigen. In other embodiments, the epitopes are from different antigens. In non-limiting embodiments at least one epitope binding site is specific for a determinant expressed on an immune effector cell (e.g. CD3, CD16, CD32, CD64, etc.) which are expressed on T lymphocytes, natural killer (NK) cells or other mononuclear cells. In one embodiment, the multispecific molecule binds to the effector cell determinant and also activates the effector cell. In this regard, multispecific molecules of the invention (e.g., bispecific antibodies, bispecific diabodies, trivalent binding molecules, etc.) may exhibit Ig-like functionality independent of whether they further comprise an Fc domain (e.g., as assayed in any effector function assay known in the art or exemplified herein).

[0150] In certain embodiments, the multispecific molecule comprises an HIV-1 envelope binding fragment, for example but not limited to an HIV-1 envelope binding fragment from any of the antibodies described herein. In other embodiments, the multispecific molecule further comprises a second antigen-interaction-site/fragment. In other embodiments, the multispecific molecule further comprises at least one effector cell targeting arm.

[0151] In certain embodiments the multispecific molecules engage cells for Antibody- Dependent Cell-mediated Cytotoxicity (ADCC). In certain embodiments the multispecific molecules engage natural killer cells, neutrophil polymorphonuclear leukocytes, monocytes and macrophages. In certain embodiments the multispecific molecules are T-cell engagers. In certain embodiments, the bispecific molecule comprises an HIV-1 envelope binding fragment and CD3 binding fragment. Various CD3 antibodies are provided herein (see, e.g., Table 4) and others are known in the art. See for example US Patent 8,784,821, and United States Patent Publications No. 2014-0099318 providing various disclosure on various CD3 antibodies, which disclosure is incorporated by reference in its entirety. In certain

embodiments, the bispecific molecule comprises an HIV-1 envelope binding fragment and CD 16 binding fragment. Various CD 16 antibodies are provided herein (see e.g., Table 4) and others are known in the art. See for example WO 03/101485, which disclosure is

incorporated by reference in its entirety.

[0152] In certain embodiments, the invention provides molecules or fragments comprising a CDR(s) of the VH and/or VL chains, or VH and/or VL chains of any suitable HIV-1 antibody, as the HIV-1 binding arm(s) of multispecific molecules, e.g., but not limited to bispecific antibodies, bispecific diabodies, trivalent binding molecules, etc., or toxin labeled HIV-1 binding molecules. Exemplary HIV-1 antibodies are provided in Table 4.

[0153] The invention also includes variants of the antibodies (and fragments) disclosed herein, including variants that retain the ability to bind to recombinant Env protein, the ability to bind to the surface of virus-infected cells and/or ADCC-mediating properties of the antibodies specifically disclosed, and methods of using same to, for example, reduce HIV-1 infection risk. Combinations of the antibodies, or fragments thereof, disclosed herein can also be used in the generation of the multispecific molecules of the invention.

[0154] Homologs and variants of a VL or a VH of an antibody that specifically binds a polypeptide are typically characterized by possession of at least about 75%, for example at least about 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% sequence identity counted over the full length alignment with the amino acid sequence of interest.

Proteins with even greater similarity to the reference sequences will show increasing percentage identities when assessed by this method, such as at least 80%, at least 85%, at least 90%), at least 95%, at least 98%, or at least 99% sequence identity. When less than the entire sequence is being compared for sequence identity, homologs and variants will typically possess at least 80% sequence identity over short windows of 10-20 amino acids, and may possess sequence identities of at least 85% or at least 90% or 95% depending on their similarity to the reference sequence. Methods for determining sequence identity over such short windows are available at the NCBI website on the internet. One of skill in the art will appreciate that these sequence identity ranges are provided for guidance only; it is entirely possible that strongly significant homologs could be obtained that fall outside of the ranges provided. [0155] In certain embodiments, the invention provides multispecific molecules comprising the VL and VH domains of antibodies which are 99%, 98%, 97%, 96%, 95%, 94%, 93%, 92%, 91%, 90%, 89%, 88%, 87%, 86%, 85%, 84%, 83%, 82%, 81%, 80% identical to the VH and VL amino acid sequences of the antibodies described herein and still maintain their epitope binding breadth and/or potency. In certain embodiments, the invention provides multispecific molecules comprising the CDR 1 , 2, and/or 3 of the VH and CDR1, 2, and/or 3 of the VL which are 99%, 98%, 97%, 96%, 95%, 94%, 93%, 92%, 91%, 90%, 89%, 88%, 87%, 86%, 85%, 84%, 83%, 82%, 81%, 80% identical to the CDR1, 2, and/or 3 of VH and CDR1, 2, and/or 3 VL amino acid sequences of the antibodies described herein and still maintain their epitope binding breadth and/or potency.

[0156] In certain embodiments, the invention provides multispecific molecules comprising polypeptide chains which are 99%, 98%, 97%, 96%, 95%, 94%, 93%, 92%, 91%, 90%, 89%, 88%, 87%, 86%, 85%, 84%, 83%, 82%, 81%, 80% identical to SEQ ID NOs: 555, 557, 559, 561, 562, 563, 596, or 597.

[0157] In certain embodiments, the invention provides multispecific molecules comprising polypeptide chains which are 99%, 98%, 97%, 96%, 95%, 94%, 93%, 92%, 91%, 90%, 89%, 88%, 87%, 86%, 85%, 84%, 83%, 82%, 81%, 80% identical to SEQ ID NOs: 500-597.

[0158] In some aspects the invention provides recombinant, multispecific molecules, polyclonal or monoclonal antibodies, variants, fusion proteins comprising an antibody portion with an antigen recognition site of the required specificity, humanized antibodies, and chimeric antibodies, and any other modified configuration of the immunoglobulin molecule that comprises an antigen recognition site of the required specificity. Throughout this application, the numbering of amino acid residues of the light and heavy chains of antibodies is according to the EU index as in Kabat et al. (1992) SEQUENCES OF PROTEINS OF

IMMUNOLOGICAL INTEREST, National Institutes of Health Publication No. 91-3242. Amino acids from the Variable Domains of the mature heavy and light chains of immunoglobulins are designated by the position of an amino acid in the chain. Kabat described numerous amino acid sequences for antibodies, identified an amino acid consensus sequence for each subgroup, and assigned a residue number to each amino acid. The Kabat numbering scheme is extendible to antibodies not included in his compendium by aligning the antibody in question with one of the consensus sequences in Kabat by reference to conserved amino acids. This method for assigning residue numbers has become standard in the field and readily identifies amino acids at equivalent positions in different antibodies, including chimeric or humanized variants.

[0159] In some embodiments, antigen-binding fragment of an antibody is a portion of an antibody that possesses an at least one antigen recognition site. Fragments include for example but not limited to, Fab, Fab', F(ab')₂ Fv), and single chain (scFv).

[0160] In certain embodiments the invention provides recombinant molecules. In certain embodiments, recombinant molecules encompasses not only intact monoclonal antibodies and full-length monoclonal antibodies, but also fragments thereof (such as Fab, Fab', F(ab')₂ Fv), single chain (scFv), mutants thereof, fusion proteins and multispecific molecules comprising an antibody portion, humanized monoclonal antibodies, chimeric monoclonal antibodies, and any other modified configuration of the immunoglobulin molecule that comprises an antigen recognition site of the required specificity and the ability to bind to an antigen. Recombinant molecules are not limited as regards to the source of the molecule or the manner in which it is made (e.g., by hybridoma, phage selection, recombinant expression, transgenic animals, etc.).

[0161] Methods of making recombinant molecules are known in the art. In certain embodiments, the molecules are produced recombinantly by any means known in the art. In one embodiment, the polynucleotide sequence encoding such a recombinant molecule is cloned into a vector for expression or propagation. The sequence encoding an antibody of interest may be maintained in a vector in a host cell and the host cell can then be expanded and frozen for future use. The polynucleotide sequence of such antibodies may be used for genetic manipulation to generate the multispecific molecules of the invention (e.g., bispecific antibodies, bispecific diabodies, trivalent binding molecules, etc.) as well as a chimeric antibody, a humanized antibody, or a caninized antibody, to improve the affinity, or other characteristics of the antibody. The general principle in humanizing an antibody involves retaining the basic sequence of the antigen-binding portion of the antibody, while swapping the non-human remainder of the antibody with human antibody sequences. There are four general steps to humanize a monoclonal antibody. These are: (1) determining the nucleotide and predicted amino acid sequence of the starting antibody light and heavy variable Domains (2) designing the humanized antibody or caninized antibody, i.e., deciding which antibody framework region to use during the humanizing or canonizing process (3) the actual humanizing or caninizing methodologies/techniques and (4) the transfection and expression of the humanized antibody. See, for example, U.S. Patents Nos. 4,816,567; 5,807,715;

5,866,692; and 6,331,415. [0162] The antibodies described herein, or fragments thereof, or molecules comprising such fragments, may be recombinantly produced in prokaryotic or eukaryotic expression systems. These systems are well described in the art. In general, protein therapeutics are produced from mammalian cells. The most widely used host mammalian cells are Chinese hamster ovary (CHO) cells and mouse myeloma cells, including NSO and Sp2/0 cells. Two derivatives of the CHO cell line, CHO-K1 and CHO pro-3, gave rise to the two most commonly used cell lines in large scale production, DUKX-X11 and DG44. {See, e.g., Kim, J., et al., "CHO cells in biotechnology for production of recombinant proteins: current state and further potential," Appl. Microbiol. Biotechnol., 2012, 93 :917-30, which is hereby incorporated-by-reference.) Other mammalian cell lines for recombinant protein expression include, but are not limited to, COS, HeLa, HEK293T, U20S, A549, HT1080, CAD, P19, NIH 3T3, L929, N2a, HEK 293, MCF-7, Y79, SO-Rb50, HepG2, J558L, and BHK. If the aim is large-scale production, the most currently used cells for this application are CHO cells. Guidelines to cell engineering for mAbs production were also reported. (Costa et al.,

"Guidelines to cell engineering for monoclonal antibody production," Eur JPharm

Biopharm, 2010, 74: 127-38, which is hereby incorporated-by-reference.) Using heterologous promoters, enhancers and amplifiable genetic markers, the yields of antibody and antibody fragments can be increased. Similar methods are utilized for the expression of the

multispecific molecules of the invention. Thus, in certain embodiments, the invention provides an antibody, or antibody fragment, or molecule comprising such fragment, that is recombinantly produced from a mammalian cell-line, including a CHO cell-line. In certain embodiments, the invention provides a composition comprising an antibody, or antibody fragment, or molecule comprising such fragment, wherein the antibody, or antibody fragment, or molecule comprising such fragment was recombinantly produced in a mammalian cell-line, and wherein the antibody, or antibody fragment, or molecule comprising such fragment is present in the composition at a concentration of at least 1, 10, 100, 1000 micrograms/mL, or at a concentration of at least 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 20, 50, or 100 milligrams/mL.

[0163] Furthermore, large-scale production of therapeutic-grade molecules are much different than those for laboratory scale. There are extreme purity requirements for therapeutic- grade. Large-scale production of therapeutic-grade molecules requires multiples steps, including product recovery for cell-culture harvest (removal of cells and cell debris), one or more chromatography steps for antibody purification, and formulation (often by tangential filtration). Because mammalian cell culture and purification steps can introduce variants that are unique to the recombinant production process (i.e., protein aggregates, N- and C- terminal variants, acidic variants, basic variants, different glycosylation profiles), there are recognized approaches in the art for analyzing and controlling these variants. (See, Fahrner, et al., Industrial purification of pharmaceutical antibodies: Development, operation, and validation of chromatography processes, Biotech. Gen. Eng. Rev., 2001, 18:301-327, which is hereby incorporated-by-reference.) In certain embodiments of the invention, the composition comprises less than 1, 2, 3, 4, 5, 10, 15, 20, 25, 30, 35, 50, or 100 nanograms of host cell protein (i.e., proteins from the cell-line used to recombinantly produce the molecule). In other embodiments, the composition comprises less than 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 15, 20, or 25 ng of protein A per milligram of molecule (i.e., protein A is a standard approach for purifying antibodies and other Fc bearing molecules from recombinant cell culture, but steps should be done to limit the amount of protein A in the composition, as it may be immunogenic). (See, e.g., U.S. Patent No. 7,458,704, Reduced protein A leaching during protein A affinity chromatography; which is hereby incorporated-by-reference.)

[0164] Multispecific molecules

[0165] Various non-limiting multispecific molecule designs are provides in Figures 1-9.

[0166] As provided herein, the invention contemplates designs of multispecific molecules, which include, but are not limited to bispecific antibodies, bispecific diabodies, Fc Domain bearing diabodies, trivalent binding molecules, Fc Domain bearing trivalent bind molecules etc. The multispecific molecules provided herein comprise various domains, including, but not limited to peptide linkers, Heterodimer Promoting Domains, VL and VH domains, and Fc Domains. Specific non-limiting embodiments of exemplary multispecific molecules are provided herein. Alternative combinations of the various domains described herein can be employed in the multispecific molecules of the invention.

[0167] As provided herein, the invention contemplates designs of multispecific molecules with various peptide linkers (also referred to herein as "intervening peptide linkers") separating the domains comprised in the polypeptide chains. Any of a variety of peptide linkers can be used to separate the domains in the polypeptide chains of the multispecific molecules of the invention. Typically, such peptide linkers will comprise 1-20, 1-19, 1-18, 1- 17, 1-16, 1-15, 1-14, 1-13, 1-12, 1-11, 1-10, 1-9, 1-8, 1-7, 1-6, 1-5, 1-4, 1-3, 1-2, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, or 20 amino acid residues. Such polypeptide linkers can include a series of glycine residues (Gly) and/or Serine (Ser) residues and may optionally comprise cysteine residue(s). Specific non-limiting embodiments of exemplary polypeptide linkers (e.g., Peptide Linker 1, Peptide Linker 2, Spacer Linker 3, etc.) are provided herein. Alternative peptide linkers are well-known in the art and can be employed in the multispecific molecules of the invention. Other linkers can be readily determined. Some additional examples of linkers are disclosed in US Pub 20100174053, incorporated by reference in its entirety.

[0168] Typically, the VH and VL domains of a polypeptide chain of the multispecific molecules are linked so that they do not associate with each other. In certain embodiments the length of the peptide linker, which separates such VL and VH domains of a polypeptide chain is selected to substantially or completely prevent such VL and VH domains from binding to one another. Thus the VL and VH domains of a polypeptide chain are

substantially or completely incapable of binding to one another. In certain embodiments this is due to the peptide linker which separates the VH and VL domains. As provided herein, the invention also contemplates designs of multispecific molecules wherein the domains comprising the polypeptide chains (e.g., Heterodimer Promoting Domains, VL and VH domains, and Fc Domains etc.) are directly linked (i.e. no peptide linker is used between the domain). In such multispecific molecules the domains can be linked by a peptide bond.

[0169] One embodiment of the present invention relates to multispecific molecules, which are bispecific, that are capable of binding to a "first epitope" and a "second epitope," such epitopes not being identical to one another. Such bispecific molecules comprise "VL1" / "VH1" domains that are capable of binding to the first epitope, and "VL2" / "VH2" domains that are capable of binding to the second epitope. The notation "VL1" and "VH1" denote respectively, the Variable Light Chain Domain and Variable Heavy Chain Domain of such bispecific molecules that bind the "first" epitope. Similarly, the notation "VL2" and "VH2" denote respectively, the Light Chain Variable Domain and Heavy Chain Variable Domain of such bispecific molecules that bind the "second" epitope. It is irrelevant whether a particular epitope is designated as the first vs. the second epitope; such notation having relevance only with respect to the presence and orientation of domains of the polypeptide chains of such multispecific molecules of the present invention. In one embodiment, one of such epitopes is an epitope of HIV-1 Env (for example but not limited to a V3 glycan and/or a CD4 binding site epitope), and the other is an epitope of a molecule that is not HIV-1 Env. In particular embodiments, one of such epitopes is an epitope of HIV-1 Env and the other is an epitope of a molecule (e.g., CD2, CD3, CD8, CD16, T-Cell Receptor (TCR), KG2D, etc.) present on the surface of an effector cell, for example but not limited to a T lymphocyte, a natural killer (NK) cell or other mononuclear cell (see, e.g., Figures 1 and 3A-3B). In certain embodiments, a bispecific molecule comprises more than two epitope-binding sites (see, e.g., Figures 2A-2C). Such bispecific molecules will bind at least one epitope of HIV-1 Env and at least one epitope of a molecule that is not HIV-1 Env.

[0170] One embodiment of the present invention also relates to trivalent binding molecules that are capable of binding to a "first epitope," a "second epitope," and a "third epitope," wherein at least one of such epitopes is not identical to another. Such trivalent binding molecules comprise VLl / VHl domains that are capable of binding to the first epitope, VL2 / VH2 domains that are capable of binding to the second epitope, and further comprise "VL3" / "VH3" domains that are capable of binding to the third epitope, wherein the notation "VL3" and " VH3" denote respectively, the Variable Light Chain Domain and Variable Heavy Chain Domain of such trivalent binding molecules that bind the "third" epitope. The capacity to bind a third epitope provides additional and/or enhanced functionality. In one embodiment, one (or two) of such epitopes is an epitope of HIV-1 Env (particularly aV3 glycan or a CD4 binding site epitope), and two (or one) of such epitopes is an epitope of a molecule that is not HIV-1 Env. In particular embodiments, one (or two) of such epitopes is an epitope of HIV-1 Env and two (or one) of such epitopes is an epitope of a molecule (e.g., CD2, CD3, CD8, CD 16, T-Cell Receptor (TCR), KG2D, etc.) present on the surface of an effector cell, such as a T lymphocyte, a natural killer (NK) cell or other mononuclear cell (see, e.g., Figures 4A- 4F). Such trivalent binding molecules will bind at least one epitope of HIV-1 Env and at least one epitope of a molecule that is not HIV-1 Env, and may bind two epitopes of HIV-1 Env and one epitope of a molecule that is not HIV-1 Env or may bind one epitope of HIV-1 Env and two epitopes that are not epitopes of HIV-1 Env.

[0171] In some embodiments, such molecules comprise two polypeptide chains, wherein each of the two polypeptide chains comprises three Domains (Figure 1). The first polypeptide chain comprises: (i) a Domain that comprises a binding region of a light chain variable Domain of a first immunoglobulin (VLl), (ii) a second Domain that comprises a binding region of a heavy chain variable Domain of a second immunoglobulin (VH2), and (iii) a third Domain that serves to promote heterodimerization with the second polypeptide chain and to covalently bond the first polypeptide to the second polypeptide chain of the molecule. The second polypeptide chain contains a complementary first Domain (a VL2 Domain), a complementary second Domain (a VHl Domain) and a third Domain that complexes with the third Domain of the first polypeptide chain in order to promote heterodimerization and covalent bonding with the first polypeptide chain. Such molecules are stable, potent and have the ability to simultaneously bind two or more antigens. They are able to promote redirected T cell (CD3) or K (CD16) cell mediated killing of cells expressing target antigens.

[0172] In certain embodiments, the HIV-1 multispecific molecules of the present invention are composed of two polypeptide chains which associate with one another to form one binding site specific for an epitope of HIV-1, and one binding site specific for an epitope of CD2, CD3, CD8, CD16, T-Cell Receptor (TCR), KG2D, etc., so as to be capable of simultaneously binding to HIV-1 and, for example, to CD3. Thus, such diabodies bind to a "first epitope," which may be either an epitope of CD3 or HIV-1, and a "second epitope," which is an epitope of HIV-1 when the first epitope is an epitope of CD3, and is an epitope of CD3 when the first epitope is from HIV-1. Alternatively, such diabodies bind to a "first epitope," which may be either an epitope of CD16 or HIV-1, and a "second epitope," which is an epitope of HIV-1 when the first epitope is of CD 16, and is an epitope of CD 16 when the first epitope is an epitope of HIV-1.

[0173] In certain embodiments, the first of such two polypeptide chains will contain, in the N- terminal to C-terminal direction, an N-terminus, the Antigen-Binding Domain of a Light Chain Variable Domain (VL) of an antibody that binds to a "first" epitope of a "first" antigen {e.g., either CD3 or HIV-1 envelope), the Antigen-Binding Domain of a Heavy Chain

Variable Domain (VH) of an antibody that binds to a "second" epitope of a "second" antigen (HIV-1, if the first antigen was CD3; CD3, if the first antigen was HIV-1), a

Heterodimerization-Promoting Domain, and a C-terminus. An intervening peptide linker (Peptide Linker 1) separates the Antigen-Binding Domain of the Light Chain Variable Domain from the Antigen-Binding Domain of the Heavy Chain Variable Domain. In certain embodiments the Antigen-Binding Domain of the Heavy Chain Variable Domain is linked to the Heterodimerization-Promoting Domain by an intervening peptide linker (Peptide Linker 2). In certain embodiments the first of the two polypeptide chains will thus contain, in the N- terminal to C-terminal direction: VLpirst Antigen - Peptide Linker 1 - VHs_ec0nd Antigen - Peptide Linker 2 - Heterodimerization-Promoting Domain.

[0174] The terms VL/VH for first and second antigens, VL_{first a}ntigen/VH_{first an}tigen VL_seCond antigen/VH_second antigen, and VLl/VHl and VL2/VH2 are used interchangeably throughout the application. [0175] In certain embodiments, the second of such two polypeptide chains will contain, in the N-terminal to C-terminal direction, an N-terminus, the Antigen-Binding Domain of a Light Chain Variable Domain (VL) of an antibody that binds to the second epitope of the second antigen, the Antigen-Binding Domain of a Heavy Chain Variable Domain (VH) of an antibody that binds to the first epitope of the first antigen, a Heterodimerization-Promoting Domain and a C-terminus. An intervening peptide linker (Peptide Linker 1) separates the Antigen-Binding Domain of the Light Chain Variable Domain from the Antigen-Binding Domain of the Heavy Chain Variable Domain. In certain embodiments, the Antigen-Binding Domain of the Heavy Chain Variable Domain is linked to the Heterodimerization-Promoting Domain by an intervening peptide linker (Peptide Linker 2). In certain embodiments the second of the two polypeptide chains will thus contain, in the N-terminal to C-terminal direction: VL_Second Antigen - Peptide Linker 1 - VH_Fir_St Antigen - Peptide Linker 2 - Heterodimerization-Promoting Domain.

[0176] The Antigen-Binding Domain of the Light Chain Variable Domain of the first polypeptide chain interacts with the Antigen-Binding Domain of the Heavy Chain Variable Domain of the second polypeptide chain in order to form a functional antigen-binding site that is specific for the first antigen (e.g., either HIV-1 envelope or CD3). Likewise, the Antigen- Binding Domain of the Light Chain Variable Domain of the second polypeptide chain interacts with the Antigen-Binding Domain of the Heavy Chain Variable Domain of the first polypeptide chain in order to form a second functional antigen-binding site that is specific for the second antigen (e.g., either CD3 or HIV-1 envelope, depending upon the identity of the first antigen). Thus, the selection of the Antigen-Binding Domain of the Light Chain Variable Domain and the Antigen-Binding Domain of the Heavy Chain Variable Domain of the first and second polypeptide chains are coordinated, such that the two polypeptide chains collectively comprise Antigen-Binding Domains of light and Heavy Chain Variable Domains capable of binding to the intended targets, in certain embodiments e.g. HIV-1 envelope and CD3, or CD 16.

[0177] In certain embodiments the length of Peptide Linker 1, which separates such VL and VH domains of a polypeptide chain is selected to substantially or completely prevent such VL and VH domains from binding to one another. Thus the VL and VH domains of the first polypeptide chain are substantially or completely incapable of binding to one another.

Likewise, the VL and VH domains of the second polypeptide chain are substantially or completely incapable of binding to one another. In certain embodiments this is due to the peptide linker which separates the VH and VL domains. In certain embodiments, the peptide linker is 5, 6, 7, 8, 9, 10, 1 1, 12, 13, 14, 15, 16, 17, 18, 19, 20, but no more than 20 amino acids. In some embodiments, the peptide linker is less than 12 amino acids in length. In certain embodiments an intervening spacer peptide (Peptide Linker 1) has the sequence (SEQ ID NO:508): GGGSGGGG.

[0178] Peptide Linker 2 separates the VH Domain of a polypeptide chain from the

Heterodimer-Promoting Domain of that polypeptide chain. Any of a variety of linkers can be used for the purpose of Peptide Linker 2. The length and composition of Peptide Linker 2 may be selected based on the choice of heterodimer-promoting domains. Typically, the second intervening peptide linker (Peptide Linker 2) will comprise 1-20 amino acid residues. In certain embodiments, where the heterodimer-promoting domains do not comprise a cysteine residue a cysteine-containing second intervening peptide linker (Peptide Linker 2) is utilized. Optionally, both a cysteine-containing Peptide Linker 2 (Peptide Linker 2-C) and a cysteine-containing Heterodimer-Promoting Domain are used. In certain embodiments a sequence for such Peptide Linker 2 has the amino acid sequence: GGCGGG (SEQ ID NO:509), which has a cysteine residue that may be used to covalently bond the first and second polypeptide chains to one another via a disulfide bond. In certain embodiments, a sequence for Peptide Linker 2-C has the amino acid sequence: ASTKG (SEQ ID NO: 510). Peptide Linker 2 and Peptide Linker 2-C could be used interchangeably.

[0179] The formation of heterodimers of the first and second polypeptide chains can be driven by the inclusion of specific sequences referred to as Heterodimer-Promoting Domains (HPDs). Such domains include without limitation GVEPKSC (SEQ ID NO: 511) or VEPKSC (SEQ ID NO:512) on one polypeptide chain and GFNRGEC (SEQ ID NO:513) or FNRGEC (SEQ ID NO:514) on the other polypeptide chain (See US2007/0004909 herein incorporated by reference in its entirety). Various HPD sequences are contemplated by the invention and disclosed in the specification. Table 4 discloses non-limiting embodiments of various HPDs. In some embodiments the HPD include E/K-coils (SEQ ID NOs: 518, 520) or cysteine engineered E/K-coils (SEQ ID NOs: 519, 521). In some embodiments HPD includes combinations of SEQ ID NOs: 51 1, 512, 513, and 514 (e.g., SEQ ID NOs: 51 1 and 513; SEQ ID NOs: 512 and 513; SEQ ID NOs: 51 1 and 514; SEQ ID NOs: 512 and 514). In some embodiments HPDs include any suitable sequences with a Cysteine residue to permit disulfide bond. In some embodiments HPDs includes suitable CHI and CL domains (See for example CH domain (SEQ ID NO: 515) and kappa and lambda light chain constant domains (SEQ ID NOs: 516 and 517).

[0180] In certain embodiments, the Heterodimer-Promoting Domains of the present invention are formed from one, two, three or four tandemly repeated coil domains of opposing charge that comprise a sequence of at least six, at least seven or at least eight charged amino acid residues (Apostolovic, B. et al. (2008) "pH-Sensitivity of the E3/K3 Heterodimeric Coiled Coil " Biomacromolecules 9:3173-3180; Arndt, K M. et al. (2001) "Helix-stabilized Fv (hsFv) Antibody Fragments: Substituting the Constant Domains of a Fab Fragment for a Heterodimeric Coiled-coil Domain " J. Molec. Biol. 312:221-228; Arndt, K M. et al. (2002) "Comparison of In Vivo Selection and Rational Design of Heterodimeric Coiled Coils " Structure 10: 1235-1248; Boucher, C. et al. (2010) "Protein Detection By Western Blot Via Coiled-Coil Interactions " Analytical Biochemistry 399: 138-140; Cachia, P.J. et al. (2004) "Synthetic Peptide Vaccine Development: Measurement Of Polyclonal Antibody Affinity And Cross-Reactivity Using A New Peptide Capture And Release System For Surface Plasmon Resonance Spectroscopy " J. Mol. Recognit. 17:540-557; De Crescenzo, G.D. et al. (2003) "Real-Time Monitoring of the Interactions of Two-Stranded de novo Designed Coiled-Coils: Effect of Chain Length on the Kinetic and Thermodynamic Constants of Binding "

Biochemistry 42: 1754-1763; Fernandez-Rodriquez, J. et al. (2012) "Induced

Heterodimerization And Purification Of Two Target Proteins By A Synthetic Coiled-Coil Tag," Protein Science 21 :511-519; Ghosh, T.S. et al. (2009) "End-To-End And End-To- Middle Interhelical Interactions: New Classes Of Interacting Helix Pairs In Protein

Structures " Acta Crystallographica D65: 1032-1041; Grigoryan, G. et al. (2008) "Structural Specificity In Coiled-Coil Interactions " Curr. Opin. Struc. Biol. 18:477-483; Litowski, J.R. et al. (2002) "Designing Heterodimeric Two-Stranded a-Helical Coiled-Coils: The Effects Of Hydrophobicity And a-Helical Propensity On Protein Folding, Stability, And Specificity " J. Biol. Chem. 27 ' -31212-3121% Steinkruger, J.D. et al. (2012) "The d'-d-d' Vertical Triad is Less Discriminating Than the a'— a— a' Vertical Triad in the Antiparallel Coiled-coil Dimer Motif" J. Amer. Chem. Soc. 134(5):2626-2633; Straussman, R. et al. (2007) "Kinking the Coiled Coil— Negatively Charged Residues at the Coiled-coil Interface " J. Molec. Biol. 366: 1232-1242; Tripet, B. et al. (2002) "Kinetic Analysis of the Interactions between

Troponin C and the C-terminal Troponin I Regulatory Region and Validation of a New Peptide Delivery/Capture System used for Surface Plasmon Resonance " J. Molec. Biol. 323 :345-362; Woolfson, D.N. (2005) "The Design Of Coiled-Coil Structures And Assemblies " Adv. Prot. Chem. 70:79-112; Zeng, Y. et al. (2008) "A Ligand-Pseudoreceptor System Based On de novo Designed Peptides For The Generation Of Adenoviral Vectors With Altered Tropism," J. Gene Med. 10:355-367).

[0181] Such repeated coil domains may be exact repeats or may have substitutions. For example, the Heterodimer-Promoting Domain of the first polypeptide chain may comprise a sequence of eight negatively charged amino acid residues and the Heterodimerization- Promoting Domain of the second polypeptide chain may comprise a sequence of eight negatively charged amino acid residues. It is immaterial which coil is provided to the first or second polypeptide chains, provided that a coil of opposite charge is used for the other polypeptide chain.

[0182] In certain embodiments a multispecific molecule of the present invention has a first polypeptide chain having a negatively charged coil. The positively charged amino acid may be lysine, arginine, histidine, etc. and/or the negatively charged amino acid may be glutamic acid, aspartic acid, etc. In certain embodiments the positively charged amino acid is lysine and/or the negatively charged amino acid is glutamic acid. It is possible for only a single Heterodimer-Promoting Domain to be employed (since such domain will inhibit

homodimerization and thereby promote heterodimerization). In certain embodiments both the first and second polypeptide chains of the multispecific molecules of the present invention contain Heterodimer-Promoting Domains.

[0183] In certain embodiments, one of the Heterodimer-Promoting Domains will comprise four tandem "E-coil" helical domains (SEQ ID NO:518 (EVAALEK-EVAALEK- EVAALEK-EVAALEK)), whose glutamate residues will form a negative charge at pH 7, while the other of the Heterodimer-Promoting Domains will comprise four tandem "K-coil" domains (SEQ ID NO:520 (KVAALKE-KVAALKE-KVAALKE-KVAALKE)), whose lysine residues will form a positive charge at pH 7. The presence of such charged domains promotes association between the first and second polypeptides, and thus fosters

heterodimerization. In other embodiments, one of the four tandem "E-coil" helical domains of SEQ ID NO: 518 has been modified to contain a cysteine residue: EVAA EK- EVAALEK-EVAALEK-EVAALEK (SEQ ID NO: 519). In some embodiments, a

Heterodimer-Promoting Domain in which one of the four tandem "K-coil" helical domains of SEQ ID NO: 520 has been modified to contain a cysteine residue: KVAA£KE-KVAALKE- KVAALKE-KVAALKE (SEQ ID NO: 521). Such cysteine modified Heterodimer- Promoting Domains may be used to covalently bond the first and second polypeptide chains to one another via a disulfide bond.

[0184] In some embodiments, the number of K coil and E coil domains can vary and a skilled artisan can readily determine whether a different number of K-coil or E-coil domain lead to heterodimerization.

[0185] In certain embodiments, the multispecific molecules of the invention, for example but not limited to bispecific monovalent diabodies and trivalent binding molecules, are engineered so that their first and second polypeptide chains covalently bond to one another via one or more cysteine residues positioned along their length. Such cysteine residues may be introduced into the intervening peptide linker that separates the VL and VH domains of the polypeptides. Alternatively, as provided above Peptide Linker 2 and/or the HPDs may contain such cysteine residues.

[0186] The invention also includes variants of the multispecific molecules, or fragments thereof disclosed herein, including variants that retain the ability to bind to recombinant Env protein, the ability to bind to the surface of virus-infected cells and/or ADCC-mediating properties of the antibodies specifically disclosed, and methods of using same to, for example, reduce HIV-1 infection risk. Combinations of the multispecific molecules, antibodies, or fragments thereof, disclosed herein can also be used in the methods of the invention.

[0187] Formation of multispecific molecule as described herein requires the interaction of differing polypeptide chains. Such interactions are difficult to achieve with efficiency within a single cell recombinant production system, due to the many variants of potential chain mispairings. One solution to decrease the probability of mispairings, is to engineer "knobs- into-holes" type mutations into the desired polypeptide chain pairs. Such mutations favor heterodimerization over homodimerization. For example, with respect to Fc-Fc-interactions, an amino acid substitution (preferably a substitution with an amino acid comprising a bulky side group forming a knob e.g., tryptophan) can be introduced into the CH2 or CH3 domain such that steric interference will prevent interaction with a similarly mutated domain and will obligate the mutated domain to pair with a domain into which a complementary, or accommodating mutation has been engineered, i.e., 'the hole' (e.g., a substitution with glycine). Such sets of mutations can be engineered into any pair of polypeptides comprising the multispecific molecule, and further, engineered into any portion of the polypeptides chains that comprise a multispecific molecule. Methods of protein engineering to favor

heterodimerization over homodimerization are well known in the art, in particular with respect to the engineering of immunoglobulin-like molecules, and are encompassed herein (see e.g., Ridgway et al. (1996) " Knobs-Into-Holes^' Engineering Of Antibody CH3 Domains For Heavy Chain Heterodimerization," Protein Engr. 9:617-621, Atwell et al. (1997) "Stable Heterodimers From Remodeling The Domain Interface Of A Homodimer Using A Phage Display Library," J. Mol. Biol. 270: 26-35, and Xie et al. (2005) "A New Format Of

Bispecific Antibody: Highly Efficient Heterodimerization, Expression And Tumor Cell Lysis," J. Immunol. Methods 296:95-101; each of which is hereby incorporated herein by reference in its entirety).

[0188] In some embodiments the invention provides multispecific molecules comprising variant Fc domain (or portions thereof), which variant Fc domain comprises at least one amino acid modification (e.g. substitution, insertion deletion) relative to a comparable wild- type Fc domain or hinge-Fc domain (or portion thereof). Molecules comprising variant Fc domains or hinge-Fc domains (or portion thereof) (e.g., antibodies) normally have altered phenotypes relative to molecules comprising wild-type Fc domains or hinge-Fc domains or portions thereof. The variant phenotype may be expressed as altered serum half-life, altered stability, altered susceptibility to cellular enzymes or altered effector function as assayed in an NK dependent or macrophage dependent assay. Fc domain variants identified as altering effector function are known in the art. For example International Application WO04/063351, U.S. Patent Application Publications 2005/0037000 and 2005/0064514.

[0189] In some embodiments the invention provides multispecific molecules comprising a hinge domain. The hinge domain be derived from any immunoglobulin isotype or allotype including IgA, IgD, IgG, IgE and IgM. In preferred embodiments, the hinge domain is derived from IgG, wherein the IgG isotype is IgGl, IgG2, IgG3 or IgG4, or an allotype thereof. The hinge domain may be engineered into a polypeptide chain comprising the multispecific molecule together with an Fc domain such that the multispecific molecule comprises a hinge-Fc domain. In certain embodiments, the hinge and Fc domain are independently selected from any immunoglobulin isotype known in the art or exemplified herein. In other embodiments the hinge and Fc domain are separated by at least one other domain of the polypeptide chain, e.g., the VL domain.

[0190] In another aspect, the invention provides multispecific molecules which include Fc domain(s)-- Fc bearing multispecific molecules. While some of the disclosure regarding Fc domain(s) refers to specific designs, a skilled artisan appreciates that the Fc disclosure is pertinent to any Fc bearing design of multispecific molecules, including but not limited to the designs described in Figures 1-9.

[0191] Fc bearing multispecific molecules, for example but not limited to Fc bearing diabodies are heavier, and could bind neonatal Fc receptor, increasing their circulating half- life. See Garber "Bispecific antibodies rise again" Nature Reviews Drug Discovery 13, 799- 801 (2014), Figure la; See US Pub 20130295121, US Pub 20140099318 incorporated by reference in their entirety. In certain embodiments, the invention encompasses multispecific molecules comprising an Fc domain or portion thereof (e.g. a CH2 domain, or CH3 domain). The Fc domain or portion thereof may be derived from any immunoglobulin isotype or allotype including, but not limited to, IgA, IgD, IgG, IgE and IgM. In some embodiments, the Fc domain (or portion thereof) is derived from IgG. In some embodiments, the IgG isotype is IgGl, IgG2, IgG3 or IgG4 or an allotype thereof. In some embodiments, the multispecific molecule comprises an Fc domain, which Fc domain comprises a CH2 domain and CH3 domain independently selected from any immunoglobulin isotype (i.e. an Fc domain comprising the CH2 domain derived from IgG and the CH3 domain derived from IgE, or the CH2 domain derived from IgGl and the CH3 domain derived from IgG2, etc.). In some embodiments, the Fc domain may be engineered into a polypeptide chain comprising the multispecific molecule of the invention in any position relative to other domains or portions of the polypeptide chain (e.g., the Fc domain, or portion thereof, may be C-terminal to both the VL and VH domains of the polypeptide of the chain; may be N-terminal to both the VL and VH domains; or may be N-terminal to one domain and C-terminal to another (i.e., between two domains of the polypeptide chain)).

[0192] Other modifications of the multispecific molecules are contemplated to increase the half-life of the molecules. In some embodiments, these modifications include addition of a polypeptide portion of a serum binding protein. See US20100174053 Al, incorporated by reference.

[0193] In some embodiments, the Fc variants of the multispecific molecules of the invention are expected to have increased serum half-life compared to the non-Fc variants. Skilled artisan can readily carry out various assays, including pharmacokinetic studies, to determine the half-life of these molecules.

[0194] In some embodiments, the polypeptide chains in multispecific molecules further comprise an Fc domain. Dimerization of the Fc domains leads to formation of a multispecific molecule that exhibits immunoglobulin-like functionality, i.e., Fc mediated function (e.g., Fc- Fc.gamma.R interaction, complement binding, etc.).

[0195] As provided in Figures 2-3, one or both of the polypeptide chains of bispecific diabodies may additionally comprise the sequence of a CH2-CH3 Domain, such that complexing between the two diabody polypeptides forms an Fc Domain that may be capable of binding to the Fc receptor of cells (such as B lymphocytes, dendritic cells, natural killer cells, macrophages, neutrophils, eosinophils, basophils and mast cells). Similarly, the first and third polypeptide chains of trivalent binding molecules can comprise the sequence of a CH2- CH3 Domain, such that complexing between these two polypeptide chains forms an Fc Domain. As provided in more detail below, the CH2 and/or CH3 Domains of such polypeptide chains need not be identical in sequence, and advantageously are modified to foster complexing between the two polypeptide chains. Many variations of such molecules have been described (see, e.g., United States Patent Publications No. 2014-0099318; 2013- 0295121; 2010-0174053 and 2009-0060910; European Patent Publication No. EP 2714079; EP 2601216; EP 2376109; EP 2158221 and PCT Publications No. WO 2015/026894; WO 2015/026892; WO 2015/021089; WO 2014/159940; WO 2012/162068; WO 2012/018687; WO 2010/080538, the content of each of these publications in herein incorporated by reference in its entirety).

[0196] In some embodiments, Fc -bearing bispecific diabodies may comprise two pairs of polypeptide chains (or four different chains, as provided below). The first and third polypeptide chains of such a bispecific molecule {e.g., diabodies) contain three domains: (i) a VL1 -containing Domain, (ii) a VH2-containing Domain, (iii) Heterodimer-Promoting Domain and (iv) a Domain containing a CH2-CH3 sequence. The second and fourth polypeptide chains contain: (i) a VL2-containing Domain, (ii) a VHl -containing Domain and (iii) a Heterodimer-Promoting Domain, where the Heterodimer-Promoting Domains promote the dimerization of the first/third chains with the second/fourth chains. The VL and/or VH Domains of the third and fourth polypeptide chains, and VL and/or VH Domains of the first and second polypeptide chains may be the same or different so as to permit tetravalent binding that is either monospecific, bispecific or tetraspecific. Such molecules are tetravalent and have enhanced potency. The general structure of the polypeptide chains of a

representative four-chain Fc Domain-containing multispecific molecules of invention is provided in Table 1: Table 1

i_fi T_tera_specc_iifi B_specc 2^nd Chain NH₂-VL2-VH 1 -Heterodimer-Promoting Domain-COOH

1^st Chain NH₂-VLl-VH2-Heterodimer-Promoting Domain-CH2-CH3-COOH

2^nd Chain NH₂-VL2-VH 1 -Heterodimer-Promoting Domain-COOH

1^st Chain NH₂-VLl-VH2-Heterodimer-Promoting Domain-CH2-CH3-COOH

3^rd Chain NH₂-VL3-VH4-Heterodimer-Promoting Domain-CH2-CH3-COOH

4^th Chain NH₂-VL4-VH3 -Heterodimer-Promoting Domain-COOH

[0197] HIV-1 bispecific bivalent Fc bearing diabodies can be composed of two pairs of polypeptide chains (i.e., two first polypeptide chain and two second polypeptide chains) which associate with one another to form two binding sites specific for an epitope of HIV-1 and two binding sites specific for an epitope, for example but not limited to CD3 (see,

Figures 2A-2C), so as to be capable of simultaneously binding to HIV-1 and to CD3. Thus such molecules binding to a "first" epitope, on a "first" antigen which may be either CD3 or HIV-1, and a "second " epitope on a "second" antigen, which is HIV-1 when the first epitope is CD3, and is CD3 when the first epitope is HIV-1.

[0198] As shown in Figures 2A-2C, the first polypeptide chain comprises (in the N-terminal to C-terminal direction): an N-terminus, the Antigen-Binding Domain of a Light Chain Variable Domain (VL1) of an antibody that binds to a "first" epitope of a "first" antigen (either an effector cell epitope such as but not limited to CD3, or HIV-1), the Antigen-Binding Domain of a Heavy Chain Variable Domain (VH2) of an antibody that binds to a "second" epitope of a "second" antigen (HIV-1, if the first antigen as CD3; CD3, if the first antigen was HIV-1), a Heterodimer-Promoting Domain which may comprise a cysteine residue, the CH2- CH3 domains of an Fc Domain ("Fc Domain") and a C-terminus. The second polypeptide contains (in the N-terminal to C-terminal direction): an N-terminus, the Antigen-Binding Domain the Light Chain Variable Domain (VL2) of an antibody that binds to the second epitope of the second antigen (VL2), the Antigen-Binding Domain of the Heavy Chain Variable Domain (VH1) of an antibody that binds to the first epitope of the first antigen, a Heterodimer-Promoting Domain that promotes heterodimerization with the first polypeptide chain, and a C-terminus. Here, two first polypeptides complex with each other to form an Fc Domain. An intervening peptide linker (Peptide Linker 1) separates the Antigen-Binding Domain of the Light Chain Variable Domain from the Antigen-Binding Domain of the Heavy Chain Variable Domain. In non-limiting embodiments the Antigen-Binding Domain of the Heavy Chain Variable Domain is linked to the Heterodimer-Promoting Domain by an intervening peptide linker (Peptide Linker 2). In other non-limiting embodiments, the Heterodimer-Promoting Domain is linked to the Fc Domain by an intervening peptide linker (Peptide Linker 3) or by an intervening spacer-linker peptide (Spacer-Linker 3). In certain embodiments, the first and second polypeptide chains form a disulfide bond between cysteine residues, which may be present in Peptide Linker 2 (e.g., but not limited to Peptide Linker 2- C) and/or in the Heterodimer-Promoting Domains (e.g. but not limited to E-coil-C/K-coil-C). Figures 2A-2C provide schematics of three variations of such diabodies utilizing different Heterodimer-Promoting Domains. In non-limiting embodiments, the first polypeptide chains will contain, in the N-terminal to C-terminal direction: VLl - Peptide Linker 1 - VH2 - Peptide Linker 2 - Heterodimer-Promoting Domain - Spacer-Linker 3 - Fc Domain, and the second polypeptide chains will contain, in the N-terminal to C-terminal direction: VL2 - Peptide Linker 1 - VH1 - Peptide Linker 2 - Heterodimer-Promoting Domain.

[0199] In some embodiments, Fc bearing diabodies may comprise three polypeptide chains. The first polypeptide of such a molecule contains three Domains: (i) a VLl -containing Domain, (ii) a VH2-containing Domain, (iii) a Domain that promotes heterodimerization and covalent bonding with the diabody' s first polypeptide chain and (iv) a Domain containing a CH2-CH3 sequence. The second polypeptide of such diabodies contains: (i) a VL2- containing Domain, (ii) a VH1 -containing Domain and (iii) a Domain that promotes heterodimerization and covalent bonding with the diabody' s first polypeptide chain. The third polypeptide of such diabodies comprises a CH2-CH3 sequence. Thus, the first and second polypeptide chains of such diabodies associate together to form a VL1/VH1 binding site that is capable of binding to the epitope, as well as a VL2/VH2 binding site that is capable of binding to the second epitope. The first and second polypeptides are bonded to one another through a disulfide bond involving cysteine residues in their respective third Domains.

Notably, the first and third polypeptide chains complex with one another to form an Fc Domain that is stabilized via a disulfide bond. Such diabodies have enhanced potency. Such Fc bearing diabodies may have either of two orientations (Table 2): Table 2

First 3^ri Chain NH₂-CH2-CH3-COOH

Orientation 1^st Chain NH₂-VLl-VH2-Heterodimer Promoting Domain-CH2-CH3-COOH

2^nd Chain NH₂-VL2-VHl-Heterodimer Promoting Domain-COOH

Second 3^ri Chain NH₂-CH2-CH3-COOH

Orientation 1^st Chain NH₂-CH2-CH3- VLl-VH2-Heterodimer Promoting Domain-COOH

2^nd Chain NH₂-VL2-VHl-Heterodimer Promoting Domain-COOH

[0200] HIV-1 bispecific monovalent Fc bearing diabodies can be composed of three polypeptide chains which associate with one another to form one binding site specific for an epitope of HIV-1 and one binding site specific for another epitope, for example but not limited to an epitope of CD3 (see, Figure 3A-3B), so as to be capable of simultaneously binding to HIV-1 and to CD3. Thus, such molecules bind to a "first" antigen, which may be either CD3 or HIV-1, and a "second" antigen, which is HIV-1 when the first epitope is CD3, and is CD3 when the first epitope is HIV-1.

[0201] As shown in Figure 3A, the first of such three polypeptide chains will contain, in the N-terminal to C-terminal direction, an N-terminus, the Antigen-Binding Domain of a Light Chain Variable Domain (VL) of an antibody that binds to a "first" epitope of a "first" antigen (for example but not limited to either CD3 or HIV-1), the Antigen-Binding Domain of a Heavy Chain Variable Domain (VH) of an antibody that binds to a "second" epitope of a "second" antigen (for example but not limited to HIV-1, if the first antigen was CD3; CD3, if the first antigen was HIV-1), a Heterodimerizati on-Promoting Domain, and a C-terminus. An intervening peptide linker (Peptide Linker 1) separates the Antigen-Binding Domain of the Light Chain Variable Domain from the Antigen-Binding Domain of the Heavy Chain Variable Domain. In non-limiting embodiments, the Antigen-Binding Domain of the Heavy Chain Variable Domain is linked to the Heterodimerization-Promoting Domain by an intervening peptide linker (Peptide Linker 2). In a non-limiting example of an HIV-1 bispecific monovalent Fc bearing diabody, the C-terminus of the Heterodimerization- Promoting Domain is linked to the CH2-CH3 domains of an Fc Domain ("Fc Domain") by an intervening peptide linker (Peptide Linker 3) or by an intervening spacer-linker peptide (Spacer-Linker 3). In non-limiting embodiments, the first of the three polypeptide chains will thus contain, in the N-terminal to C-terminal direction: VLpirst Antigen - Peptide Linker 1 - Hsecond Antigen - Peptide Linker 2 - Heterodimerization-Promoting Domain - Spacer-Linker 3 - Fc Domain. [0202] Alternatively, as shown in Figure 3B, the first of such three polypeptide chains will contain, in the N-terminal to C-terminal direction, an N-terminus, Peptide Linker 3, the CH2-CH3 domains of an Fc Domain ("Fc Domain"), an intervening spacer peptide (Peptide Linker 4), having, for example the amino acid sequence: APSSS (SEQ ID NO:524) or the amino acid sequence APSSSPME (SEQ ID NO:525), the Antigen-Binding Domain of a Light Chain Variable Domain (VL) of an antibody that binds to the first epitope of the first antigen (for example but not limited to CD3 or HIV-1), the Antigen-Binding Domain of a Heavy Chain Variable Domain (VH) of an antibody that binds to the second epitope of the second antigen (for example but not limited to HIV-1, if the first antigen was CD3; CD3, if the first antigen was HIV-1), a Heterodimerizati on-Promoting Domain, and a C-terminus. An intervening peptide linker (Peptide Linker 1) separates the Antigen-Binding Domain of the Light Chain Variable Domain from the Antigen-Binding Domain of the Heavy Chain

Variable Domain. In non-limiting embodiments, the Antigen-Binding Domain of the Heavy Chain Variable Domain is linked to the Heterodimerization-Promoting Domain by an intervening peptide linker peptide (Peptide Linker 2). In non-limiting embodiments, the first of the three polypeptide chains will thus contain, in the N-terminal to C-terminal direction: Peptide Linker 3 - Fc Domain - Peptide Linker 4 - VLpirst Antigen - Peptide Linker 1 - Hsecond Antigen - Peptide Linker 2 - Heterodimerization-Promoting Domain.

[0203] In non-limiting embodiments, the second of such three polypeptide chains will contain, in the N-terminal to C-terminal direction, an N-terminus, the Antigen-Binding Domain of a Light Chain Variable Domain (VL) of an antibody that binds to the second epitope of the second antigen, the Antigen-Binding Domain of a Heavy Chain Variable Domain (VH) of an antibody that binds to the first epitope of the first antigen, a

Heterodimerization-Promoting Domain and a C-terminus. An intervening peptide linker (Peptide Linker 1) separates the Antigen-Binding Domain of the Light Chain Variable Domain from the Antigen-Binding Domain of the Heavy Chain Variable Domain. In non- limiting embodiments, the Antigen-Binding Domain of the Heavy Chain Variable Domain is linked to the Heterodimerization-Promoting Domain by an intervening peptide linker

(Peptide Linker 2). In non-limiting embodiments, the second of the three polypeptide chains will thus contain, in the N-terminal to C-terminal direction: VLsecond Antigen - Peptide Linker 1 - VH_First Antigen - Peptide Linker 2 - Heterodimerization-Promoting Domain. [0204] In non-limiting embodiments, the third of such three polypeptide chains will contain a peptide linker (Peptide Linker 3) and the CH2-CH3 domains of an Fc Domain ("Fc

Domain").

[0205] For the various Fc bearing multispecific molecules described herein, the Antigen- Binding Domain of the Light Chain Variable Domain of the first polypeptide chain interacts with the Antigen-Binding Domain of the Heavy Chain Variable Domain of the second polypeptide chain in order to form a functional antigen-binding site that is specific for the first antigen (e.g., either HIV-1 or CD3). Likewise, the Antigen-Binding Domain of the Light Chain Variable Domain of the second polypeptide chain interacts with the Antigen-Binding Domain of the Heavy Chain Variable Domain of the first polypeptide chain in order to form a second functional antigen-binding site that is specific for the second antigen (e.g., either CD3 or HIV-1, depending upon the identity of the first antigen). Thus, the selection of the

Antigen-Binding Domain of the Light Chain Variable Domain and the Antigen-Binding Domain of the Heavy Chain Variable Domain of the first and second polypeptide chains are coordinated, such that the two polypeptide chains collectively comprise Antigen-Binding Domains of light and Heavy Chain Variable Domains capable of binding to the first and second antigens (e.g., HIV-1 and CD3).

[0206] The Fc Domain of the Fc bearing multispecific molecules, including but not limited to bispecific and trispecific molecules (e.g. bispecific antibodies, bispecific diabodies, and trivalent binding molecules), of the present invention may be either a complete Fc Domain (e.g., a complete IgG Fc Domain) or only a fragment of a complete Fc Domain. In some embodiments the Fc Domain of the molecules of the present invention may possess the ability to bind to one or more Fc receptors (e.g., FcyR(s)). In non-limiting embodiments the Fc Domain will cause reduced binding to FcyRIA (CD64), FcyRIIA (CD32A), FcyRIIB

(CD32B), FcyRIIIA (CD 16a) or FcyRIIIB (CD 16b) (relative to the binding exhibited by a wild-type Fc Domain) or will substantially eliminate the ability of such Fc Domain to bind to such receptor(s). The Fc bearing multispecific molecules of the present invention may include some or all of the CH2 Domain and/or some or all of the CH3 Domain of a complete Fc Domain, or may comprise a variant CH2 and/or a variant CH3 sequence (that may include, for example, one or more insertions and/or one or more deletions with respect to the CH2 or CH3 domains of a complete Fc Domain). The Fc Domain of the Fc bearing multispecific molecules of the present invention may comprise non-Fc polypeptide portions, or may comprise portions of non-naturally complete Fc Domains, or may comprise non-naturally occurring orientations of CH2 and/or CH3 domains (such as, for example, two CH2 domains or two CH3 domains, or in the N-terminal to C-terminal direction, a CH3 Domain linked to a CH2 Domain, etc.).

[0207] Polymorphisms have been observed at a number of different positions within antibody constant regions (e.g., Fc positions, including but not limited to positions 270, 272, 312, 315, 356, and 358 as numbered by the EU index as set forth in Kabat), and thus slight differences between the presented sequence and sequences in the prior art can exist. Polymorphic forms of human immunoglobulins have been well-characterized. At present, 18 Gm allotypes are known: Glm (1, 2, 3, 17) or Glm (a, x, f, z), G2m (23) or G2m (n), G3m (5, 6, 10, 11, 13, 14, 15, 16, 21, 24, 26, 27, 28) or G3m (bl, c3, b3, bO, b3, b4, s, t, gl, c5, u, v, g5) (Lefranc, et. al, "The Human IgG Subclasses: Molecular Analysis Of Structure, Function And

Regulation." Pergamon, Oxford, pp. 43-78 (1990); Lefranc, G. et. al., 1979, Hum. Genet. : 50, 199-211). It is specifically contemplated that the molecules of the present invention may be incorporate any allotype, isoallotype, or haplotype of any immunoglobulin gene, and are not limited to the allotype, isoallotype or haplotype of the sequences provided herein.

Furthermore, in some expression systems the C-terminal amino acid residue (bolded above) of the CH3 Domain may be post-translationally removed. Accordingly, the C-terminal residue of the CH3 Domain is an optional amino acid residue. Specifically encompassed by the instant invention are molecules lacking the C-terminal residue of the CH3 Domain. Also specifically encompassed by the instant invention are such constructs comprising the C- terminal lysine residue of the CH3 Domain.

[0208] In non-limiting embodiments the first and third polypeptide chains of the Fc bearing multispecific molecules of the present invention each comprise CH2-CH3 domains that complex together to form an immunoglobulin (IgG) Fc Domain. The amino acid sequence of the CH2-CH3 domain of human IgGl is (SEQ ID NO: 527):

APELLGGPSV FLFPPKPKDT LMI SRTPEVT CVWDVSHED PEVKFNWYVD GVEVHNAKTK

PREEQYNSTY RWSVLTVLH QDWLNGKEYK CKVSNKALPA PI EKTI SKAK GQPREPQVYT

LPPSREEMTK NQVSLTCLVK GFYPSDIAVE WESNGQPENN YKTTPPVLDS DGS FFLYSKL

TVDKSRWQQG NVFSCSVMHE ALHNHYTQKS LSLS PGK

[0209] Thus the CH2 and/or CH3 Domains of the first and third polypeptide chains may both be composed of SEQ ID NO: 527, or a variant thereof (e.g., SEQ ID NO: 528, 529, 530).

[0210] In non-limiting embodiments the CH2-CH3 domains of the first and third polypeptide chains of the Fc bearing multispecific molecules of the present invention exhibit decreased (or substantially no) binding to FcyRIA (CD64), FcyRIIA (CD32A), FcyRIIB (CD32B), FcyRIIIA (CD 16a) or FcyRIIIB (CD 16b) (relative to the binding exhibited by the wild-type Fc Domain). Fc variants and mutant forms capable of mediating such altered binding are well known in the art and include amino acid substitutions at positions 234 and 235, a substitution at position 265 or a substitution at position 297 (see, for example, US Patent No. 5,624,821, herein incorporated by reference). In non-limiting embodiments the CH2-CH3 Domain of the first and/or third polypeptide chains of the Fc bearing multispecific molecules of the present invention include a substitution at position 234 with alanine and 235 with alanine.

[0211] The CH2 and/or CH3 Domains of the first and third polypeptide chains need not be identical in sequence, and in some embodiment are modified to foster complexing between the two polypeptide chains. For example, an amino acid substitution (for example a substitution with an amino acid comprising a bulky side group forming a 'knob', e.g., tryptophan) can be introduced into the CH2 or CH3 Domain such that steric interference will prevent interaction with a similarly mutated domain and will obligate the mutated domain to pair with a domain into which a complementary, or accommodating mutation has been engineered, i.e., 'the hole' (e.g., a substitution with glycine). Such sets of mutations can be engineered into any pair of polypeptides comprising Fc bearing multispecific molecules of the invention, and further, engineered into any portion of the polypeptides chains of the pair. Methods of protein engineering to favor heterodimerization over homodimerization are well known in the art, in particular with respect to the engineering of immunoglobulin-like molecules, and are encompassed herein (see e.g., Ridgway et al. (1996) " 'Knobs-Into-Holes ' Engineering Of Antibody CH3 Domains For Heavy Chain Heterodimerization, " Protein Engr. 9:617-621, Atwell et al. (1997) "Stable Heterodimers From Remodeling The Domain

Interface Of A Homodimer Using A Phage Display Library, " J. Mol. Biol. 270: 26-35, and Xie et al. (2005) "A New Format Of Bispecific Antibody: Highly Efficient

Heterodimerization, Expression And Tumor Cell Lysis, " J. Immunol. Methods 296:95-101; each of which is hereby incorporated herein by reference in its entirety). In non-limiting embodiments the 'knob' is engineered into the CH2-CH3 Domains of the first polypeptide chain and the 'hole' is engineered into the CH2-CH3 Domains of the third polypeptide chain. Thus, the 'knob' will help in preventing the first polypeptide chain from homodimerizing via its CH2 and/or CH3 Domains. In non-limiting embodiments, as the third polypeptide chain contains the 'hole' substitution it will heterodimerize with the first polypeptide chain as well as homodimerize with itself. In non-limiting embodiments a knob is created by modifying a native IgG Fc Domain to contain the modification T366W. In non-limiting embodiments a hole is created by modifying a native IgG Fc Domain to contain the modification T366S, L368A and Y407V. To aid in purifying the third polypeptide chain homodimer from the final bispecific monovalent Fc bearing diabody comprising the first, second and third polypeptide chains, the protein A binding site of the CH2 and CH3 Domains of the third polypeptide chain is mutated by amino acid substitution at position 435 (H435R). Thus, the third polypeptide chain homodimer will not bind to protein A, whereas the bispecific monovalent Fc bearing diabody will retain its ability to bind protein A via the protein A binding site on the first polypeptide chain.

[0212] In non-limiting embodiments a sequence for the CH2 and CH3 Domains of the first polypeptide chain of the Fc bearing multispecific molecules of the present invention will have the "knob-bearing" sequence (SEQ ID NO: 531):

APEAAGGPSV FLFPPKPKDT LMI SRTPEVT CVWDVSHED PEVKFNWYVD GVEVHNAKTK

PREEQYNSTY RWSVLTVLH QDWLNGKEYK CKVSNKALPA PI EKTI SKAK GQPREPQVYT

LPPSREEMTK NQVSLWCLVK GFYPSDIAVE WESNGQPENN YKTTPPVLDS DGS FFLYSKL

TVDKSRWQQG NVFSCSVMHE ALHNHYTQKS LSLS PGK

[0213] In non-limiting embodiments a sequence for the CH2 and CH3 Domains of the third polypeptide chain of the Fc bearing multispecific molecules of the present invention will have the "hole-bearing" sequence (SEQ ID NO: 533):

APEAAGGPSV FLFPPKPKDT LMI SRTPEVT CVWDVSHED PEVKFNWYVD GVEVHNAKTK

PREEQYNSTY RWSVLTVLH QDWLNGKEYK CKVSNKALPA PI EKTI SKAK GQPREPQVYT

LPPSREEMTK NQVSLSCAVK GFYPSDIAVE WESNGQPENN YKTTPPVLDS DGS FFLVSKL

TVDKSRWQQG NVFSCSVMHE ALHNRYTQKS LSLS PGK

[0214] As will be noted, the CH2-CH3 Domains of SEQ ID NO: 531 and SEQ ID NO: 533 include a substitution at position 234 with alanine and 235 with alanine, and thus form an Fc Domain exhibit decreased (or substantially no) binding to FcyRIA (CD64), FcyRIIA

(CD32A), FcyRIIB (CD32B), FcyRIIIA (CD 16a) or FcyRIIIB (CD 16b) (relative to the binding exhibited by the wild-type Fc Domain (SEQ ID NO: 527).

[0215] In non-limiting embodiments, the first polypeptide chain will have a "knob-bearing" CH2-CH3 sequence, such as that of SEQ ID NO: 531 or 532. However, as will be recognized, a "hole-bearing" CH2-CH3 Domain (e.g., SEQ ID NO: 533 or 534) could be employed in the first polypeptide chain, in which case, a "knob-bearing" CH2-CH3 Domain (e.g., SEQ ID NO: 531 or 532) would be employed in the third polypeptide chain.

[0216] In non-limiting embodiments, the Fc domain can be modified by amino acid substitution to increase binding to the neonatal Fc receptor and therefore the half-life of the antibody when administered to a subject. The Fc domain can be an IgA, IgM, IgD, IgE or IgG Fc domain. The Fc domain can be an optimized Fc domain, as described in U.S.

Published Patent Application No. 20100093979, incorporated herein by reference. In certain embodiments the Fc bearing multispecific molecules comprise amino acid alterations, or combinations thereof, for example in the Fc domain(s) outside of epitope binding, which alterations can improve their properties. Various Fc modifications are known in the art. Amino acid numbering is according to the EU Index in Kabat. In some embodiments, the invention contemplates antibodies comprising mutations that affect neonatal Fc receptor (FcRn) binding, antibody half-life, and localization and persistence of antibodies at mucosal sites. See e.g. Ko SY et al., Nature 514: 642-45, 2014, at Figure la and citations therein; Kuo, T. and and Averson, V., mAbs 3(5): 422-430, 2011, at Table 1, US Pub 20110081347 (an aspartic acid at Kabat residue 288 and/or a lysine at Kabat residue 435), US Pub

20150152183 for various Fc Domain mutation, incorporated by reference in their entirety.

[0217] In certain embodiments, the Fc bearing multispecific molecules comprise AAAA substitution in and around the Fc Domain of the Fc bearing multispecific molecule that has been reported to enhance ADCC via NK cells (AAA mutations) containing the Fc Domain aa of S298A as well as E333A and K334A (Shields RI et al. JBC, 276: 6591-6604, 2001) and the 4^th A (N434A) is to enhance FcR neonatal mediated transport of the IgG to mucosal sites (Shields RI et al. ibid).

[0218] Other mutations have been reported to improve antibody half-life or function or both and can be incorporated into the Fc Domain of the Fc bearing multispecific molecules. These include the DLE set of mutations (Romain G, et al. Blood 124: 3241, 2014), the LS mutations M428L/N434S, alone or in a combination with other Fc Domain mutations, (Ko SY et al. Nature 514: 642-45, 2014, at Figure la and citations therein; Zlevsky et al., Nature

Biotechnology, 28(2): 157-159, 2010; US Pub 20150152183); the YTE Fc mutations (Robbie G et al Antimicrobial Agents and Chemotherapy 12: 6147-53, 2013) as well as other engineered mutations to the Fc Domain such as QL mutations, IHH mutations (Ko SY et al. Nature 514: 642-45, 2014, at Figure la and relevant citations; See also Rudicell R et al. J. Virol 88: 12669-82, 201). In some embodiments, modifications, such as but not limited to fucosylation, which may affect interaction with Fc receptors (See e.g. Moldt, et al. JVI 86(11): 66189-6196, 2012). In some embodiments, the Fc bearing multispecific molecules can comprise modifications, for example but not limited to glycosylation, which reduce or eliminate polyreactivity of such a molecule. See e.g. Chuang, et al. Protein Science 24: 1019- 1030, 2015. In some embodiments the Fc bearing multispecific molecules can comprise modifications in the Fc domain such that the Fc domain exhibits, as compared to an unmodified Fc domain enhanced antibody dependent cell mediated cytotoxicity (ADCC); increased binding to FcyRIIA or to FcyRIIIA; decreased binding to FcyRIIB; or increased binding to FcyRIIB. See e.g. US Pub 20140328836.

[0219] In another aspect, the invention provides trivalent structures incorporating two diabody-type binding domains and one non-diabody-type domain and an Fc Domain (see, e.g., Figures 4A-4F and PCT Publication Nos. WO 2015/184207 and WO 2015/184203). Such trivalent binding molecules may be utilized to generate monospecific, bispecific or trispecific molecules. The ability to bind three different epitopes provides enhanced capabilities.

[0220] A further embodiment of the present invention relates to trivalent binding molecules comprising an Fc Domain. The Fc Domain bearing trivalent binding molecules can simultaneously bind a first epitope, a second epitope, and a third epitope, wherein at least one of such epitopes is not identical to another. Such trivalent binding molecules comprise three epitope-binding sites, two of which are Diabody-Type Binding Domains, which provide binding Site A and binding Site B, and one of which is a Fab-Type Binding Domain, or an scFv-Type Binding Domain, which provides binding Site C (see, e.g., Figures 4A-4F, and PCT Publication Nos. WO 2015/184207 and WO 2015/184203). Such trivalent binding molecules thus comprise "VL1" / "VH1" domains that are capable of binding to the first epitope and "VL2" / "VH2" domains that are capable of binding to the second epitope and "VL3" and "VH3" domains that are capable of binding to the "third" epitope of such trivalent binding molecule. A "Diabody-Type Binding Domain" is the type of epitope-binding site present in a diabody, and especially, a DART® diabody, as described above. Each of a "Fab- Type Binding Domain" and an "scFv-Type Binding Domain" are epitope-binding sites that are formed by the interaction of the VL Domain of an immunoglobulin light chain and a complementing VH Domain of an immunoglobulin heavy chain. Fab-Type Binding Domains differ from Diabody-Type Binding Domains in that the two polypeptide chains that form a Fab-Type Binding Domain comprise only a single epitope-binding site, whereas the two polypeptide chains that form a Diabody-Type Binding Domain comprise at least two epitope- binding sites. Similarly, scFv-Type Binding Domains also differ from Diabody-Type Binding Domains in that they comprise only a single epitope-binding site. Thus, as used herein Fab- Type, and scFv-Type Binding Domains are distinct from Diabody-Type Binding Domains. [0221] Typically, the trivalent binding molecules of the present invention will comprise four different polypeptide chains (see Figures 4A-4B), however, the molecules may comprise fewer or greater numbers of polypeptide chains, for example by fusing such polypeptide chains to one another (e.g., via a peptide bond) or by dividing such polypeptide chains to form additional polypeptide chains, or by associating fewer or additional polypeptide chains via disulfide bonds. Figures 4C-4F illustrate this aspect of the present invention by

schematically depicting such molecules having three polypeptide chains. As provided in

Figures 4A-4F, the trivalent binding molecules of the present invention may have alternative orientations in which the Diabody-Type Binding Domains are N-terminal (Figures 4A, 4C and 4D) or C-terminal (Figures 4B, 4E and 4F) to an Fc Domain.

[0222] In certain embodiments, the first polypeptide chain of such trivalent binding molecules of the present invention contains: (i) a VL1 -containing Domain, (ii) a VH2-containing Domain, (iii) a Heterodimer-Promoting Domain, and (iv) a Domain containing a CH2-CH3 sequence. The VL1 and VL2 Domains are located N-terminal or C-terminal to the CH2-CH3- containing domain as presented in Table 3 (also see, Figures 4A and 4B). The second polypeptide chain of such embodiments contains: (i) a VL2-containing Domain, (ii) a VH1- containing Domain, and (iii) a Heterodimer-Promoting Domain. The third polypeptide chain of such embodiments contains: (i) a VH3 -containing Domain, (ii) a CHI -containing Domain and (iii) a Domain containing a CH2-CH3 sequence. The third polypeptide chain may be the heavy chain of an antibody that contains a VH3 and a heavy chain constant region, or a polypeptide that contains such domains. The fourth polypeptide of such embodiments contains: (i) a VL3 -containing Domain and (ii) a CL-containing Domain. The fourth polypeptide chains may be a light chain of an antibody that contains a VL3 complementary to the VH3 of the third polypeptide chain, or a polypeptide that contains such domains. The third or fourth polypeptide chains may be isolated from naturally occurring antibodies.

Alternatively, they may be constructed recombinantly, synthetically or by other means.

[0223] The Light Chain Variable Domain of the first and second polypeptide chains are separated from the Heavy Chain Variable Domains of such polypeptide chains by an intervening spacer peptide having a length that is too short to permit their VL1/VH2 (or their VL2/VH1) domains to associate together to form epitope-binding site capable of binding to either the first or second epitope. A preferred intervening peptide linker (Peptide Linker 1) for this purpose has the sequence (SEQ ID NO:508): GGGSGGGG. Other Domains of the trivalent binding molecules may be separated by one or more intervening peptide linkers (Peptide Linkers), optionally comprising a cysteine residue. In particular, as provided above, such Peptide Linkers will typically be incorporated between Variable Domains (i.e., VH or VL) and peptide Heterodimer-Promoting Domains (e.g., an E-coil or K-coil) and between such peptide Heterodimer-Promoting Domains (e.g., an E-coil or K-coil) and CH2-CH3 Domains. Exemplary peptide linkers (e.g., Peptide Linker 2, Peptide Linker 2-C, Peptide Linker 3, Spacer Linker 3, etc.) useful for the generation of trivalent binding molecules are provided above. Such linkers are also provided in PCT Publication Nos. WO 2015/184207 and WO 2015/184203. Thus, the first and second polypeptide chains of such trivalent binding molecules associate together to form a VL1/VH1 binding site capable of binding a first epitope, as well as a VL2/VH2 binding site that is capable of binding to a second epitope. The third and fourth polypeptide chains of such trivalent binding molecules associate together to form a VL3/VH3 binding site that is capable of binding to a third epitope.

[0224] As described above, the trivalent binding molecules of the present invention may comprise three polypeptides. Trivalent binding molecules comprising three polypeptide chains may be obtained by linking the domains of the fourth polypeptide N-terminal to the VH3 -containing Domain of the third polypeptide (e.g., using an intervening spacer peptide (Peptide Linker 5)). Alternatively, a third polypeptide chain of a trivalent binding molecule of the invention containing the following domains is utilized: (i) a VL3 -containing Domain, (ii) a VH3 -containing Domain, and (iii) a Domain containing a CH2-CH3 sequence, wherein the VL3 and VH3 are spaced apart from one another by an intervening spacer peptide that is sufficiently long (at least 9 or more amino acid residues) so as to allow the association of these domains to form an epitope-binding site. One preferred intervening spacer peptide for this purpose has the sequence: GGGGS GGGGS GGGGS (SEQ ID NO: 526).

[0225] It will be understood that the VL1/VH1, VL2/VH2, and VL3/VH3 Domains of such trivalent binding molecules may be different so as to permit binding that is monospecific, bispecific, or trispecific. In particular, the VL and VH Domains may be selected such that a trivalent binding molecule comprises two binding sites for a first epitope and one binding sites for a second epitope, or one binding site for a first epitope and two binding sites for a second epitope, or one binding site for a first epitope, one binding site for a second epitope and one binding site for a third epitope. [0226] In one embodiment, these domains are selected so as to bind an epitope of HIV- 1 Env, an epitope of second molecule, and an epitope of a third molecule, wherein the second molecule and the third molecule (e.g., CD2, CD3, CD8, CD16, T-Cell Receptor (TCR),

KG2D, etc.) are present on the surface of an effector cell, such as a T lymphocyte, a natural killer (NK) cell or other mononuclear cell.

[0227] The general structure of the polypeptide chains of representative trivalent binding molecules of invention is provided in Figures 4A-4F and in Table 3:

HPD = Heterodimer-Promoting Domain

[0228] One embodiment of the present invention relates to trivalent binding molecules that comprise two epitope-binding sites for HIV-1 Env and one epitope-binding site for a second molecule. The two epitope-binding sites for HIV-1 Env may bind the same epitope or different epitopes. Another embodiment of the present invention relates to trivalent binding molecules that comprise, one epitope-binding site for HIV-1 Env and two epitope-binding sites for a second molecule. The two epitope-binding sites for the second molecule may bind the same epitope or different epitopes of the second molecule. A further embodiment of the present invention relates to trispecific trivalent binding molecules that comprise, one epitope- binding site for HIV-1 Env, one epitope-binding site for a second molecule, and one epitope- binding site for a third molecule. In certain embodiments, the second molecule is a molecule (e.g., CD2, CD3, CD8, CD16, T-Cell Receptor (TCR), KG2D, etc.) present on the surface of an effector cell, such as a T lymphocyte, a natural killer (NK) cell or other mononuclear cell. In certain embodiments, the second molecule is CD3 and the third molecule is CD8. As provided above, such trivalent binding molecules may comprise three, four, five, or more polypeptide chains.

[0229] In one embodiment, these domains are selected so as to bind two epitopes of HIV-1 Env, which may be the same epitopes or different epitopes, and an epitope of second molecule, wherein the second molecule is a molecule (e.g., CD2, CD3, CD8, CD16, T-Cell Receptor (TCR), KG2D, etc.) present on the surface of an effector cell, such as a T lymphocyte, a natural killer (NK) cell or other mononuclear cell. In a specific embodiment, the two epitopes of HIV-1 Env are the same, and the second molecule is CD3. In an alternative embodiment, the two epitopes of HIV-1 Env are different, and the second molecule is CD3.

[0230] The molecules, and fragments thereof, described above can be formulated as a composition (e.g., a pharmaceutical composition). Suitable compositions can comprise the molecules (or fragments thereof) in a pharmaceutically acceptable carrier e.g., dissolved or dispersed in an aqueous medium, or lyophilized. The compositions can be sterile and can be in an injectable form (e.g. but not limited to a form suitable for intravenous injection, or intramuscular injection). The molecules (and fragments thereof) can also be formulated as a composition appropriate for topical administration to the skin or mucosa. Such compositions can take the form of liquids, ointments, creams, gels and pastes. The molecules (and fragments thereof) can also be formulated as a composition appropriate for intranasal administration. The molecules (and fragments thereof) can be formulated so as to be administered as a post-coital douche or with a condom. Standard formulation techniques can be used in preparing suitable compositions.

[0231] In certain embodiments the invention provides multispecific molecules such as but not limited to bispecific and trispecific molecules (e.g., bispecific antibodies, bispecific diabodies, trivalent binding molecules, etc.) comprising the binding domains from human, humanized and/or chimeric antibodies. Methods to construct such antibodies are well known in the art. [0232] In certain aspects the invention provides use of the multispecific molecules of the invention such as but not limited to bispecific and trispecific molecules (e.g., bispecific antibodies, bispecific diabodies, trivalent binding molecules, etc.), in methods of treating and preventing HIV-1 infection in an individual, comprising administering to the individual a therapeutically effective amount of a composition comprising a multispecific molecule of the invention in a pharmaceutically acceptable form. In certain embodiment, the methods include a composition which includes more than one HIV-1 targeting multispecific molecule. In certain embodiments, the HIV-1 targeting multispecific molecule in such combination bind different epitopes on the HIV-1 envelope. In certain embodiments, such combinations of multispecific molecule targeting more than one HIV-1 epitope provide increased killing of HIV-1 infected cells. In other embodiments, such combinations of multispecific molecule targeting more than one HIV-1 epitope provide increased breadth in recognition of different HIV-1 subtypes.

[0233] The various multispecific molecule described herein have utility, for example, in settings including but not limited to the following:

i) in the setting of anticipated known exposure to HIV-1 infection, the multispecific molecule described herein can be administered prophylactically (e.g., IV, topically or intranasally) as a microbiocide,

ii) in the setting of known or suspected exposure, such as occurs in the setting of rape victims, or commercial sex workers, or in any homosexual or heterosexual transmission without condom protection, the multispecific molecule described herein can be administered as post-exposure prophylaxis, e.g., IV or topically, and

iii) in the setting of Acute HIV-1 infection (AHI), the multispecific molecule described herein can be administered as a treatment for AHI to control the initial viral load, or for the elimination of virus-infected CD4 T cells.

[0234] In accordance with the invention, the multispecific molecules described herein can be administered prior to contact of the subject or the subject's immune system/cells with HIV-1 or within about 48 hours of such contact. Administration within this time frame can maximize inhibition of infection of vulnerable cells of the subject with HIV-1.

[0235] In addition, various forms of the multispecific molecules described herein can be administered to chronically or acutely infected HIV-1 patients and used to kill remaining virus infected cells by virtue of these multispecific molecule binding to the surface of virus infected cells and being able to mediate redirected cell killing of such infected cells. [0236] In certain embodiments, the multispecific molecules of the invention can be administered in combination with latency activating agents, so as to activate latent reservoir of HIV- 1 -infected cells. The expectation is that by activating latent proviral HIV-1 DNA in resting cells, once inactive cells will start producing new virus and they will be recognized and eliminated by the immune system. Non-limiting examples of latency activating agents are HDAC inhibitors, e,g, vorinostat, romidepsin, panobinostat, disulfiram, JQ1, bryostatin, PMA, inonomycin, or any combination thereof. See Bullen et al. Nature Medicine 20, 425- 429 (2014).

[0237] In certain embodiments the multispecific molecules of the invention can be administered in combination with anti-retroviral agents.

[0238] Suitable dose ranges can depend on the multispecific molecule and on the nature of the formulation and route of administration. Optimum doses can be determined by one skilled in the art without undue experimentation. For example, doses of antibodies in the range of 1-50 mg/kg of unlabeled or labeled antibody (with toxins or radioactive moieties) can be used. If antibody fragments, with or without toxins are used or antibodies are used that can be targeted to specific CD4 infected T cells, then less antibody can be used (e.g., from 5 mg/kg to 0.01 mg/kg). If multispecific molecules are used, doses in the range of 0.01 μg/kg to about 30 mg/kg or more of the subject's body weight can be used. Suitable dose ranges can depend on the antibody (or fragment, or multispecific molecule) and on the nature of the formulation and route of administration. Optimum doses can be determined by one skilled in the art without undue experimentation. For example but not limited, doses of multispecific molecules in the range of 0.01-100 μg/kg, 0.1-50 mg/kg, 1-50 mg/kg, 1-10 mg/kg, 1, 1.25, 1.5, 2, 2.5, 3, 3.5, 4, 4.5, 5, 5.5, 6, 6.5, 7, 7.5, 8, 8.5, 9, 9.5, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, or 30 mg/kg of unlabeled or labeled multispecific molecule (with toxins or radioactive moieties) can be used. If antibody fragments, with or without toxins are used or antibodies are used that can be targeted to specific CD4 infected T cells, then less antibody can be used (e.g., from 5 mg/kg to 0.01 mg/kg). In other

embodiments, the molecules of the invention can be administered at a suitable fixed dose, regardless of body size or weight. See Bai et al. Clinical Pharmacokinetics February 2012, Volume 51, Issue 2, pp 119-135.

[0239] Multispecific molecules of the invention can be produced recombinantly using nucleic acids comprising nucleotide sequences encoding VH and VL sequences selected from those shown in the figures and examples, or those known in the art. [0240] In certain embodiments the invention provides multispecific binding molecules comprising antigen binding fragments. Typically, multispecific binding molecules compete with the intact antibody from which they were derived for specific binding to the target including separate heavy chains, light chains Fab, Fab', F(ab').sub.2, F(ab)c, diabodies, Dabs, nanobodies, and Fv. Fragments that can be produced by recombinant DNA techniques, or by enzymatic or chemical separation of intact immunoglobulins.

[0241] Nucleic acid sequences encoding polypeptides for the production of multispecific molecules with specificities as described herein can be used to produce plasmids for stable expression of such multispecific molecules. Methods for recombinant expression and purification are known in the art. In certain embodiments of Fc bearing multispecific molecules, the plasmids also comprise any of the changes to the Fc portion described herein.

[0242] In certain embodiments, the nucleic acids are optimized for recombinant expression in a suitable host cell. In certain embodiments, the vector is suitable for gene delivery and expression. There are numerous expression systems available for expression of proteins including E. coli, other bacterial hosts, yeast, and various higher eukaryotic cells such as the COS, CHO, HeLa and myeloma cell lines.

[0243] Any suitable cell line can be used for expression of the polypeptides of the invention, including but not limited to CHO cells, 293T cells. In some aspects, the invention provides nucleic acids encoding these antibodies, expression cassettes and vectors including these nucleic acids, and isolated cells that express the nucleic acids which encode the multispecific molecules of the invention are also provided. The polypeptides of the invention can be purified by any suitable method for purification of polypeptides and/or antibodies.

[0244] Table 4. Summary listing of various sequences listed throughout the specification; starting at SEQ ID NO: 500

501 hCD3 VH gaggtgcagctggtggagtctgggggaggcttggtccagcctggaggg (hXR32VH) tccctgagactctcctgtgcagcctctggattcaccttcagcacatac (nucleoti gctatgaattgggtccgccaggctccagggaaggggctggagtgggtt de ggaaggatcaggtccaagtacaacaattatgcaacctactatgccgac sequence ) tctgtgaagggtagattcaccatctcaagagatgattcaaagaactca ctgtatctgcaaatgaacagcctgaaaaccgaggacacggccgtgtat tactgtgtgagacacggtaacttcggcaattcttacgtgtcttggttt gcttattggggacaggggacactggtgactgtgtcttcc

502 hCD3 VL QAVVTQEPSLTVSPGGTVTLTCRSSTGAVTTSNYA WVQQKPGQAPRG

LIGGTNKRAPWTPARFSGSLLGGKAALTITGAQAEDEADYYCALWYSN

(hXR32VL) LWVFGGGTKLTVL

(amino

acid

sequence )

503 hCD3 VL caggctgtggtgactcaggagccttcactgaccgtgtccccaggcgga actgtgaccctgacatgcagatccagcacaggcgcagtgaccacatct

(hXR32VL) aactacgccaattgggtgcagcagaagccaggacaggcaccaaggggc ctgatcgggggtacaaacaaaagggctccctggacccctgcacggttt

(nucleoti

tctggaagtctgctgggcggaaaggccgctctgactattaccggggca de

caggccgaggacgaagccgattactattgtgctctgtggtatagcaat sequence )

ctgtgggtgttcgggggtggcacaaaactgactgtgctg

504 hCD16 VH QVTLRESGPALVKPTQTLTLTCTFSGFSLSTSGMGVGWIRQPPGKALE

WLAHIWWDDDKRYNPALKSRLTISKDTSK QVVLTMTNMDPVDTATYY

(h3G8VH5 ) CAQINPAWFAYWGQGTLVTVSS

(amino

acid

sequence )

505 hCD16 VH caggttaccctgagagagtctggccctgcgctggtgaagcccacacag accctcacactgacttgtaccttctctgggttttcactgagcacttct

(h3G8VH5 ) ggtatgggtgtaggctggattcgtcagcctcccgggaaggctctagag tggctggcacacatttggtgggatgatgacaagcgctataatccagcc

(nucleoti

ctgaagagccgactgacaatctccaaggatacctccaaaaaccaggta de

gtcctcacaatgaccaacatggaccctgtggatactgccacatactac sequence )

tgtgctcaaataaaccccgcctggtttgcttactggggccaagggact ctggtcactgtgagctccgg

506 hCD16 VL DIVMTQSPDSLAVSLGERATINCKASQSVDFDGDSFMNWYQQKPGQPP

KLLIYT SNLESGVPDRFSGSGSGTDFTL ISSLQAEDVAVYYCQQSN

(h3G8VLl) EDPYTFGQGTKLEIK

(amino

acid

sequence ) 507 hCD16 VL atgggtgacaatgacatccactttgcctttctctccacaggtgtccac (h3G8VLl) tccgacatcgtgatgacccaatctccagactctttggctgtgtctcta (nucleoti ggggagagggccaccatcaactgcaaggccagccaaagtgttgatttt de gatggtgatagttttatgaactggtaccaacagaaaccaggacagcca sequence ) cccaaactcctcatctatactacatccaatctagaatctggggtccca gacaggtttagtggcagtgggtctgggacagacttcaccctcaccatc agcagcctgcaggctgaggatgtggcagtttattactgtcagcaaagt aatgaagatccgtacacgttcggacaggggaccaagcttgagatcaaa

508 Peptide GGGSGGGG

Linker 1

509 Peptide GGCGGG

Linker 2- C

510 Peptide AS KG

Linker 2

511 HPD 1 GVEPKSC

512 HPD 2 VEPKSC

513 HPD 3 GFNRGEC

514 HPD 4 FNRGEC

515 IgG CHI- ASTKGPSVFPLAPSSKSTSGGTAALGCLVKDYFPEPVTVSWNSGALTS hinge GVHTFPAVLQSSGLYSLSSVVTVPSSSLGTQTYICNVNHKPSNTKVDK Domain RVEPKSCDKTHTCPPCP

516 Ig CL RTVAAPSVFI FPPSDEQLKSG ASVVCLLNNFYPREAKVQWKVDNALQ

Kappa SGNSQESVTEQDSKDSTYSLSSTLTLSKADYEKHKVYACEVTHQGLSS Domain PVTKSFNRGEC

517 Ig CL QPKAAPSVTLFPPSSEELQANKATLVCLISDFYPGAVTVAWKADSSPV

Lambda KAGVETTPSKQSNNKYAASSYLSLTPEQWKSHRSYSCQVTHEGSTVEK

Domain TVAPTECS

518 HPD 5: E- EVAALEKEVAALEKEVAALEKEVAALEK

coil

519 HPD 6: E- EVAACEKEVAALEKEVAALEKEVAALEK

coil-C

520 HPD 7: K- KVAALKEKVAALKEKVAALKEKVAALKE

coil 521 HPD 8: K- KVAACKEKVAALKEKVAALKEKVAALKE

coil-C

522 Spacer GGGDKTHTCPPCP

Linker 3

523 Peptide DKTHTCPPCP

Linker 3

524 Peptide APSSS

Linker 4

525 Peptide APSSSPME

Linker

4.1

526 Peptide GGGGSGGGGSGGGGS

Linker 5

527 IgGl Fc APELLGGPSVFLFPPKPKDTLMISRTPEVTCVVVDVSHEDPEVKFNWY

WT VDGVEVHNAKTKPREEQYNSTYRVVSVLTVLHQDWLNGKEYKCKVSNK ALPAPIEKTISKAKGQPREPQVYTLPPSREEMTKNQVSLTCLVKGFYP SDIAVEWESNGQPENNYKTTPPVLDSDGSFFLYSKLTVDKSRWQQGNV FSCSVMHEALHNHYTQKSLSLSPGX

wherein, X is a lysine (K) or is absent

528 IgGl Fc APEAAGGPSVFLFPPKPKDTLMISRTPEVTCVVVDVSHEDPEVKFNWY

(AA) VDGVEVHNAKTKPREEQYNSTYRVVSVLTVLHQDWLNGKEYKCKVSNK

ALPAPIEKTISKAKGQPREPQVYTLPPSREEMTKNQVSLTCLVKGFYP SDIAVEWESNGQPENNYKTTPPVLDSDGSFFLYSKLTVDKSRWQQGNV FSCSVMHEALHNHYTQKSLSLSPGX

wherein, X is a lysine (K) or is absent

529 IgGl Fc APELLGGPSVFLFPPKPKDTLYITREPEVTCVVVDVSHEDPEVKFNWY

YTE VDGVEVHNAKTKPREEQYNSTYRVVSVLTVLHQDWLNGKEYKCKVSNK ALPAPIEKTISKAKGQPREPQVYTLPPSREEMTKNQVSLTCLVKGFYP SDIAVEWESNGQPENNYKTTPPVLDSDGSFFLYSKLTVDKSRWQQGNV FSCSVMHEALHNHYTQKSLSLSPGX

wherein, X is a lysine (K) or is absent

530 IgGl Fc APEAAGGPSVFLFPPKPKDTLYITREPEVTCVVVDVSHEDPEVKFNWY

(AA/YTE) VDGVEVHNAKTKPREEQYNSTYRVVSVLTVLHQDWLNGKEYKCKVSNK

wherein, X is a lysine (K) or is absent 531 IgGl Fc APEAAGGPSVFLFPPKPKDTLMISRTPEVTCVVVDVSHEDPEVKFNWY Knob (AA) VDGVEVHNAKTKPREEQYNSTYRVVSVLTVLHQDWLNGKEYKCKVSNK

ALPAPIEKTISKAKGQPREPQVYTLPPSREEMTKNQVSLWCLVKGFYP SDIAVEWESNGQPENNYKTTPPVLDSDGSFFLYSKLTVDKSRWQQGNV FSCSVMHEALHNHYTQKSLSLSPGX

wherein, X is a lysine (K) or is absent

532 IgGl Fc APEAAGGPSVFLFPPKPKDTLYITREPEVTCVVVDVSHEDPEVKFNWY

Knob VDGVEVHNAKTKPREEQYNSTYRVVSVLTVLHQDWLNGKEYKCKVSNK (AA/YTE) ALPAPIEKTISKAKGQPREPQVYTLPPSREEMTKNQVSLWCLVKGFYP

SDIAVEWESNGQPENNYKTTPPVLDSDGSFFLYSKLTVDKSRWQQGNV FSCSVMHEALHNHYTQKSLSLSPGX

wherein, X is a lysine (K) or is absent

533 IgGl Fc APEAAGGPSVFLFPPKPKDTLMISRTPEVTCVVVDVSHEDPEVKFNWY

Hole VDGVEVHNAKTKPREEQYNSTYRVVSVLTVLHQDWLNGKEYKCKVSNK (AA/R) ALPAPIEKTISKAKGQPREPQVYTLPPSREEMTKNQVSLSCAVKGFYP

SDIAVEWESNGQPENNYKTTPPVLDSDGSFFLVSKLTVDKSRWQQGNV FSCSVMHEALHNRYTQKSLSLSPGX

wherein, X is a lysine (K) or is absent

534 IgGl Fc APEAAGGPSVFLFPPKPKDTLYITREPEVTCVVVDVSHEDPEVKFNWY

Hole VDGVEVHNAKTKPREEQYNSTYRVVSVLTVLHQDWLNGKEYKCKVSNK

(AA/R/YTE ALPAPIEKTISKAKGQPREPQVYTLPPSREEMTKNQVSLSCAVKGFYP ) SDIAVEWESNGQPENNYKTTPPVLDSDGSFFLVSKLTVDKSRWQQGNV

FSCSVMHEALHNRYTQKSLSLSPGX

wherein, X is a lysine (K) or is absent

535 Pali VH QVTLRESGPALVKPTQTLTLTCTFSGFSLSTSGMSVGWIRQPPGKALE

(amino WLADIWWDDKKDYNPSLKSRLTISKDTSKNQVVLKVTNMDPADTATYY acid CARSMITNWYFDVWGAGTTVTVSS

sequence )

536 Pali VH caggttacacttagagaatctggtcctgctcttgtgaagcctactcag

(nucleoti actcttactcttacctgcaccttctctggtttcagcctttctacttct de ggtatgtctgtgggttggattagacaacctcctggtaaggctcttgaa sequence ) tggcttgctgatatttggtgggatgataagaaggattacaacccaagc ctgaagtctaggctgaccatttctaaggataccagcaagaatcaggtg gtgcttaaggtgaccaatatggaccctgctgatactgctacttactac tgcgctagatccatgatcaccaactggtactttgatgtttggggtgct ggtactactgtgaccgtttcttcc 537 Pali VL DIQMTQSPSTLSASVGDRVTITCRASQSVGYMHWYQQKPGKAPKLLIY

(amino DTSKLASGVPSRFSGSGSGTEFTLTISSLQPDDFATYYCFQGSGYPFT acid FGGGTKLEIK

sequence )

538 Pali VL gatattcagatgactcagtctccttctaccctgtctgcttctgttggt

(nucleoti gatagggttaccattacatgcagggcttctcaatctgtgggttacatg de cactggtaccaacaaaagcctggtaaggctcctaagctgcttatctac sequence ) gatacctctaagcttgcttctggtgtgccttctaggttttctggttct ggatctggtactgagttcacccttaccatttctagccttcagcctgat gatttcgctacctactattgctttcagggtagcggttaccctttcact tttggtggtggtactaagcttgagattaag

539 A32 VH QVQLQESGPGLVKPSQTLSLSCTVSGGSSSSGAHYWSWIRQYPGKGLE

(amino WIGYIHYSGNTYYNPSLKSRITISQHTSENQFSLKLNSVTVADTAVYY acid CARGTRLRTLRNAFDIWGQGTLVTVSS

sequence )

540 A32 VH caggtgcagctgcaggagtccggccccggactggtcaaaccctctcag

(nucleoti actctgtctctgtcatgtaccgtgtcaggcggctcttccagctccggg de gcacactactggagctggatcaggcagtatcccggcaaggggctggag sequence ) tggatcggatacattcattatagcggcaacacatactataatccttct ctgaagagtcggatcactatttcacagcacaccagcgaaaaccagttc agcctgaagctgaacagcgtgaccgtcgccgacacagccgtgtactat tgcgcccggggcaccagactgagaactctgagaaacgcatttgacatc tggggacaggggacactggtgacagtgagctcc

541 A32 VL QSALTQPPSASGSPGQSVTISCTGTSSDVGGYNYVSWYQHHPGKAPKL

(amino IISEVNNRPSGVPDRFSGSKSGNTASLTVSGLQAEDEAEYYCSSYTDI acid HNFVFGGGTKLTVL

sequence )

542 A32 VL cagagcgcactgactcagcccccttccgcctccgggtctcctggacag

(nucleoti agcgtgacaatctcatgcactgggacttcaagcgatgtgggcgggtac de aactatgtgagttggtaccagcaccatcccgggaaggcacctaaactg sequence ) atcattagcgaagtgaacaatcgaccaagcggcgtccccgaccggttc agcggcagcaagtctggcaataccgccagtctgacagtctcaggcctg caggccgaggatgaagctgagtactattgctcatcatacactgacatc cataacttcgtcttcggcggcggaactaaactgaccgtgctg

543 OK 8 VH QVQLLESGPELLKPGASVKMSCKASGYTFTDYNMHWVKQSHGKSLEWI

GYIYPYTGGTGYNQKFKNKATLTVDSSSSTAYMELRSLTSEDSAVYYC

(CD8 VH) ARNFRYTYWYFDVWGQGTTVTVSS

(amino

acid

sequence ) 544 OK 8 VH caggtgcagctgttggagtctggacctgagctgttgaaacctggggcc tcagtgaagatgtcctgcaaggcttctggatacacattcactgactac

(CD8 VH) aacatgcactgggtgaagcagagccatggaaagagccttgagtggatt

(nucleoti ggatatatttatccttacactggtggcactggctacaaccagaagttc de aagaacaaggccacattgactgtagacagttcctccagcacagcctac sequence ) atggagctccgcagcctgacatctgaggactctgcagtctattactgt gcaaggaactttaggtacacctactggtacttcgatgtctggggccaa ggcaccacggtcaccgtctcctca

545 OK 8 VL DIVMTQSPASLAVSLGQRATISCRASESVDSYDNSLMHWYQQKPGQPP

KVLIYLASNLESGVPARFSGSGSRTDFTLTIDPVEADDAATYYCQQNN

(CD8 VL) EDPYTFGGGTKLEIK

(amino

acid

sequence )

546 OK 8 VL gacatcgtgatgacccagtctccagcttctttggctgtgtctctaggg cagagggccaccatatcctgcagagccagtgaaagtgttgatagttat

(CD8 VL) gacaatagtcttatgcactggtaccagcagaaaccaggacagccaccc

(nucleoti aaagtcctcatctatcttgcatccaacctagaatctggggtccctgcc de aggttcagtggcagtgggtctaggacagacttcaccctcaccattgat sequence ) cctgtggaggctgatgatgctgcaacctattactgtcagcaaaataat gaggatccgtacacgttcgggggggggaccaagctggagatcaaacgt

547 TRX2 VH QVQLVESGGGVVQPGRSLRLSCAASGFTFSDFGMNWVRQAPGKGLEWV

AL

(CD8 VH) IYYDGSNKFYADSVKGRFTISRDNSKNTLYLQMNSLRAEDTAVYYCAK

(amino PHYDGYYHFFDSWGQGTLVTVSS

acid

sequence )

548 TRX2 VH caggttcaattggtggagtctggaggaggcgttgtacagcctggaagg tccctgagactctcatgtgcagcttctggattcactttcagtgacttt

(CD8 VH) ggcatgaactgggttcgacaggctcccgggaaggggctggaatgggtg

(nucleoti gcactgatttactatgatggtagtaacaagttctatgcagactctgtg de aagggtcgattcaccatctccagggacaattctaagaacaccctatac sequence ) ctgcaaatgaacagcctgagagctgaggacacagccgtgtattactgt gcaaaaccccactatgatggttattatcacttctttgattcctggggc caagggacactagtcacagtctcctca

549 TRX2 VL DIQMTQSPSSLSASVGDRVTITCKGSQDINNYLAWYQQKPGKAPKLLI

YNTDILHTGVPSRFSGSGSGTDF FTISSLQPEDIATYYCYQYNNGYT

(CD8 VL) FGQGTKVEIK

(amino

acid

sequence ) 550 TRX2 VL gacatccagatgacccagagcccaagcagcctgagcgccagcgtgggt gacagagtgaccatcacctgtaaaggaagtcaggatattaacaattac

(CD8 VL) ttagcctggtaccagcagaagccaggtaaggctccaaagctgetgate

(nucleoti tacaatacagacattttgcacacgggtgtgccaagcagattcagcggt de agcggtagcggtaccgacttcaccttcaccatcagcagcctccagcca sequence ) gaggacatcgccacctactactgctatcagtataacaacgggtacacg ttcggccaagggaccaaggtggaaatcaaa

551 CH557 VH See Figure 12 (SEQ ID NO: 124)

(amino

acid

sequence )

552 CH557 VH See Figure 12 (SEQ ID NO: 126)

(nucleoti

de

sequence )

553 CH557 VL See Figure 12 (SEQ ID NO: 125)

(amino

acid

sequence )

554 CH557 VL See Figure 12 (SEQ ID NO: 127)

(nucleoti

de

sequence )

555 CH557xCD3 EIVLTQSPATLSASPGERVTLTCRASRSVRNNVAWYQHKGGQSPRLLI

Fc YDASTRAAGVPARFSGSASGTEFTLAISNLESEDFTVYFCLQYNNWWT

FGQGTRVDIKGGGSGGGGEVQLVESGGGLVQPGGSLRLSCAASGFTFS

CHAIN 1 TYAMNWVRQAPGKGLEWVGRIRSKYNNYATYYADSVKGRFTISRDDSK (amino NSLYLQMNSLKTEDTAVYYCVRHGNFGNSYVSWFAYWGQGTLVTVSSA acid STKGEVAACEKEVAALEKEVAALEKEVAALEKGGGDKTHTCPPCPAPE sequence ) AAGGPSVFLFPPKPKDTLMISRTPEVTCVVVDVSHEDPEVKFNWYVDG

VEVHNAKTKPREEQYNSTYRVVSVLTVLHQDWLNGKEYKCKVSNKALP APIEKTISKAKGQPREPQVYTLPPSREEMTKNQVSLWCLVKGFYPSDI AVEWESNGQPENNYKTTPPVLDSDGSFFLYSKLTVDKSRWQQGNVFSC SVMHEALHNHYTQKSLSLSPGK 556 CH557xCD3 gaaattgtgttgacgcagtctccagccaccctgtccgcgtctccaggg Fc gaaagagtcaccctaacttgcagggccagtcggagtgtccgaaacaac gtggcctggtatcagcacaagggtggccagagtcccaggctcctcatt

CHAIN 1 tatgatgcgtccacgagggccgctggtgtcccagccaggttcagcggc (nucleoti agtgcatctgggacagagttcactctcgccatcagcaacttggagtct de gaagattttacagtctacttctgtctgcagtataataactggtggacc sequence ) ttcggccaagggaccagggtggacatcaaaggtggaggatccggcggc ggaggcgaggtgcagctggtggagtctgggggaggcttggtccagcct ggagggtccctgagactctcctgtgcagcctctggattcaccttcagc acatacgctatgaattgggtccgccaggctccagggaaggggctggag tgggttggaaggatcaggtccaagtacaacaattatgcaacctactat gccgactctgtgaagggtagattcaccatctcaagagatgattcaaag aactcactgtatctgcaaatgaacagcctgaaaaccgaggacacggcc gtgtattactgtgtgagacacggtaacttcggcaattcttacgtgtct tggtttgcttattggggacaggggacactggtgactgtgtcttccgcc tccaccaagggcgaagtggccgcatgtgagaaagaggttgctgctttg gagaaggaggtcgctgcacttgaaaaggaggtcgcagccctggagaaa ggcggcggggacaaaactcacacatgcccaccgtgcccagcacctgaa gccgcggggggaccgtcagtcttcctcttccccccaaaacccaaggac accctcatgatctcccggacccctgaggtcacatgcgtggtggtggac gtgagccacgaagaccctgaggtcaagttcaactggtacgtggacggc gtggaggtgcataatgccaagacaaagccgcgggaggagcagtacaac agcacgtaccgtgtggtcagcgtcctcaccgtcctgcaccaggactgg ctgaatggcaaggagtacaagtgcaaggtctccaacaaagccctccca gcccccatcgagaaaaccatctccaaagccaaagggcagccccgagaa ccacaggtgtacaccctgcccccatcccgggaggagatgaccaagaac caggtcagcctgtggtgcctggtcaaaggcttctatcccagcgacatc gccgtggagtgggagagcaatgggcagccggagaacaactacaagacc acgcctcccgtgctggactccgacggctccttcttcctctacagcaag ctcaccgtggacaagagcaggtggcagcaggggaacgtcttctcatgc tccgtgatgcatgaggctctgcacaaccactacacgcagaagagcctc tccctgtctccgggtaaa

557 CH557xCD3 QAVVTQEPSLTVSPGGTVTLTCRSSTGAVTTSNYANWVQQKPGQAPRG

Fc LIGGTNKRAPWTPARFSGSLLGGKAALTITGAQAEDEADYYCALWYSN

LWVFGGGTKLTVLGGGGSGGGGQVRLAQYGGGVKRLGATMTLSCVASG

CHAIN 2 YTFNDYYIHWVRQAPGQGFELLGYIDPANGRPDYAGALRERLSFYRDK (amino SMETLYMDLRSLRYDDTAMYYCVRNVGTAGSLLHYDHWGSGSPVIVSS acid ASTKGKVAACKEKVAALKEKVAALKEKVAALKE

sequence ) 558 CH557xCD3 caggctgtggtgactcaggagccttcactgaccgtgtccccaggcgga Fc actgtgaccctgacatgcagatccagcacaggcgcagtgaccacatct aactacgccaattgggtgcagcagaagccaggacaggcaccaaggggc

CHAIN 2 ctgatcgggggtacaaacaaaagggctccctggacccctgcacggttt (nucleoti tctggaagtctgctgggcggaaaggccgctctgactattaccggggca de caggccgaggacgaagccgattactattgtgctctgtggtatagcaat sequence ) ctgtgggtgttcgggggtggcacaaaactgactgtgctgggagggggt ggatccggcggaggtggacaggtccgactagcccaatatggtggtggg gtgaagaggctaggggccacaatgaccctttcctgcgtggcatctgga tacaccttcaacgactactacatacattgggtgcggcaggcccctgga caaggctttgagttgttgggatacatcgaccccgctaatggtcgccca gactacgcaggggcgttgagggagagactctccttctacagggacaag tccatggagacgctgtacatggacctgaggagcctaagatatgacgac acggccatgtattattgtgttagaaatgtggggaccgctggcagcttg ctgcattatgaccactggggctcgggaagcccggtcatcgtctcctcc gcctccaccaagggcaaagtggccgcatgtaaggagaaagttgctgct ttgaaagagaaggtcgccgcacttaaggaaaaggtcgcagccctgaaa gag

559 CH557xCD3 DKTHTCPPCPAPEAAGGPSVFLFPPKPKDTLMISRTPEVTCVVVDVSH

Fc or EDPEVKFNWYVDGVEVHNAKTKPREEQYNSTYRVVSVLTVLHQDWLNG DH542xCD3 KEYKCKVSNKALPAPIEKTISKAKGQPREPQVYTLPPSREEMTKNQVS Fc LSCAVKGFYPSDIAVEWESNGQPENNYKTTPPVLDSDGSFFLVSKLTV

DKSRWQQGNVFSCSVMHEALHNRYTQKSLSLSPGK

CHAIN 3

(amino

acid

sequence )

560 CH557xCD3 Gacaaaactcacacatgcccaccgtgcccagcacctgaagccgcgggg

Fc ggaccgtcagtcttcctcttccccccaaaacccaaggacaccctcatg atctcccggacccctgaggtcacatgcgtggtggtggacgtgagccac

CHAIN 3 gaagaccctgaggtcaagttcaactggtacgtggacggcgtggaggtg (nucleoti cataatgccaagacaaagccgcgggaggagcagtacaacagcacgtac de cgtgtggtcagcgtcctcaccgtcctgcaccaggactggctgaatggc sequence ) aaggagtacaagtgcaaggtctccaacaaagccctcccagcccccatc gagaaaaccatctccaaagccaaagggcagccccgagaaccacaggtg tacaccctgcccccatcccgggaggagatgaccaagaaccaggtcagc ctgagttgcgcagtcaaaggcttctatcccagcgacatcgccgtggag tgggagagcaatgggcagccggagaacaactacaagaccacgcctccc gtgctggactccgacggctccttcttcctcgtcagcaagctcaccgtg gacaagagcaggtggcagcaggggaacgtcttctcatgctccgtgatg catgaggctctgcacaaccgctacacgcagaagagcctctccctgtct ccgggtaaa 561 Chain 3 EVQLQQSGAELVKPGASVKLSCTASGFNIKDTYIHFVRQRPEQGLEWI of GRIDPANDNTLYASKFQGKATITADTSSNTAYMHLCSLTSGDTAVYYC exemplary GRGYGYYVFDHWGQGTTLTVSSASTKGPSVFPLAPSSKSTSGGTAALG four CLVKDYFPEPVTVSWNSGALTSGVHTFPAVLQSSGLYSLSSVVTVPSS chain SLGTQTYICNVNHKPSNTKVDKRVEPKSCDKTHTCPPCPAPEAAGGPS trivalent VFLFPPKPKDTLMISRTPEVTCVVVDVSHEDPEVKFNWYVDGVEVHNA binding KTKPREEQYNSTYRVVSVLTVLHQDWLNGKEYKCKVSNKALPAPIEKT molecule ISKAKGQPREPQVYTLPPSREEMTK QVSLSCAVKGFYPSDIAVEWES

(amino NGQPENNYKTTPPVLDSDGSFFLVSKLTVDKSRWQQGNVFSCSVMHEA acid LHNRYTQKSLSLSPG

sequence )

562 Chain 4 DVQINQSPSFLAASPGE I INCRTSRSISQYLAWYQEKPGKTNKLLI of YSGSTLQSGIPSRFSGSGSGTDFTLTISGLEPEDFAMYYCQQHNENPL exemplary TFGAGTKLELRRTVAAPSVFIFPPSDEQLKSGTASVVCLLNNFYPREA four KVQWKVDNALQSGNSQESVTEQDSKDSTYSLSSTLTLSKADYEKHKVY chain ACEVTHQGLSSPV KSFNRGEC

trivalent

binding

molecule

(amino

acid

sequence )

563 Chain 3 DVQINQSPSFLAASPGE I INCRTSRSISQYLAWYQEKPGKTNKLLI of YSGSTLQSGIPSRFSGSGSGTDFTLTISGLEPEDFAMYYCQQHNENPL exemplary TFGAGTKLELRGGGGSGGGGSGGGGSEVQLQQSGAELVKPGASVKLSC three TASGFNIKDTYIHFVRQRPEQGLEWIGRIDPANDNTLYASKFQGKA I chain TADTSSNTAYMHLCSLTSGDTAVYYCGRGYGYYVFDHWGQGTTLTVSS trivalent ASTKGPSVFPLAPSSKSTSGGTAALGCLVKDYFPEPVTVSWNSGALTS

GVHTFPAVLQSSGLYSLSSVVTVPSSSLGTQTYICNVNHKPSNTKVDK

binding

RVEPKSCDKTHTCPPCPAPEAAGGPSVFLFPPKPKDTLMISRTPEVTC

molecule

VVVDVSHEDPEVKFNWYVDGVEVHNAKTKPREEQYNSTYRVVSVLTVL

(amino

HQDWLNGKEYKCKVSNKALPAPIEKTISKAKGQPREPQVYTLPPSREE

acid

MTK QVSLSCAVKGFYPSDIAVEWESNGQPENNYKTTPPVLDSDGSFF

sequence ) LVSKLTVDKSRWQQGNVFSCSVMHEALHNRYTQKSLSLSPG

564 CH556 VH See Fig. 15A (SEQ ID NO 138)

(amino

acid

sequence )

565 CH556VL See Fig. 158 (SEQ ID NO 149)

(amino

acid

sequence ) 566 CH555VH See Fig. 15A (SEQ ID NO 136)

(amino

acid

sequence )

567 CH555VL See Fig. 15B (SEQ ID NO 148)

(amino

acid

sequence )

568 CH493VH See Fig. 18 (SEQ ID NO 162)

(amino

acid

sequence )

569 CH493VH See Fig. 19 (SEQ ID NO 168)

(nucleoti

de

sequence )

570 CH493VL* See Fig. 18 (SEQ ID NO 163)

(amino *CH493VL is VL from CH236 acid

sequence )

571 CH493VL* See Fig. 19 (SEQ ID NO 169)

(nucleoti *CH493VL is VL from CH236 de

sequence )

572 CH492VH See Fig. 18 (SEQ ID NO 161)

(amino

acid

sequence )

573 CH492VH See Fig. 19 (SEQ ID NO 167)

(nucleoti

de

sequence ) 574 CH492VL* See Fig. 18 (SEQ ID NO 164)

(amino *CH492VL is VL from CH241 acid

sequence )

575 CH492VL* See Fig. 19 (SEQ ID NO 170)

(nucleoti *CH492VL is VL from CH241 de

sequence )

576 CH491VH See Fig. 18 (SEQ ID NO 160)

(amino

acid

sequence )

577 CH491VH See Fig. 19 (SEQ ID NO 166)

(nucleoti

de

sequence )

578 CH491VL* See Fig. 18 (SEQ ID NO 163)

(amino *CH491VL is VL from CH236 acid

sequence )

579 CH491VL* See Fig. 19 (SEQ ID NO 169)

(nucleoti *CH491VL is VL from CH236 de

sequence )

580 CH490VH See Fig. 18 (SEQ ID NO 159)

(amino

acid

sequence )

581 CH490VH See Fig. 19 (SEQ ID NO 165)

(nucleoti

de

sequence ) 582 CH490VL* See Fig. 18 (SEQ ID NO 164)

(amino *CH490VL VL from CH241 acid

sequence )

583 CH490VL* See Fig. 19 (SEQ ID NO 170)

(nucleoti *CH490VL is VL from CH241 de

sequence )

584 DH542VH See Fig. 1 (SEQ ID NO 3)

(amino

acid

sequence )

585 DH542VH See Fig. 1 (SEQ ID NO 1)

(nucleoti

de

sequence )

586 DH542VL See Fig. 1 (SEQ ID NO 4)

(amino

acid

sequence )

587 DH542VL See Fig. 1 (SEQ ID NO 2)

(nucleoti

de

sequence )

588 DH542 QSA See Ex.10 SEQ ID NO 179)

VH

(amino

acid

sequence ) 589 DH542 QSA See Ex.10 (SEQ ID NO 178) VH

(nucleoti

de

sequence )

590 DH542 QSA See Ex.10 (SEQ ID NO 181)

VL

(amino

acid

sequence )

591 DH542 QSA See Ex.10 (SEQ ID NO 180)

VL

(nucleoti

de

sequence )

592 DH542 L4 See Fig. 1 (SEQ ID NO 3)

VH*

*DH542 L4VH is DH542VH

(amino

acid

sequence )

593 DH542 L4 See Fig. 1 (SEQ ID NO 1)

VH*

*DH542 L4VH is DH542VH

(nucleoti

de

sequence )

594 DH542 L4 See Fig. 2 (SEQ ID NO 51)

VL*

*DH542 L4VL is VL of DH429

(amino

acid

sequence )

595 DH542 L4 See Fig. 2 (SEQ ID NO 39)

VL*

*DH542 L4VL is VL of DH429

(nucleoti

de

sequence ) 596 DH542xCD3 TSLLTQPASVSGSPGQS IT I SCTGTKYDVGSHDLVSWYQQYPGKVPKY Fc MIYEVNKRPSGVSNRFSGSKSGNTASLT I SGLRAEDEADYYCCS FGGS

ATVVCGGGTKVTVLGGGSGGGGEVQLVE SGGGLVQPGGSLRLSCAASG

CHAIN 1 FT FSTYAMNWVRQAPGKGLEWVGRIRSKYNNYATYYADSVKGRFT I SR

DDSKNSLYLQMNSLKTEDTAVYYCVRHGNFGNSYVSWFAYWGQGTLVT VS SASTKGE VAACEKE VAALEKE VAALEKE VAALEKGGGDKTHTCPPC PAPEAAGGPSVFLFPPKPKDTLMI SRTPEVTCVVVDVSHEDPEVKFNW YVDGVE VHNAKTKPREEQYNSTYRVVSVLTVLHQDWLNGKEYKCKVSN KALPAP IEKT I SKAKGQPRE PQVYTLPPSREEMTKNQVSLWCLVKGFY PSDIAVEWESNGQPENNYKTTPPVLDSDGS FFLY SKLTVDKSRWQQGN VFSCSVMHEALHNHYTQKSLSLSPGK

597 DH542xCD3 QAVVTQEPSLTVSPGGTVTLTCRS STGAVTTSNYANWVQQKPGQAPRG

Fc LIGGTNKRAPWT PARFSGSLLGGKAALT ITGAQAEDEADYYCALWY SN

LWVFGGGTKLTVLGGGGSGGGGQVQLVQSGAQMKNPGASVKVSCAPSG

CHAIN 2 YT FTDFY I HWLRQAPGQGLQWMGWMNPQTGRTNTARNFQGRVTMTRDT

S IGTAYMELRSLTSDDTAIYYCTTGGWI SLYYDS SYYPNFDHWGQGTL LTVS SASTKG KV AAC KE KVAAL KE KV AAL KE KVAAL KE

[0245] Various antigen binding domains capable of binding to an epitope of HIV- 1, CD3, CD 16 and CD8 are contemplated by the invention and are disclosed in the specification. Table 4 discloses non-limiting embodiments of such antigen binding domains which may be incorporated into the multispecific molecules of the invention. Additional, alternative antigen binding domains from other antibodies having specificity for the desired antigens may be utilized. Many such antibodies are known in the art, for example additional anti-CD3 antibodies are described in WO2012/162067 and WO 2014/110601 the contents of each of which are hereby incorporated by reference; additional anti-CD 16 antibodies are described in WO 03/101485 the contents of which is hereby incorporated by reference; and additional anti- CD8 antibodies are described in WO 2014/164553 the contents of which is hereby

incorporated by reference.

[0246] Various Heterodimer-Promoting Domain (HPD) sequences are contemplated by the invention and disclosed in the specification. Table 4 discloses non-limiting embodiments of various HPDs. In some embodiments the HPD include E/K-coils (SEQ ID NOs: 518, 510) or cysteine engineered E/K-coils (SEQ ID NOs: 519, 521). In some embodiments HPD includes combinations of SEQ ID NOs: 511, 512, 513, and 514 sequences (e.g., SEQ ID NOs: 511 and 513; SEQ ID NOs: 512 and 513; SEQ ID NOs: 511 and 514; SEQ ID NOs: 512 and 514); kappa and lambda light chain constant domains (SEQ ID NOs: 516 and 517). In some embodiments HPD include any suitable sequences with a Cystein residue to permit disulfide bond. In some embodiments HPD includes suitable CHI and CL domains.

[0247] A skilled artisan readily appreciates that the disclosure throughout the application of various design elements such as but not limited to HPDs and sequences, linkers, CH2-CH3 domain and Fc domain and their variants is applicable and could be used in any of the designs of the multispecific molecules of the invention that comprise these design elements.

[0248] HIV-1 Antibodies

[0249] Broadly neutralizing and potent HIV-1 envelope glycoprotein (Env) antibodies are now being developed for both prevention of HIV-1 (Rudicell RS et al. J. Virol 88: 12669,-82, 2014) and for treatment of HIV-1 infected individuals (Barouch DH, et al. Nature 503 : 224-8, 2013; Shingai M et al. Nature 503 : 277-80, 2013). Thus, human recombinant antibodies either alone or in combinations have great prophylactic and therapeutic potential for the prevention and treatment of HIV-1 infection. Moreover, antibodies that bind with high affinity to Env may be useful in eliminating the latent pool of HIV-1 -infected CD4 T cells and curing HIV-1 infection, when either used to sensitize HIV-1 expressing target cells with bispecific bnAbs for NK or CD8 T cell killing or when bnAbs are conjugated with toxins or radionucleotides.

[0250] In certain aspects the invention provides fully human antibodies and fragments that specifically bind to and potently neutralize various isolates of HIV-1. In some embodiments, the antibodies bind to HIV-1 env V3 glycan. In some embodiments, the antibodies of the invention bind to HIV-1 gpl20 Env CD4 binding site.

[0251] In certain aspects the invention provides pharmaceutical compositions including these human antibodies and a pharmaceutically acceptable carrier. In certain aspects the invention provides antibodies for passive immunization against HIV/ AIDS. Nucleic acids encoding these antibodies, expression cassettes and vectors including these nucleic acids, and isolated cells that express the nucleic acids which encode the antibodies of the invention are also provided.

[0252] In some embodiments, the invention provides antibodies which are clonal variants . In some embodiments, clonal variants are sequences that differ by one or more nucleotides or amino acids, and have a V region with shared mutations compared to the germline, identical VHDJH or VJH gene usage, identical or similar HCDR3 length, and the same VL and JL usage. The germline sequence (unmutated common ancestor "UCA") is intended to be the sequence coding for the antibody/immunoglobulin (or of any fragment thereof) deprived of mutations, for example somatic mutations. Antibodies in a clone that are designated as UCA and/or I (for "Intermediate") are typically not identified from a biological sample, but are derived computationally based on VH and/or VL sequences identified from subjects infected with HIV-1.

[0253] Compositions including the human antibodies of the invention, including V3 glycan and CD4 binding site antibodies, can be used for any purpose including but not limited to research, diagnostic and therapeutic purposes.

[0254] The neutralization breadth of the inventive antibodies is demonstrated by the diversity of viruses which are neutralized in the TZMbl Env pseudovirus inhibition assay. In certain embodiments, the neutralization breadth and/or binding of the antibodies of the invention can be maintained in the presence of tolerate changes to the epitope. Comparing the sequences of the neutralized viruses, versus viruses that are not neutralized, a skilled artisan can readily determine the % virus changes, including changes in the epitope, which can be tolerated while neutralization and/or binding is maintained.

[0255] Comparing the sequences of the antibodies and their neutralization properties, a skilled artisan can readily determine sequence identity, compare sequence length and determine the % sequence identity and/or changes, including % sequence identity and/or changes in the VH and VL sequences, including % sequence identity and/or changes in the CDRs, as well as the specific positions and types of substitutions which can be tolerated while neutralization potency and breadth is maintained.

[0256] Various algorithms for sequence alignment are known in the art. The similarity between amino acid sequences is expressed in terms of the similarity between the sequences, otherwise referred to as sequence identity. Sequence identity is frequently measured in terms of percentage identity (or similarity or homology); the higher the percentage, the more similar the two sequences are. Homologs or variants of a polypeptide will possess a relatively high degree of sequence identity when aligned using standard methods.

[0257] Methods of alignment of sequences for comparison are well known in the art. Various programs and alignment algorithms are described in: Smith and Waterman, Adv. Appl. Math. 2:482, 1981; Needleman and Wunsch, J. Mol. Biol. 48:443, 1970; Pearson and Lipman, Proc. Natl. Acad. Sci. U.S.A. 85:2444, 1988; Higgins and Sharp, Gene 73 :237, 1988; Higgins and Sharp, CABIOS 5: 151, 1989; Corpet et al., Nucleic Acids Research 16: 10881, 1988; and Pearson and Lipman, Proc. Natl. Acad. Sci. U.S.A. 85:2444, 1988. Altschul et al., Nature Genet. 6: 119, 1994, presents a detailed consideration of sequence alignment methods and homology calculations.

[0258] The NCBI Basic Local Alignment Search Tool (BLAST) (Altschul et al., J. Mol. Biol. 215:403, 1990) is available from several sources, including the National Center for

Biotechnology Information (NCBI, Bethesda, Md.) and on the internet, for use in connection with the sequence analysis programs blastp, blastn, blastx, tblastn and tblastx. A description of how to determine sequence identity using this program is available on the NCBI website on the internet.

[0259] Homologs and variants of a VL or a VH of an antibody that specifically binds a polypeptide are typically characterized by possession of at least about 75%, for example at least about 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98% or 99% sequence identity counted over the full length alignment with the amino acid sequence of interest.

Proteins with even greater similarity to the reference sequences will show increasing percentage identities when assessed by this method, such as at least 80%, at least 85%, at least 90%), at least 95%, at least 98%, or at least 99% sequence identity. When less than the entire sequence is being compared for sequence identity, homologs and variants will typically possess at least 80% sequence identity over short windows of 10-20 amino acids, and may possess sequence identities of at least 85% or at least 90% or 95% depending on their similarity to the reference sequence. Methods for determining sequence identity over such short windows are available at the NCBI website on the internet. One of skill in the art will appreciate that these sequence identity ranges are provided for guidance only; it is entirely possible that strongly significant homologs could be obtained that fall outside of the ranges provided.

[0260] As used herein, reference to antibodies, without explicit mention of antibody fragments and antibody-fragment comprising molecules, may encompass antibody fragments and antibody-fragment comprising molecules.

[0261] In certain embodiments, the invention provides antibodies and antibody-fragment comprising molecules, including multispecific molecules, such as, but not limited to bispecific and trispecific molecules (e.g., bispecific antibodies, bispecific diabodies, trivalent binding molecules, etc.) which are 99%, 98%, 97%, 96%, 95%, 94%, 93%, 92%, 91%, 90%, 89%, 88%, 87%, 86%, 85%, 84%, 83%, 82%, 81%, 80% identical to the VH and VL amino acid sequences of the antibodies described herein and still maintain the neutralization breadth, biding and/or potency. In certain embodiments, the invention provides antibodies and antibody-fragment comprising molecules, including multispecific molecules, such as, but not limited to bispecific and trispecific molecules (e.g., bispecific antibodies, bispecific diabodies, trivalent binding molecules, etc.) which are 99%, 98%, 97%, 96%, 95%, 94%, 93%, 92%, 91%, 90%, 89%, 88%, 87%, 86%, 85%, 84%, 83%, 82%, 81%, 80% identical to the CDRl, 2, and/or 3 of VH and CDRl, 2, and/or 3 VL amino acid sequences of the antibodies described herein and still maintain the neutralization breadth, biding and/or potency.

[0262] In certain embodiments, the invention provides antibodies and antibody-fragment comprising molecules, including multispecific molecules, such as, but not limited to bispecific and trispecific molecules (e.g., bispecific antibodies, bispecific diabodies, trivalent binding molecules, etc.) which can tolerate a larger percent variation in the sequences outside of the VH and/VL sequences of the antibodies. In certain embodiments, the invention provides antibodies which are 99%, 98%, 97%, 96%, 95%, 94%, 93%, 92%, 91%, 90%, 89%, 88%, 87%, 86%, 85%, 84%, 83%, 82%, 81%, 80%, 79%, 78%, 77%, 76%, 75%, 74%, 73%, 72%, 71%, 70%, 69%, 68%, 67%, 66%, 65% identical, wherein the identity is outside of the VH or VL regions, or the CDRs of the VH or VL chains of the antibodies described herein.

[0263] Antibodies of the invention are expected to have the same binding specificity, for example as intact immunoglobulins and antigen binding variants or fragments e.g. as a number of well characterized fragments produced by digestion with various peptidases. For instance and without limitation, Fabs, Fvs, scFvs are fragments which are expected to have the same binding specificities as intact antibodies. Binding specificity can be determined by any suitable assay in the art, for example but not limited competition binding assays, epitope mapping, etc. A scFv protein is a fusion protein in which a light chain variable region of an immunoglobulin and a heavy chain variable region of an immunoglobulin are bound by a linker, while in dsFvs, the chains have been mutated to introduce a disulfide bond to stabilize the association of the chains. Provided are also genetically engineered forms such as chimeric antibodies and heteroconjugate antibodies such as bispecific antibodies. See also, Pierce Catalog and Handbook, 1994-1995 (Pierce Chemical Co., Rockford, 111.); Kuby,

Immunology, 3.sup.rd Ed., W.H. Freeman & Co., New York, 1997.

[0264] In certain embodiments the invention provides antibody fragments and molecules comprising antibody fragments,, which have the binding specificity and/or properties of the inventive antibodies. Non-limiting examples include: (1) Fab, the fragment which contains a monovalent antigen-binding fragment of an antibody molecule produced by digestion of whole antibody with the enzyme papain to yield an intact light chain and a portion of one heavy chain; (2) Fab', the fragment of an antibody molecule obtained by treating whole antibody with pepsin, followed by reduction, to yield an intact light chain and a portion of the heavy chain; two Fab' fragments are obtained per antibody molecule; (3) (Fab¹). sub.2, the fragment of the antibody obtained by treating whole antibody with the enzyme pepsin without subsequent reduction; (4) F(ab').sub.2, a dimer of two Fab' fragments held together by two disulfide bonds; (5) Fv, a genetically engineered fragment containing the variable region of the light chain and the variable region of the heavy chain expressed as two chains; and (6) single chain antibody ("SCA"), a genetically engineered molecule containing the variable region of the light chain, the variable region of the heavy chain, linked by a suitable polypeptide linker as a genetically fused single chain molecule. In certain embodiments, the antibody fragments can be produces recombinantly.

[0265] In certain embodiments, VH refers to the variable region of an immunoglobulin heavy chain, including but not limited to that of an antibody fragment, such as Fv, scFv, dsFv or Fab. In certain embodiments, VL refers to the variable region of an immunoglobulin light chain, including but not limited to that of an Fv, scFv, dsFv or Fab.

[0266] Any of the nucleic acids encoding any of the antibodies, or fragment thereof can be expressed in a recombinantly engineered cell such as bacteria, plant, yeast, insect and mammalian cells. The nucleic acid sequences include any sequence necessary for expression, including but not limited to a promoter, a leader sequence. These antibodies can be expressed as individual VH and/or VL chain, or can be expressed as a fusion protein. In certain embodiments, the antibodies can be expressed by viral vector mediated delivery of genes encoding the antibodies of the invention (See e.g. Yang et al. Viruses 2014, 6, 428-447).

[0267] The present invention also encompasses molecules comprising a hinge domain. The hinge domain be derived from any immunoglobulin isotype or allotype including IgA, IgD, IgG, IgE and IgM. In preferred embodiments, the hinge domain is derived from IgG, wherein the IgG isotype is IgGl, IgG2, IgG3 or IgG4, or an allotype thereof. The hinge domain may be engineered into a polypeptide chain comprising the diabody molecule together with an Fc domain such that the diabody molecule comprises a hinge-Fc domain. In certain

embodiments, the hinge and Fc domain are independently selected from any immunoglobulin isotype known in the art or exemplified herein. In other embodiments the hinge and Fc domain are separated by at least one other domain of the polypeptide chain, e.g., the VL domain. [0268] It is readily understood that the nucleic acid sequences disclosed in the application are non-limiting embodiments of representative nucleotide sequences encoding the respective amino acid sequences.

[0269] The contents of the various publications cited throughout the specification are incorporated by reference in their entirety.

[0270] The following examples are provided to illustrate particular features of certain embodiments, but the scope of the claims should not be limited to those features exemplified.

EXAMPLES

Example 1: Isolating antibodies from natural HIV-1 infected

individuals

[0271] Methods to identify and isolate antigen specific reactive antibodies were carried out essentially as described in Liao HX et al. J. Virol. Methods 158: 171-9, 2009. Specific hooks are designed to identify antibodies which bind to specific HIV-1 envelope targets/antigens. Using such hooks, with fluorophore labeled streptavidin in two colors, cells are sorted by flow cytometry, into single wells, and the diagonally (that reacted with both colors hooks) reactive memory B cells are picked. B cells enriched from PBMC are sorted, and plated at limiting dilution (as single cell per well). Optionally, these cultures are grown and supernatants are functionally characterized.

[0272] PCR on these cells is carried out according to the protocol in Liao HX et al. J. Virol. Methods 158: 171-9, 2009. PCR amplifications are carried out to amplify rearranged VH and VL fragment pairs from the diagonally sorted memory B cells (Liao et al JVM). Overlapping PCR is used to construct full length Ig heavy and Ig light linear genes comprising the rearranged VH and VL fragment pairs. RT-PCR and PCR reactions is carried out essentially as described in Liao HX et al. J. Virol. Methods 158: 171-9, 2009, see for example Figure 1, Section 3.3. Sequence analysis of the VH and VL genes was carried out to determine the VH and VL gene usage, CDR lengths, the % mutation of HCDR3 and LCDR3. Based on this sequence analysis, one to two pairs of linear VH and VL genes are selected and made in linear cassettes (essentially as described in Liao HX et al. J. Virol. Methods 158: 171-9, 2009, see for example Figure 1, Section 3.3) to produce recombinant monoclonal antibodies by transient transfection, e.g. in 293T cells. [0273] Recombinant antibodies are grown and supernatants and/or purified antibodies are functionally characterized.

[0274] Pairs of VH and VL genes as selected above can also be used to produce plasmids for stable expression of recombinant antibodies. In certain embodiments, the plasmids or linear constructs for recombinant antibody expression also comprise AAAA substitution in and around the Fc region of the antibody that has been reported to enhance ADCC via K cells (AAA mutations) containing the Fc region aa of S298A as well as E333A and K334A

(Shields RI et al JBC , 276: 6591-6604, 2001) and the 4^th A (N434A) is to enhance FcR neonatal mediated transport of the IgG to mucosal sites (Shields RI et al. ibid).

[0275] The antibodies of the invention were selected based on a combination of criteria including sequence analyses, and functional analyses including but not limited as

neutralization breadth, and potency.

[0276] In certain embodiments, the antibodies of the invention comprise naturally rearranged VH and VL fragment pairs, wherein the rest of the Ig gene is not naturally occurring with the identified rearranged VH and VL fragments. In certain embodiments, the antibodies of the invention are recombinantly produced.

Example 2: TZM-bl cells pseudo-viruses neutralization assay

[0277] TZMbl neutralization assay is a standard way to evaluate antibody breadth and potency. See Montefiori, D. Methods Mol Biol. 2009;485:395-405; HIV-1 Env- pseudoviruses infection of TZM-bl cells. Exemplary pseudovirus neutralization assays and panels of HIV-1 pseudovirus are described for example, in Li et al., J Virol 79, 10108-10125, 2005, Seaman et al, J. Virol., 84: 1439-1452, 2010; Sarzotti-Kelsoe et al., J. Immunol.

Methods, 409: 131-46, 2014; and WO2011/038290, each of which is incorporated by reference herein. Various HIV-1 isolates, both Tier 1 and Tier 2 viruses can be included in this assay.

[0278] The TZMbl assay was conducted to determine neutralization potency and breadth of the various antibodies of the invention on different HIV-1 pseudoviruses.

[0279] Figure 27A shows summary results of neutralization data of CH557, CH235, VRCOl, VRC07-523-LS, N6, 3BNC117, 8ANC131, CH103, F105, and DH522 against a panel of HIV-1 isolates in the Luc/TZM-bl neutralization assay. Values represent IC50 in μg/ml.

[0280] Figure 27B shows the mean IC50 and percent of isolates neutralized at different IC50 values. Figure 28A shows summary results of neutralization data of CH557, CH235, VRCOl, VRC07-523-LS, N6, 3BNC 1 17, 8ANC131, CH103, F 105, and DH522 against a panel of HIV-1 isolates in the Luc/TZM-bl neutralization assay. Values represent IC80 in μg/ml.

[0281] Figure 28B shows the mean IC80 and percent of isolates neutralized at different IC80<50ug/ml values.

Example 3: Epitope mapping of antibodies

[0282] Binding and/or neutralization assays using various envelop antigens can be used to determine the envelop epitope recognized by these antibodies.

Example 4: Kd determination

[0283] Kd measurements of antibody binding to HIV-1 envelope, e.g. gpl20 or any other suitable peptide, will be determined by Surface Plasmon Resonance measurements, for example using Biacore, or any other suitable technology which permits detection of interaction between two molecules in a quantitative way.

[0284] Various assays and experiments can be designed to analyze prevention, treatment and/or cure.

Example 5: Assay for self-reactivity

[0285] Table 9 below summarizes some of the known types of disease associated antibodies.

I l i slones Drug i nduced S I T! [0286] Various assays for self-reactivity of human antibodies are known in the art. AtheNA Multi-Lyte ANA Plus Test System is one such assay. This is luminex-based assay, which is also used to screen patient sera.

Example 6: Antibodies from CH235 lineage

DH493 (also referred as CH493) and DH491 (also referred as CH491)

[0287] This example describes the design and making of non-naturally occurring CD4bs broad neutralizing HIV-1 antibodies

[0288] Monoclonal antibody CH493 was designed as follows: the heavy chain VDJ rearrangement was derived from genomic DNA deep sequencing performed on memory B cells isolated from PBMCs of the 703-01-050-5 subject (CHAVIOOl protocol) obtained 152 weeks post-infection. Other V-heavy chain VDJ rearrangement sequences were retrieved with this technology from multiple time points. Figure 20 shows the heavy chain phylogenetic tree including all the sequences retrieved with deep sequencing (in black: all except for CH240, CH239, CH235, CH236, CH241). V-light chains were not identified.

[0289] Figure 20 shows the heavy chain phylogenetic tree including all the sequences retrieved from RNA of cultured memory B cells isolated from PBMCs obtained 41 weeks post-infection (in red: CH240, CH239, CH235, CH236, CH241). V-light chains of these antibodies from cultures were also identified.

[0290] Four recombinant non-naturally occurring antibodies were produced using V-heavy VDJ rearrangements identified from deep sequencing and they are called CH490, CH491, CH492 and CH493. These V-heavy sequences were chosen because they are the most mutated.

[0291] The V-heavy chains were paired with the V-light chains of the antibodies identified from memory B cell cultures that are closest in the phylogenetic tree shown above.

Therefore, CH491 and CH493 heavy chains were paired with CH236 light chain, while CH490 and CH492 heavy chains were paired with the CH241 light chain,

[0292] CH235 antibodies from week 41 are characterized by their inability to bind and neutralize CH505.TF envelope mutants with specific point mutations in the D loop (Gao, Bonsignori, Liao et al. Cell 158, 481-491, July 31, 2014, see also U.S. Provisional Application No. 62/027,427 filed July 22, 2014, and U.S. Provisional Application No. 61/972,531 filed March 31, 2014). [0293] Figure 21 shows that CH493 restores almost completely the ability to bind the mutants not recognized by the early members of the lineage, indicating that this antibody is not constrained by the amino acid makeup of the D loop as tightly as the naturally occurring CH235 early lineage antibodies. CH493 still retained differential binding to the CH505.TF gpl20 delta 3711 mutant. Binding dependence to 1371 is a hallmark of neutralizing CD4bs antibodies.

[0294] Most notably, CH493 neutralized 20 of 24 tier-2 HIV-1 viruses (83%) in a multiclade virus panel optimized to represent diversity among globally circulating viruses. Naturally occurring antibodies in the same lineage neutralized only 25% of the viruses in the same panel and other engineered antibodies neutralized max 46% (Figure 22).

[0295] CH235, CH236, CH239, CH240 and CH241 are all Abs with natural pairs VH and VL from week 41 of infection of CH505 individual. CH241 is the most mutated at ~11 %.

CH490, CH491, CH492 and CH493 are antibodies which comprise VH chains identified by deep sequencing and are near 20% mutated. These VH chains were paired with VL chains from the closest natural pair antibodies. So in the natural tree, CH241 is the most mutated from week 41 and it hit 38% of isolates.

[0296] For Abs CH491 and CH493, the VH chains were complemented with the VL of

CH236. For antibodies CH490 and CH492 that were complemented with the VL of CH241.

CH491 neutralized 46% of isolates and CH493 neutralized 83%.

[0297] In summary these antibodies were designed to include heavy chains form 454 sequencing and the heavy chains were paired with VL from observed antibodies:

[0298] CH490 and CH492 VH from 454 were paired with CH241 VL.

[0299] CH491 and CH493 VH from 454 were paired with CH236 VL. Pairing was done with the observed VL that was closer to the 454 sequence in the phylogenetic tree.

[0300] VH chains source: CH490 is from week 66; CH491 and CH492 are from week 100;

CH493 is from week 152. The heavy chains were selected because were the most mutated ones.

Example 7: Antibodies from CH235 lineage Antibodies CH555, CH556, CH557and CH558

[0301] CH505 transmitted/founder (CH505.TF) gpl20 Env-specific memory B cells were isolated from 20 million PBMCs using fluorescent-activated single cell sorting (FACS) collected from chronically HIV-1 infected African CHAVI subject 703-01-050-5 264 and 323 weeks post-infection (10 million PBMCs/timepoint). Viable memory B cells were defined as AquaVital Dye neg, CD16neg, CD14neg, CD3neg, CD19pos, IgDneg cells. Sorted cells were cultured overnight in RPMI + 10%FCS supplemented with 2.5 ug/mL ODN2006, 2.5 ug/mL CFK2-inhibitor, 50ng/mL rHu IL-21 and 1 : 1 EBV-containing supematants in a 96-well plates well containing 5000 CD40L-expressing MS40L feeder cells. Cells were then plated at limiting dilution in 96-well plates containing feeder cells, 2.5 ug/mL ODN2006, 2.5 ug/mL CFD 2-inhibitor and 50ng/mL rHu IL-21, and cultured for 14 days, with medium refresh at days 3, 7 and 10. Culture supematants were collected at day 14 and tested for neutralization of CH505.TF virus and binding to CH505.TF gpl20, CH505.TF gpl20 delta371I, RSC3 protein, RSC3 protein delta371I and CH505.TF mutant envelopes M6, M8 and M20

(described in Gao, Bonsignori, Liao et al. Cell 2014. Jul 31;158(3):481-91). From selected positive cultures, recombinant monoclonal antibodies were produced as previously described (Bonsignori et al. J Virol. 2011. Oct;85(19):9998-10009). Monoclonal antibodies CH557 was identified from a culture that neutralized 91% infectivity of CH505.TF virus, differentially bound to CH505.TF gpl20 Env and CH505.TF gpl20 delta371I Env, bound to M6, weakly bound to M8 but did not bind to M20 CH505.TF mutant gpl20s. CH557 was identified from memory B cells collected 323 weeks post-infection. CH557 is a member of the previously described CH235 clonal lineage (Gao, Bonsignori, Liao et al. Cell 2014. Jul 31; 158(3):481- 91). From the same experiment we also identified monoclonal antibodies CH555, CH556 and CH558 - all members of the CH235 clonal lineage. All monoclonal antibodies but CH555, which was identified from memory B cells collected 236 weeks post-infection, are from week 323 post-infection.

Example 8: Maturation Pathway from Germline to Broad HIV-1

Neutralizer of a CD4-Mimic Antibody

[0302] See Bonsignori et al. Cell 165, 1-15, April 7, 2016, published on-line March 3, 2016, the contents of which are hereby incorporated by reference in its entirety. Antibodies with ontogenies from V_H1-2 or V_Hl-46-germline genes dominate the broadly neutralizing response against the CD4-binding site (CD4bs) on HIV-1. Here we define with longitudinal sampling from time-of-infection the development of a V_Hl-46-derived antibody lineage that matured to neutralize 90% of HIV-1 isolates. Structures of lineage antibodies CH235 (week 41 from time-of-infection, 18% breadth), CH235.9 (week 152, 77%) and CH235.12 (week 323, 90%) demonstrated the maturing epitope to focus on the conformationally invariant portion of the CD4bs. Similarities between CH235 lineage and five unrelated CD4bs lineages in epitope focusing, length-of-time to develop breadth, and extraordinary levels of somatic

hypermutation suggested commonalities in maturation among all CD4bs antibodies.

Fortunately, the required CH235-lineage hypermutation appeared substantially guided by the intrinsic mutability of the V_H1-46 gene, which closely resembled V_H1-2. The CH235-lineage findings were integrated with a second broadly neutralizing lineage and HIV-1 co-evolution to suggest a vaccination strategy for inducing both lineages.

[0303] INTRODUCTION

[0304] Understanding the pathways and mechanisms of broadly neutralizing antibody (bnAb) induction is a critical goal of HIV-1 vaccine development (Bonsignori et al., 2012; Haynes, 2015; Haynes and Bradley, 2015; Haynes et al., 2012; Mascola and Haynes, 2013). In chronic HIV-1 infections, breadth of plasma neutralization follows a uniform distribution and broad neutralization arises in -50% of individuals after 5 years or more of infection (Hraber et al., 2014). The delayed appearance of bnAbs suggests roadblocks to their development, and one vaccine approach is to decipher these roadblocks and devise strategies to overcome them. It is possible that - because of the high diversity of antibodies resulting from recombination and somatic hypermutation (SHM) - different bnAb lineages may have different developmental pathways and roadblocks. However, for the CD4-binding site (CD4bs), a population-level analysis on 14 donors indicated only two general types of CD4bs bnAbs: V_H-gene restricted and CDR H3-dominated (Zhou et al., 2015).

[0305] The V_H-gene restricted classes arise from two highly similar V_H -genes: V_H1-2 and V_H1-46 (Scheid et al., 2011; Wu et al., 2011). V_Hl-2*02 and V_H1-46*01 share 93.4%

(269/288) nucleotide sequence identity. Both classes give rise to antibodies that recognize the CD4bs via V_H structural mimicry of the immunoglobulin-like N-terminal domain of CD4 (Zhou et al., 2010; Zhou et al., 2015). For the V_H1-2 gene-derived antibodies, analysis of their ontogeny suggests two roadblocks based on: (i) a requirement for high levels of SHM (Klein et al., 2013; Scheid et al., 2009; Scheid et al., 2011; Wu et al., 2010), and (ii) weak binding of the inferred unmutated common ancestor (UCA) to gpl20 (Jardine et al., 2013; McGuire et al., 2013; Scheid et al., 2011; Wu et al., 2011; Zhou et al., 2010; Zhou et al., 2015), although a definitive analysis from time-of-infection had not yet provided detail. In addition, several of the CD4bs bnAbs are autoreactive with ubiquitinase enzymes (Bonsignori et al., 2014; Liao et al., 2013; Liu et al., 2015). [0306] Structure-based design of UCA-interacting immunogens has recently demonstrated a means to overcome this second roadblock, with priming of V_Hl-2 bnAb lineages in knock-in mice (Dosenovic et al., 2015; Jardine et al., 2015). However, the maturation of primed V_H1-2 CD4bs B cell lineages to broad neutralization as well as the mechanism for the development of breadth remain unresolved.

[0307] For the V_Hl-46-derived antibodies, far less is known. Two chronically HIV-infected individuals, RU1 and RU8, have developed V_Hl-46-derived bnAbs, 1B2530 and 8ANC131 (Scheid et al., 2011). An African individual (donor CH505) was recently described who, over time, developed a CD4bs bnAb lineage (the CHI 03 lineage) that recognized the CD4 supersite through a CDR H3-dominated mode of interaction (Liao et al., 2013). Analysis of the co-evolution between virus and CHI 03 lineage demonstrated a second B cell lineage (the CH235 lineage) that cooperated by selection of escape mutants from the CH235 lineage that drove the CH103 bnAb lineage (Gao et al., 2014). Described herein is the finding that the CH235 lineage itself progressed to bnAb over 5 years of affinity maturation. Described herein are sequences of the CH235 lineage that were identified through longitudinal samples of 17 time points spanning weeks (wks) 6-323 post infection, assessment of the neutralization breadth of sequential lineage members on a panel of -200 diverse isolates, and determination of Env-complexed crystal and EM structures for lineage members. The conformity (i.e. the level of shared mutation positions and identical sequence mutations) of CH235 lineage development is analyzed relative to other V_H gene-specific bnAb lineages in other donors, as well as the co-evolution of virus and CH235 lineage. Despite an early near-optimal binding orientation, the CH235 lineage required over 20% SHM to reach 90% neutralization breadth. The results described herein provide insight into the difficulties in focusing recognition to the conserved site of HIV-1 vulnerability, and suggest that CD4bs-directed antibodies, whether V_H-gene restricted or CDR H3-dominated, face similar obstacles in development. For V_HI - 46- and V_Hl-2-derived CD4-mimic antibodies, the unique genetic mutability inherent in each of these two V_H-germline genes helps to direct maturation, potentially providing an explanation for the prevalence of effective CD4bs antibodies derived from these two germline genes.

[0308] RESULTS [0309] Sequencing of B cell antibody gene rearrangements in longitudinal samples.

[0310] To understand the maturation of the cooperating CH235 lineage in donor CH505, we sought to identify sequences of lineage members at 17 time points, spanning wks 6 to 323 from time of infection. First it was asked when we could detect members of the CH235 lineage. Next-generation sequencing (NGS) of antibody heavy chain gene rearrangements amplified from genomic DNA template of blood mononuclear cells from wk 6 to 152 (15 time points) identified a total of 479,028 unique, non-duplicated V-heavy sequences. The first V-heavy sequences belonging to the CH235 B cell lineage were found at wk 14, and additional CH235 lineage members were found at all subsequent time points. Only unique sequences in the CH235 lineage were further investigated and they were assigned to the earliest time-point (time-of-appearance) in which they were identified. Four V-heavy sequences were paired with the closest V_L from identified antibodies and produced as recombinant monoclonal antibodies (mAbs) (CH235.6 through CH235.9). From cultured memory B cells collected 41 wks post-transmission we had previously identified five members of the CH235 lineage (CH235, CH236, CH239, CH240 and CH241) (Gao et al., 2014) and we have now identified four additional members with natural V_H and V_L pairing from cultured memory B cells collected at wks 264 and 323 post-transmission: CH235.10 through CH235.13 (Figure 29A, Figure 36A and Figure 40C). CH235 lineage antibodies represented 0.018% of the total memory B cell repertoire and 0.5% of the CH505 TF gpl20- specific memory B cell population.

[0311] The CH235 lineage could be separated into three clades (clade I, II and III). Clade I showed a number of early lineage members, but no additional clade I sequences were observed after wk 30; clade II showed further development and included members CH241 (wk 41) and CH235.6 (wk 66), but no additional sequences were observed after wk 66; clade III developed through wk 323 and included antibodies CH235 (wk 41), CH235.9 (wk 152), and CH235.12 (wk 323) (Figure 29A).

[0312] CH235 lineage HIV-1 neutralization.

[0313] To characterize the development of neutralization breadth in the CH235 lineage, antibodies in clade III were assessed for their ability to neutralize diverse HIV-1 isolates in a 199-isolate panel (Figure 29B and Figure 41). No isolates were neutralized by the unmutated common ancestor (UCA), whereas 18% of the viruses were neutralized by CH235 at wk 41. By wk 152, CH235.9 neutralized 77% of viruses, although with a relatively weak potency of 3 μ /ηι1. By wk 323, however, CH235.12 was able to neutralize 90% of viruses, and the neutralization 50% inhibitory concentration (IC₅₀) potency increased by 5-fold to 0.6 μ§/ιη1.

[0314] Next the heterologous neutralization pattern of these antibodies were analyzed to understand their development of broad neutralization (Figure 36B) (Georgiev et al., 2013). CH235 lineage members and previously identified HIV-1 bnAbs were clustered based on heterologous neutralization activity. CH235 neutralization activity was more similar to CD4bs bnAbs than to bnAbs with other epitope specificities. While the CH235 neutralization profile was the most divergent from other CD4bs bnAbs, CH235.9 and CH235.12 were much more similar to other CD4bs bnAbs and each other. Interestingly, despite V_HI-46 usage, the CH235.9 and CH235.12 neutralizing profile was more similar to that of V_Hl-2-derived antibodies, such as VRCOl, than V_Hl-46-derived antibodies, such as 8ANC131 (Figure 36B).

[0315] Crystal structures of CH235-lineage members with HIV-1 gpl20.

[0316] To provide structural insight into the recognition and maturation of the CH235 lineage, the antigen-binding fragments (Fabs) of antibodies CH235 (wk 41 from time of infection, 18% breadth), CH235.9 (wk 152, 77%) and CH235.12 (wk 323, 90%), were prepared and co-crystallized, solved and refined these in complex with the gpl20 core of HIV-1 isolate strain (93TH057) (Figure 30, Figure 42). We mapped the location of residues altered during SHM and observed changes throughout the variable domain (Figure 30A).

[0317] Comparison of the orientation of the V_H of CH235 in Env binding with that of CD4, VRCOl and 8ANC131 (Scheid et al., 2011) showed that the CH235 V_H domain mimicked CD4 in Env binding and was highly similar to the V_H orientation and structure of the VRCOl and 8ANC131 V_H chains: in particular, the V_HI-46 of CH235 preserved key contacts mediated by the CDR H2 loop for the CD4 binding loop and for the gpl20 D368 (Figure 37A,B).

[0318] Analysis of the angle of recognition for the CH235 lineage indicated little change during maturation, with CH235, CH235.9 and CH235.12 all clustering within the larger VRCOl-class of antibodies. Interestingly, other V_Hl-46 antibodies clustered differently, with antibody 1B2530 from HIV-1 -positive donor RU1 at a highly similar angle and 1.5 A translated, and antibodies 8ANC131 and 8ANC134 from HIV-1 -positive donor RU8 occupying a cluster about 55 degrees and 3.5 A translated related to the CD4 (Figure 37C).

[0319] These results suggest that the gpl20-antibody orientation was determined early in bnAb lineage ontogeny, with further maturation maintaining the same general orientation. Overall, the structures of CH235 lineage members with HIV-1 gpl20 Env revealed CD4 mimicry. While the V_H gene usage classifies the CH235 lineage within the V_Hl-46-derived 8ANC131 bnAb class, it is both functionally and structurally closer to the VRCOl class (Zhou et al., 2015).

[0320] Negative stain EM of CH235-lineage members with trimeric HIV-1 Env.

[0321] To visualize the recognition of the CH235 lineage in the context of the HIV-1 Env trimer, negative stain EM was used to determine 3D-reconstructions of Fabs CH235, CH235.9 and CH235.12 bound to trimeric BG505 and B41 HIV-1 Env glycoproteins (Figure 30B)(Pugach et al., 2015; Sanders et al., 2013). Notably, the stoichiometry increased with antibody maturation, with CH235 (8% V_H mutation) binding with a stoichiometry of 1 : 1 (BG505; Figure 30B, top, Figure 37D-F) or 2: 1 (B41; Figure 30B, bottom, Figure 37D-F) Fabs per trimer and CH235.9 and CH235.12 (19% and 25% V_H mutation, respectively) binding with a 3 : 1 Fab to trimer ratio (Figure 30B). Next the orientation and stoichiometry of CH235.12 Fab was compared with that of CH103, a CDR H3-dependent CD4bs bnAb identified from the same subject (Liao et al., 2013). EM analysis of either CH235.12 or CHI 03 Fab in complex with BG505 SOSIP.664 revealed structural differences between the CDR H3-dominated CH103 class bnAb and the 8ANC131-class CH235.12 bnAb and, in accordance with crystallographic results, the angle of approach of CH235 was similar to that of VRCOl and other CD4 mimicking bnAbs (Figure 37G).

[0322] Despite the CD4 mimicry by CH235, the trimer remained in a closed conformation when the CH235 lineage members were bound. However, the EM-derived model of CHI 03 in complex with BG505 revealed that CH103 either bound to or induced a more open version of the trimer. This conformation represents an intermediate state between the closed, compact trimer in complex with CH235 or VRCOl, and the CD4-induced open model in complex with soluble CD4 or 17b Fab (Figure 37G). Similar to more mature CH235 lineage bnAb Fabs, bnAb CH103 bound to BG505 with a stoichiometry of 3 Fabs per trimer. (Figure 37G).

[0323] Maturation focuses CH235 lineage recognition to a conserved site of CD4 vulnerability.

[0324] To gain insight into the structural consequences of maturation, the epitope of CH235 lineage members was mapped relative to the conformationally invariant CD4 supersite of vulnerability (Zhou et al., 2015). When the CH235 footprint was mapped on gpl20, we observed portions of the CH235-binding surface on gpl20 to be outside of the CD4 supersite of vulnerability (Figure 30C, left). This surface was reduced in CH235.9 and CH235.12 structures, especially on variable loop V5. Recognition by the CH235.12 antibody

concentrated almost entirely on the CD4 supersite of vulnerability, with little interactions with the inner domain or variable loop V5; there was, however, a large remaining interaction with the conserved loop D region (Figure 30C, middle and right).

[0325] To quantify targeting precision, the buried surface between antibodies and gpl20 co- crystal complexes was computed, for the region overlapping the CD4 supersite of

vulnerability minus the region outside the vulnerable site. Overall targeting precision correlated with neutralization breadth ( =0.0007) (Figure 30D). The CH235-lineage antibodies all showed good targeting precision. The correlation of SHM versus neutralization breadth was also analyzed ( =0.0097) (Figure 30E): While the CH235 lineage generally trended towards lower SHM relative to neutralization breadth, all CD4bs bnAbs appeared to require a high degree of SHM, independent of whether the antibody derived from a specific VH-gene or used a CDR H3 -dominated mode of recognition.

[0326] Overall, the results suggest that maturation requires a high degree of SHM to focus recognition onto the CD4 supersite of vulnerability and that this high degree of SHM is a general requirement of all CD4bs bnAb lineages, even those that begin with highly favorable orientations such as CH235.

[0327] Conformity of sequence evolution of CH235 lineage.

[0328] The mutation levels of CH235-lineage antibodies identified 41 wks post infection from memory B cell cultures was markedly lower (range 7-11%;) than that of all previously reported V_Hl-46 and V_Hl-2 CD4bs bnAbs (>25%) (Scheid et al., 2011; Sui et al., 2010; Wu et al., 2010; Zhou et al., 2015) (Figure 40C). The mutation levels of CH235 -lineage antibodies identified up to 264 wks post infection increased to -20%, but were still lower than those of most other bnAbs until 323 wks post infection (CH235.12 : 26% mutations) (Figure 31A)

[0329] To quantify the conformity of CH235-lineage antibodies to the two V_Hl-46-derived bnAbs (1B2530 from donor RU1 and 8ANC131 from donor RU8) (Scheid et al., 2011; Zhou et al., 2015), we analyzed the similarity of shared mutation positions (positional conformity) and shared identical mutations (identity conformity) of the VH genes (Figure 31B, Figure 38A). As a comparison, the positional conformity and identity conformity of non-HIV-1 targeting antibodies identified from 3 HIV-1 negative donors relative to template antibodies 1B2530 and 8ANC131 were also calculated. Positional conformity in SHM was spread over a large range (50-90%), and there did not seem to be much discrimination between V_Hl-46 in antibodies that effectively neutralized HIV-1 and those that did not (Figure 31B, top panels). Identity conformity in SHM was also spread over a large range (0-75%) (Figure 31B, bottom panels), and while little discrimination was observed between V_H1-46 in antibodies that effectively neutralized HIV-1 and those that did not for antibody 8ANC131, there was discrimination among CD4bs antibodies when 1B2530 was used as a reference (Figure 31B, bottom left panel). The differences in CH235-lineage identify conformity to 1B2530 or to 8ANC131 may reflect the greater similarity of the recognition orientation of CH235-lineage members withlB2530 (Figure 37C) and suggested that slight differences in recognition orientation can substantially alter factors associated with identity conformity.

[0330] Overall, these results indicated SHM in response to HIV-1 infection to proceed in a manner that depended less on functional selection and more on intrinsic properties of the V_Hl-46-gene, especially related to the position of residues that undergo SHM. To investigate further the contribution of the V_Hl-46 gene, we analyzed SHM observed in V_Hl-46 gene transcripts from three uninfected individuals (Figure 31C, top); notably, all 11 positions mutated in CH235, 1B2530 and 8ANC131 were also mutated among non-HIV-1 neutralizing antibodies with high frequency (>20%). Moreover, the residue substitutions in CH235 were frequently found in the top three most commonly observed substitutions for that position in the V_H1-46 gene. To quantify the impact of gene mutability, we compared the difference in probability distributions of positional and identity conformity for sequences simulated with and without taking into account the intrinsic V_H1-46 gene mutability. The simulations showed that both positional and identity conformity shifted to a higher level of similarity when considering gene mutability (Figure 38B and 38C). Notably, a substantial shift in probability was observed for the positional conformity of CH235 (Figure 38B, Figure 43 A). Similar shifts in identity conformity were also observed for CH235 (Figure 38C, Figure 43B). Thus, the intrinsic susceptibility at specific sites of the V_Hl-46 germline gene to mutation as well as to the frequency of specific mutations that existed at each of these sites appeared to be a dominant factor in the SHM alteration of the CH235 lineage. These results are in line with our previous finding that selection and mutability synergized during affinity maturation of an influenza HA-reactive clone from a non-HIV-1 infected person to hemagglutinin (HA) (Kepler et al., 2014): hence, the dominant role of intrinsic susceptibility at specific sites may be a more general biological phenomenon in dictating the course of SHM. [0331] Because V_H1-2 is genetically the most closely related germline gene to V_H1-46, we also examined the mutability of the V_Hl-2 gene (Figure 31C, bottom). Consistent with V_H1- 46 antibodies, the mutated positions among V_H1-2 derived bnAbs also showed high frequency of mutation among non-HIV-1 targeting V_H1-2 antibodies, suggesting that gene mutability contributes to V_H1-2 derived HIV-1 antibody evolution. Notably, the average mutability of the V_Hl-2 gene at positions where the CH235 antibody showed SHM was generally high: 9 of 15 positions mutated in CH235 antibody were also mutated in more than 15% of V_Hl-2- derived NGS reads. In 10 of these 15 positions, the mature V_Hl-2-derived bnAbs (VRCOl, VRC-CH31 and VRC-PG04) also showed changes. When we analyzed mutability of other V_H genes used by CD4bs bnAbs (V_Hl-69, V_H3-23, V_H3-30, and V_H4-59) (Zhou et al., 2015) (Figure 38D), we observed gene mutability patterns different from that of V_Hl-46 while, in contrast, the mutability patterns of V_H1-2 and V_H1-46 were more similar (Figure 43C).

Despite the similarity between V_H1-2 and V_H1-46, we did observe that antibody sequences from CD4bs bnAbs of each gene segregated phylogenetically (Figure 31E), indicating differences in maturation pathway between bnAbs evolving from these two germline genes.

[0332] These data suggested that for both V_H1-2 and V_H1-46 germline genes-derived bnAbs, somatic mutations that lead to neutralization breadth appeared to be primarily determined by the intrinsic mutability of V_H1-46 and V_H1-2 germline genes. The differences in the intrinsic mutabilities of these V_H genes may contribute to the high occurrence of CD4bs bnAbs that originate from either V_H1-2 or V_H1-46 (Zhou et al., 2015).

[0333] Interaction between CH235 and CH103 bnAb lineages.

[0334] While gene mutability plays a role in determining the position where SHM occurs, binding between antibody and HIV-1 Env likely also plays a role in selecting or fixing a mutation. A hallmark of cooperating B cell lineages is that they interact at the same site as the bnAb lineage that is being driven (Gao et al., 2014).

[0335] To determine a mechanism whereby the initial interaction of the early CH235 and CHI 03 lineage members bind to the same or similar epitope and result in CH235 selection of escape mutants that stimulated the CH103 bnAb lineage (Gao et al., 2014), cross-competition between early CH235 lineage antibodies and the CH103 lineage antibody CH106 in ELISA was evaluated, as an example of early CHI 03 lineage development, and measured their association rate constant with surface plasmon resonance (SPR). Since both the CH235 and CH103 lineages bound to the loop D gpl20 region, we asked if the early CH235 lineage antibodies could block the binding of the CHI 03 lineage mature antibody CHI 06, or block the binding of soluble (s)CD4 to CH505 TF gpl20 Env. CH241 was the only antibody in the CH235 lineage that strongly blocked CH106 bnAb and sCD4 binding to CH505 gpl20 (IC₅₀ = 2.6 and 1.5 μg/ml, respectively) (Figure 44A).

[0336] To confirm early dominance of the binding of CHI 03 lineage compared to the CH235 lineage to CH505 TF Env, the blocking assay was reversed and asked if bnAb CHI 06 could block the binding of biotinylated CH235, CH236, CH239, CH240 or CH241. CH106 strongly blocked the binding of all the CH235 mature antibodies with IC50s ranging from 2.3 μg/ml (for CH240) to 14.3 ug/ml (for CH241) (Figure 44B). These data suggested that the earliest maturation intermediates of the CH235 lineage antibodies could not outcompete CHI 06 bnAb for binding to CH505 TF gpl20 Env.

[0337] Affinity maturation in germinal centers is subjected to kinetic selection and involves improvement in dissociation rate constant (K_d) that is often driven by an improvement in the kinetic association rate (k_a), which is a key variable in conferring a binding advantage for the cognate epitope to an antibody over other competing antibodies (Foote and Milstein, 1991; Kepler et al., 2014). The k_a and dissociation kinetic rate (1¾) of the CH505 TF gpl20 Env binding by CH235 and CHI 03 was measured with SPR to identify differences that might explain the relative inability of the CH235 lineage to block the binding of the CHI 03 lineage bnAbs to autologous CH505 TF Env and found that the two lineages followed two distinct trajectories and modalities to increase their overall affinity.

[0338] The UCA of the CH103 lineage bound to CH505 TF Env with a K_d of 227 nM which increased one order of magnitude throughout affinity maturation (Figure 32A). The CHI 03 UCA displayed a fast association rate (k_a = 37 x 10³ M^'V¹) which was maintained across the intermediate and mature mAbs (k_a = 11.9 - 37.3 xlO³ M^'V¹), suggesting that maintaining the fast association rate was important for survival and maturation of the CHI 03 lineage (Figure 32B). In contrast, the CH235 lineage mAb K_d increased four orders of magnitude during affinity maturation (from 30.6 mM of IA4 - the earliest intermediate mAb in the CH235 lineage for which kinetic rates could be measured - to 0.7 nM of CH241) (Figure 32C). Such increase was predominantly facilitated by slower dissociation rates (k_d) observed in later intermediates and mature mAbs, which decreased from 88.1 xlO^"3 s^"1 of IA4 to 0.33 xlO^"3 s^"1 of CH241 (Figure 32D). Conversely, CH235 lineage mAbs bound to CH505 TF gpl20 Env with k_a that started off an order of magnitude slower than CHI 03 UCA and its earlier intermediates (IA4 k_a =2.9 xlO³ M^'V¹) and only modestly improved - primarily between IA1 and CH235 mAbs - with the majority of the early CH235 mAbs having slower k_a than CHI 03 mAbs (Figure 32D).

[0339] Thus, the relative inability of wk 41 CH235 lineage antibodies to block early mature CHI 03 lineage mAbs could be explained by the observed differences in their association rates, and these data provide an explanation of how the CH235 antibody lineage exerted its cooperating function in driving autologous virus toward better neutralization by the CHI 03 antibody lineage without impeding concurrent development of the CHI 03 antibody lineage itself.

[0340] Late CH235 lineage broadly neutralizing antibodies neutralize autologous loop D escape viruses selected by early CH235 lineage members.

[0341] It has previously been demonstrated that the CH235 lineage selected escape viruses with mutations in the loop D region of gpl20 Env that rendered loop D mutant viruses more sensitive to the CHI 03 bnAb lineage and that autologous virus escaped from early CH235 lineage antibodies by wk 30 after infection (Gao et al., 2014). Described herein is the isolation of autologous viruses through wk 323 and the determination of the neutralization capacity of the late CH235 lineage bnAbs. Viruses partially sensitive to the later members of the CH235 lineage (particularly bnAbs CH235.9 and CH235.12) were found as late as wk 176 (Figure 33A, Figure 45). These viruses still contained the loop D mutations that were selected by virus escape from early antibody members of the CH235 lineage (Gao et al., 2014). Hence, the ability of the late CH235 lineage bnAbs to neutralize the panel of CH505 TF loop D mutants was tested (Gao et al., 2014). Remarkably, CH235.9, CH235.12 and CH235.13 bnAbs acquired the ability to neutralize all loop D mutants that were resistant to the early members of the CH235 lineage (Figure 33B and Figure 46). In particular, CH235.9, CH235.12 and CH235.13 neutralized CH505 TF gpl20 M8, M20 and M21 (not neutralized by early lineage member CH236), which differed from CH505 TF gpl20 M6 and M10 (neutralized by CH236) by a single mutation at position 280 (N280S for M8 and M20, and N280T for M21) (Figure 33B).

[0342] In the gpl20-complexed structure, the side chains of N280 forms three hydrogen bonds with two residues in the CDR L3 and these hydrogen bonds are predicted to be disrupted by the N280S and N280T mutations (Figure 39A). Since the CH235.9 antibody had the V_L of CH236, the direct implication was that mutations in the heavy chain were responsible for the ability of CH235.9 to neutralize loop D mutant viruses. Interestingly, CH235.7, which did not neutralize autologous viruses beyond wk 53, also had the VL of CH236 but, in contrast to CH235.9, failed to neutralize CH236-resistant loop D mutants M7, M8, M9, M20 and M21.

[0343] Therefore, we reverted the 5 amino acids (aa) in CH235.9 VH at gpl20 contact positions that were different from those present in CH236 VH but not shared with CH235.7 V_H: N30T and D3 IT in CDR HI, G62Q and G65Q in FR H3 and A103E in CDR H3 (Figure 39B). Five of the six CH235.9 mutants retained the ability to potently neutralize all the CH505 TF loop D mutant viruses. In contrast, the N30T mutation in CDR HI reverted CH235.9 to the CH236 phenotype (CH236 has a threonine in position 30): M21 neutralization was abrogated, M20, M7 and M9 were near completely abrogated (CH235.9 N30T IC₅₀ >44 μg/ml) and M8 IC₅₀ increased 37-fold (CH235.9 IC₅₀ = 0.66μg/ml vs CH235.9 N30T IC₅₀ = 24.31 μg/ml) (Figure 46).

[0344] Thus, acquisition of extraordinary breadth in the CH235 bnAb lineage was associated with accumulation of somatic mutations in CDR HI that enabled late CH235 lineage antibodies to neutralize autologous loop D mutant viruses that were escape mutants from early CH235 antibodies. CH235.9 bnAb residue N30 contacts R429 in the β20-β21 loop of the C4 region of gpl20 Env, which is on the opposite face of the CD4bs from loop D (Figure 33C). In addition, CH505 TF has a glutamic acid in position 429 that is in close enough proximity to N30 to form a hydrogen bond.

[0345] These findings indicate a mechanism for acquiring the ability to neutralize loop D mutants via a compensatory mutation in the CH235 VHDJ_h which strengthens the binding to the gpl20 C4 region by introducing hydrogen bonds that correct the loss of neutralization due to disruption of the hydrogen bonds between loop D and the CH235 mAb light chain.

[0346] CH235 and CH103 lineage antibody binding to CH505 gpl20 Env.

[0347] The CH235 lineage antibodies were tested for binding to 113 recombinant CH505 gpl20 Env isolated from time of transmission to wk 160 post-transmission, including CH505 TF loop D mutant Envs (Figure 34A and Figure 47). Of note, CH235.9 and CH235.12 bound to 4/22 and 8/22 Envs isolated from wk 136 and 160 post-transmission, respectively, including Envs from viruses that were also neutralized. Env binding to the initial members of the CH103 lineage has previously been reported (Hraber et al., 2015), and described herein is the same Env binding analysis of the CHI 03 lineage with 10 additional matured bnAb members of the CHI 03 lineage (Figure 34A and Figure 47). We have used these data to select CH505 gpl20 Env quasi-species that bound to mature and precursor bnAbs of both lineages, defining a series of CH505 Env immunogens now optimized and predicted to induce both bnAb lineages (Figure 40A).

[0348] It has previously been reported that CH235 UCA weakly reacted with CH505 TF gpl20 at ~10 μΜ as determined by SPR (Gao et al., 2014). Here we show stronger binding of the CH235 UCA to 8/113 autologous CH505 gpl20 Envs measured in ELISA (Figure 34A and Figure 47). Moreover, in a panel of 15 heterologous Envs from multiple clades, CH235 UCA bound to 3/15 Envs and the introduction of only 3 mutations (W47L, G54W and S56R), which were selected based on the increase in surface area of interaction (G54W and S56R) or the reduction in clash score (W47L), increased this recognition (to 5/15 Envs), of which the dominant effect appeared to be reduction in clash (Figure 34B and Figure 40B).

[0349] Autoreactivity in the CH235 B cell lineage.

[0350] Development of auto- and poly reactivity during antibody maturation toward neutralization breadth is a critical aspect that may limit the ability of generating bnAbs during natural infection and upon vaccination (Bonsignori et al., 2014; Haynes et al., 2005; Haynes et al., 2012; Haynes and Verkoczy, 2014; Liu et al., 2015; Verkoczy et al., 2013; Verkoczy et al., 2010; Verkoczy et al., 2011). It has previously been reported that in HIV- 1 -infected individual CH505, the CD4bs CHI 03 bnAb lineage was polyreactive and, similar to VRCOl- class bnAbs, bound to human ubiquitin ligase E3 A (UBE3 A) with avidity correlated with neutralization (Liao et al., 2013; Liu et al., 2015). In addition, most of the mutations introduced in VRC07 - a somatic variant of VRCOl - that enhanced neutralizing activity also resulted in increased autoreactivity (Rudicell et al., 2014). Since CH235.12 is a potent and extremely broad CD4-mimic CD4bs bnAb, we compared the auto- and polyreactivity profile of CH235.12 with other members of the CH235 lineage. Most CH235 lineage antibodies displayed reactivity against DNA and sporadic reactivity with Scl70 (CH235.7) (Figure 35A). CH241 bound to cardiolipin (Figure 35B). In Hep-2 IF staining CH236, CH235.7 and CH235.9 were all cytoplasmic positive (Figure 35C). Conversely, CH235.12, despite being highly mutated and broadly neutralizing, did not display autoreactivity in any of these assays (Figure 35A-C) Of particular note, CH235 lineage antibodies, including CH235.12, did not react with UBE3A (Figure 35D).

[0351] These data identify CH235.12 as an antibody that has developed neutralization breadth without being itself auto- and polyreactive, while less mutated precursor antibodies (CH235 is in the same clade of CH235.12) did develop autoreactivity. Therefore, in vivo decoupling of neutralization breadth of CD4 mimic CD4bs bnAbs from auto- and polyreactivity can occur, even for bnAb lineages that have developed autoreactivity during the course of their maturation and, therefore, inducing such bnAbs from such lineages through vaccination, though difficult, is an achievable goal.

[0352] DISCUSSION

[0353] Here we have traced the ontogeny of the CH235 V_Hl-46 8ANC131 class of CD4bs bnAbs from acute infection to chronic infection and defined both the structural and functional pathways of bnAb lineage induction. That the CH235 bnAb lineage that selected virus escape mutants that drove the CHI 03 CD4bs CDR H3 -dependent bnAb lineage is itself an

8ANC131-class bnAb lineage and co-evolved with the CH103 bnAb is a remarkable demonstration of a bnAb-to virus-to bnAb interaction in the same HIV-1 infected individual. In addition, the similarity of V_Hl-46 8ANC131-like and V_Hl-2 VRCOl family CD4 supersite bnAbs demonstrates dramatic convergence of antibody structures to recognize the CD4 supersite. The CH235 lineage required over 20% SHM in heavy chain variable domain to achieve 90% breadth. Fortunately, a substantial portion of the VH-gene SHM was guided by the intrinsic mutability of the VH1-46 germline gene. Moreover, the CH235 lineage Ab that became broadly neutralizing acquired the ability to neutralize loop D mutants selected by early Ab lineage members (Gao et al., 2014) with a mechanism involving a compensatory mutation (T30N) in CDR HI, which allowed the formation of H-bonds with the HIV-1 gpl20 C4 region, thus correcting the original loss of binding.

[0354] The driving forces of the CH235 lineage were the natural transmitted/founder and M5 Envs. In addition, despite near-complete autologous virus escape from CH235 lineage antibodies by wk 100, viruses arose later during the course of infection, which were sensitive to the more mature CH235 bnAb members and likely contributed to antigen drive. It is interesting to note that many of these late viruses were less sensitive to CHI 03 CDR H3 binder bnAbs prompting the hypothesis that the CHI 03 lineage may have the capacity for cooperation with the CH235 lineage after 5-6 years of co-development. Finally, the CH235.12 antibody that evolved late in CH235 development is an extraordinary broad and potent non- autoreactive antibody and is a candidate for preventive and therapeutic uses.

[0355] In summary, the acquisition of neutralization breadth in the CH235 VRCOl -like VH I - 46 CD4 mimic bnAb occurred with the sequence of transmitted/founder and early mutant- initiated antigen drive, selection of Env loop D mutants that cooperated with the CHI 03 bnAb lineage to drive it to bnAb breadth, followed by acquisition of the ability of the CH235 lineage itself to neutralize autologous loop D mutants coincident with potent neutralization of a broad array of heterologous HIV-1 isolates. Mapping these events points to a strategy for the simultaneous induction of both CDR H3 and VRCOl -class CD4bs bnAbs, whereby sequential immunizations with transmitted founder Env followed by loop D mutant Envs comprise a rational immunization strategy.

[0356] EXPERIMENTAL PROCEDURES

[0357] Donor and sample information. Donor and sample information was previously reported (Liao et al., 2013) and is summarized in Supplemental Experimental Procedures. Memory B cell cultures were performed on PBMCs collected at 264 and 323 wks post- transmission. All work related to human subjects was in compliance with Institutional Review Board protocols approved by the Duke University Health System Institutional Review Board.

[0358] Preparation of libraries for 454 DNA pyrosequencing. 454 DNA pyrosequencing was performed on genomic DNA template isolated with Qiagen kits from PBMCs collected at 6, 7, 8, 9, 14, 20, 22, 30, 41, 53, 66, 92, 100, 144 and 152 wks post-transmission as described in (Boyd et al., 2009) and in Supplemental Experimental Procedures. Only unique V-heavy rearrangements were included in the analysis to generate the phylogeny; in the case of duplicated sequences, the earliest occurrence was included in the analysis.

[0359] Phylogenetic analysis. For clonal phylogenetics, the UCA was inferred using

Cloanalyst (Kepler, 2013), which simultaneously estimates the UCA and the phylogenetic tree relating the observed sequences to each other and to the UCA. Internally, Cloanalyst uses dnaml from the PHYLIP suite of phylogenetic software (Felsenstein, 2005). The CH235 antibody lineage clonogram was displayed using the ete2 Python package.

[0360] Isolation of CH235 Lineage Antibodies from Donor CH505. Fluorescence- activated cell sorting of antigen-specific IgG⁺ B cells from PBMC and the amplification and cloning of immunoglobulin genes were performed as described in (Bonsignori et al., 2011). CH505.TF gpl20 Env-positive memory B cells were cultured as described in Supplemental Experimental Procedures.

[0361] Neutralization assays. Neutralization of donor CH235 mAbs were measured using single-round-of-infection HIV-1 Env pseudoviruses and TZM-bl target cells as described in Supplemental Experimental Procedures. [0362] Neutralization signature. Antibody neutralization signatures were computed and compared as described in Supplemental Experimental Procedures.

[0363] Monoclonal Antibody and Antigen-Binding Fragment (Fab) Production. Ig genes of mAbs were amplified from RNA and expression plasmids for heavy and kappa chains were constructed. Expression and purification of recombinant IgG mAbs and preparation of Fab fragments are described in Supplemental Experimental Procedures.

[0364] Crystallization, X-Ray Data Collection, Structure Determination, and

Refinement of Donor CH235 Antibodies in Complex with HIV-1 gpl20. Purification, crystallization of antibody-gpl20 complexes, data collection, structure solution, refinement and analysis are described in Supplemental Experimental Procedures. Diffraction data were integrated and scaled with the HKL2000 suite (Otwinowski and Minor, 1997).

[0365] Electron microscopy data collection and processing. BG505 SOSIP.664 and B41 SOSIP.664 gpl40 trimers and donor CH235-derived Fab complex negative- stain electron microscopy images, analysis and visualization are described in the Supplemental

Experimental Procedures.

[0366] Focused maturation and conformity analysis. Focused maturation and mAb conformity analysis are described in the Supplemental Experimental Procedures.

[0367] Surface Plasmon Resonance Affinity and Kinetics Measurements. MAb binding to autologous CH505 gpl40 was measured using a BIAcore 3000 or BIAcore T200 instrument (GE Healthcare) as described in (Alam et al., 2007; Alam et al., 2009; Liao et al., 2013) and in Supplemental Experimental Procedures.

[0368] Direct-Binding ELISA. Direct-binding ELISAs were performed as described in Supplemental Experimental Procedures.

[0369] MAb CH235.9 Amino Acid Reversion. Site-directed mutagenesis of the CH235.9 mAb genes was performed using the Quikchange lightning multi-site-directed mutagenesis kit (Agilent) following manufacturer's protocol. Primers are listed in Supplemental Experimental Procedures.

[0370] Structural Modeling. Loop D mutations were structurally modeled using PyMOL with sidechains placed in the most frequently observed rotamer that did not result in steric clashing with neighboring residues. Hydrogen bonds were calculated using HBPLUS software (McDonald and Thornton, 1994). [0371] Recombinant HIV-1 Proteins. HIV-1 genes of autologous CH505 Env were determined from samples collected from 4 to 323 wks post-infection by single genome amplification (Keele et al., 2008) and produced as described in (Liao et al., 2013).

[0372] Protein Array. MAbs were screened for binding on protein microarrays (ProtoArray) (PAH0525101; Invitrogen) pre-coated with 9,400 human proteins in duplicate and screened following manufacturer's instructions and as described in (Liu et al., 2015; Yang et al., 2013).

[0373] HEp-2 cell staining. Indirect immunofluorescence binding of mAbs or plasma to HEp-2 cells (Zeuss Scientific) was performed as previously described (Bonsignori et al., 2014; Haynes et al., 2005).

[0374] Supplemental Experimental Procedures

[0375] Donor and Sample Information.

[0376] The CH505 donor, from which the CHI 03 and the CH235 antibody lineages were identified, is an African male enrolled in the CHAVIOOl acute HIV-1 infection cohort (Tomaras et al., 2008) and followed for over 6 years. During this time viral load ranged from 14,460 to 847,279 copies/ml (median = 173,667 copies/ml), and CD4 counts ranged from 69 to 431 cells/mm³ (median = 294 cells/ mm³).

[0377] The time of infection was estimated by analyzing the sequence diversity in the first available sample using the Poisson Fitter tool (Giorgi et al., 2010) as described in (Liao et al., Nature 2013). Results were consistent with a single founder virus establishing the infection and with the earliest isolated virus sequences being taken 4 weeks post-transmission.

[0378] Flow Cytometry, Memory B Cell Cultures and mAb Isolation.

[0379] The HIV-1 CH505.TF gpl20 envelope glycoprotein was produced and used in flow cytometry on PBMC collected from donor CH505 at week 264 and 323 post-transmission using a two-color technique as described (Gray et al., 2011).

[0380] CH505.TF gpl20 Env-positive memory B cells were cultured as described

(Bonsignori et al., 2011) with the following modifications: non-irradiated MS40L cells were used as feeder cells at a concentration of 3,000 cells/well and were added to wells in which memory B cells were sorted in bulk; 50 ng/ml of recombinant human (rHu) IL-21 (200-21; Peprotech, Rocky Hill, NJ) were added to the complete medium; memory B cells were distributed by limiting dilution at a calculated concentration of 2 cells/well; culture medium was refreshed every 5 days. [0381] Cell culture supernatants were screened for neutralization of autologous CH505.TF virus using the tzm-bl neutralization assay (Bonsignori et al., 2011; Montefiori, 2005) and for binding to CH505.TF gpl20 Env, CH505.TF Δ371Ι gpl20 Env mutant, HIV-1 Env resurface core protein 3 (RSC3) and RSC3 Δ371Ι (Wu et al., 2010).

[0382] MAbs CH235.10 through CH235.13 were identified from cultures that displayed differential binding of CH505.TF and CH505 TF Δ371Ι gpl20 Env, did not bind to RSC3 (Gao et al., 2014) and neutralized 13 to 99% CH505.TF infectivity.

[0383] CH235 lineage antibody frequency over total memory B cells was calculated by dividing the number of CH235 lineage antibodies identified at week 41 (n = 5; Gao et al., 2014) for the number of memory B cells analyzed (n = 27,950). CH235 lineage antibody frequency over CH505.TF gpl20 Env-specific memory B cells was calculated by dividing the number of CH235 lineage antibodies identified at weeks 264 and 323 (n = 4) for the number of CH505.Env gpl20-specific memory B cells analyzed (n = 794).

[0384] 454 Pyrosequencing of CH235 lineage heavy chains.

[0385] Antibody heavy chain gene rearrangements were PCR amplified from 6 independent lOOng genomic DNA aliquots to generate 6 barcoded libraries per sample. Multiplexed primers complementary to the IGHV FR1 or FR2 framework regions, and an IGHJ-primer were modified from the BIOMED-2 consortium primers (Boyd et al., 2009; van Dongen et al., 2003). 10-nucleotide 'barcode' sequences in the primer sets encoded sample identity and replicate library identity. AmpliTaq Gold (Roche) enzyme was used for PCR following the manufacturer's instructions, with a thermocycler program: 94°C 5 min; 35 cycles of (94°C 30 sec, 60°C 45 sec, 72°C 90 sec); and final extension at 72°C for 10 min. Following

quantitation, PCR products from each replicate library were pooled in equimolar amounts, then the pooled library was run on a 1.5% agarose gel and gel extracted (Qiagen). High- throughput sequencing was performed on the 454 (Roche) platform using Titanium chemistry.

[0386] Antibody production.

[0387] Immunoglobulin genes of mAbs CH235.10 through CH235.13 were amplified from RNA from isolated cells, expression cassettes made, and mAbs expressed as described (Gao et al. 2014). The V_H genes of mAbs CH235.6 through CH235.9 were retrieved from sequences obtained through genomic DNA 454 sequencing, which were restored to full length and complemented with the V_L of the phylogenetically closest identified antibody in the CH235 lineage (i.e. CH241 for CH235.6 and CH235.8, and CH236 for CH235.7 and

CH235.9). We have previously described the isolation of mAbs CH235, CH236, CH239, CH240 and CH241and the inference of unmutated common ancestor (UCA) and intermediate antibodies IA1 through IA4 (Gao et al., 2014; Kepler, 2013).

[0388] Heavy chain plasmids were co-transfected with appropriate light chain plasmids at an equal ratio in Expi 293 cells using either 293Fectin or ExpiFectamine 293 transfection reagents (Thermo Fisher Scientific) according to the manufacturer's protocols. Cultures were supplemented with AbBooster antibody expression enhancer media (ABI Scientific) at 10% of the final culture volume 24 h after transfection. . Cultures were then incubated at 33°C for 5 more days, and supernatants were harvested and passed over a protein A affinity column. Following PBS wash and low pH elution, the pH of eluate was neutralized with lMTris pH 8.5 and samples were dialyzed against PBS. Antibodies were then aliquoted and stored at - 80°C prior to use. Alternatively, for ExpiFectamine transfections we used the enhancer provided with the kit, transfected cultures were incubated at 37°C 8% C0₂ for 2-6 days, harvested, concentrated and incubated overnight with Protein A beads at 4°C on a rotating shaker before loading the bead mixture in columns for purification; following PBS/NaCl wash, eluate was neutralized with trizma hydrochloride and antibody concentration was determined by Nanodrop. Purified antibodies were tested in SDS-Page Coomassie and western blots, and stored at 4°C.

[0389] Direct-binding ELISA.

[0390] Direct-binding ELISAs were performed as described previously (Bonsignori et al., 2011) with the following modifications: plates were blocked for 1 h at room temperature (RT) or overnight at 4°C (both procedures were previously validated); primary purified antibodies were tested at a starting concentrations of 100μg/ml, serially three-fold diluted and incubated for 1 h at RT; HRP-conjugated human IgG antibody was added at optimized concentration of 1 :30,000 in assay diluent for 1 hour and developed using TMB substrate; plates were read at 450 nm in a SpectraMax 384 PLUS reader (Molecular Devices, Sunnyvale, CA); results are reported as logarithm area under the curve (LogAUC) unless otherwise noted.

[0391] For cell culture supernatant screening of RSC3 and RSC3 Δ371Ι HIV-1 Env core proteins reactivity, plates were coated with streptavidin ^g/ml); blocked plates were stored at -20°C until used; 10 μΐ/well of biotinylated avi -tagged RSC3 and RSC3 Δ371Ι were added at 2 μg/ml for 30 minutes at RT and culture supernatants were added at 1 :3 dilution in assay diluent; plates were developed for 10 min using SureBlue Reserve TMB (53-00-03; KPL, Gaithersburg, MD) equilibrated at RT. [0392] Competition ELISAs were performed using ΙΟμΙ of primary purified monoclonal antibody, starting at 10C^g/ml and diluted in a two-fold concentration, incubated for 1 h at RT; for CD4 binding site blocking assays, 10 μΐ of a saturating concentration soluble CD4 (Progenies Pharm Inc.) was added following antibody incubation step. Ten μΐ of biotinylated target Mab was added at the EC50 determined by a direct binding of biotinylated-Mab for one hour at RT. After background subtractions, percent inhibition was calculated as follows: 100- (sera triplicate mean/no inhibition control mean)* 100.

[0393] Autoimmune purified antigens histones (whole), Jo-1, RNP/Sm, Scl-70, Sm, SSA (Ro), SSB (all from Immuno Vision) and centromere B (Prospec) were coated at optimal concentrations determined by lot-specific checkerboard with positive controls. All plasma antibody positive controls were purchased from Immuno Vision; lot-specific optimal ranges for standard curves were determined. All antibodies were tested using the same lots for each antigen and positive controls with the protocol described above. For DNA ELISA, plates were coated with 2μg/ml poly-lysine (Sigma- Aldrich) for 2 h at RT, washed 3X with PBS and blocked with PBS/2%BSA/0.05% Tween-20 for 2h at RT. After 3X wash, DNA (LS002195, Worthington) in saline sodium citrate buffer was added for 1 h, washed and antibodies were incubated for 1 h. Secondary antibody was diluted in PBS/0.05% Tween-20. Plates were developed for 30 min. Human recombinant monoclonal antibody Ab008391 (courteously provided by David Easterhoff, Duke Human Vaccine Institute) was used as positive control. For all autoantigen ELISAs, palivizumab was used as negative control. For each antibody, LogAUC was calculated and data are presented semi-quantitatively: no binding = logAUCAb < 2X negative control logAUC_neg ctri; to quantify antibody binding we divided (logAUC_pos ctri - 2X logAUC_{neg ctr}i) in tertiles and expressed test antibody binding as weak (+), intermediate (+ +) or strong (+ + +) if logAUCAb was in the first, second or higher tertile, respectively.

[0394] Anti-cardiolipin ELISA was performed using the QUANTA Lite ACA IgG III kit (708625; INOVA Diagnostics) following manufacturer's protocol.

[0395] Assessment of virus neutralization using a large panel and calculation of neutralization dendrograms.

[0396] Neutralizing antibody assays in TZM-bl cells were performed as described previously (Montefiori, 2005). Neutralization breadth of CH235 UCA, CH235, CH235.9 and CH235.12 neutralization breadth was assessed using the 384-well plate declination of the assay using an updated panel of 199 geographically and genetically diverse Env-pseudoviruses representing the major circulating genetic subtypes and recombinant forms as described (Seaman et al., 2010; Wu et al., 2010). The data were calculated as a reduction in luminescence units compared with control wells, and reported as IC₅₀ in μg/ml (Montefiori, 2005).

[0397] Dendrograms were calculated using the neighbor-joining method, showing the protein sequence distance from the HIV-1 Env gpl60 sequences of 190 HIV-1 primary isolates. The clades of HIV-1, including circulating recombinant forms (CRTs) are indicated.

[0398] Antibody neutralization fingerprinting analysis.

[0399] Neutralization fingerprints were computed and compared for CH235, CH235.9 and CH235.12 from the CH235 lineage, other CD4-binding-site antibodies, and antibodies targeting other sites of vulnerability on HIV-1 Env. The fingerprints were computed over a common panel of 165 HIV-1 strains with neutralization data for all antibodies, and a hierarchical clustering procedure was applied for building the tree, as described in (Georgiev et al., 2013). Briefly, for each antibody, the neutralization data for the common set of 165 HIV-1 strains formed that antibody's neutralization fingerprint. The Spearman correlation coefficients for all pairs of antibody neutralization fingerprints were then computed, transforming the antibody -virus neutralization matrix into an antibody-antibody correlation matrix. This correlation matrix was then input into a hierarchical clustering procedure as a way to visualize the similarities between the neutralization fingerprints for the different antibodies. The distances in the resulting tree are thus a function of the differences between fingerprints.

[0400] VHl-46 and VHl-2 antibody dendrogram calculation.

[0401] Phylogenic trees for multiple antibodies derived from VHl-46 and VHl-2 heavy chain variable genes were calculated using the neighbor-joining method. The sequences are aligned by Clustal Omega, calculated using ClustalW2. Dendrograms were drawn in Figtree.

[0402] Production and purification of HIV-1 Env protein complexed to antigen-binding fragments.

[0403] HIV-1 gpl20 protein from clade AE 93TH057 and antibodies of CH235, CH235.9 and CH235.12 were produced and purified as described previously (Zhou et al., 2010). Fab fragments of antibodies were prepared by digesting purified IgG with Lys-C at 37°C for 2-4 h. The digestion reaction was quenched by the addition of cOmplete protase inhibitors (Roche). The digested antibodies were passed over Protein A agarose to remove the Fc fragment. The Fab was further purified over a Superdex 200 gel filtration column and concentrated aliquots were stored at -80 °C. [0404] X-ray crystallography.

[0405] The gpl20-antibody complexes were formed by mixing deglycosylated gpl20 with the antibody Fab in a 1 : 1.5 molar ratio. The complexes were purified by size exclusion chromatography (Hiload 26/60 Superdex S200 prep grade; GE Healthcare) with buffer containing 0.35 M NaCl, 2.5 mM Tris (pH 7.0), and 0.02% NaN₃. Fractions with gpl20- antibody complexes were concentrated to -10 mg/ml and used for crystallization experiments. All gpl20-Fab complexes were screened against 576 crystallization conditions using a Cartesian Honeybee crystallization robot. Initial crystals were grown by the vapor diffusion method in sitting drops at 20 °C by mixing 0.2 μΐ of protein complex with 0.2 μΐ of reservoir solution. Crystals were manually reproduced in hanging drops by mixing 0.50 μΐ protein complex solution with 0.5 μΐ reservoir solution.

[0406] The 93TH057 core_e gpl20-CH235 complex was crystallized with a reservoir solution of 25% (w/v) of PEG2000, 0.2 M of Li₂S0₄, 0.1 M of Tris-HCl pH 8.5 and 5% (v/v) of isopropanol and was flash frozen in liquid nitrogen in mother liquor supplemented with 15%> of 2R,3R-butanediol as a cryoprotectant. The 93TH057 core_e gpl20-CH235.9 complex was crystallized with a reservoir solution of 9% (w/v) of PEG8000, 19% (w/v) of PEG400, 0.1 M HEPES pH 7.5 and was flash frozen in mother liquor supplemented with an additional 15%> PEG 400 as a cryoprotectant. The 93TH057 core_e gpl20-CH235.12 complex was crystallized with a reservoir solution of 10% PEG 8000, 20% PEG 400 and 100 mM HEPES, pH7.5 and was flash frozen in mother liquor supplemented with an additional 15 - 20% PEG 400 as a cryoprotectant.

[0407] Data for all crystals were collected at a wavelength of l .OOA at SER-CAT beamlines ID-22 and BM-22 (Advanced Photon Source, Argonne National Laboratory). All diffraction data were processed with the HKL2000 suite, structures were solved by molecular

replacement using PHASER, and iterative model building and refinement were performed in COOT and PHENIX, respectively. For 93TH057core_e complexes with CH235.9 and

CH235.12, molecular replacement solutions were obtained using EAF31403.1-CH235 complex as a search model.

[0408] Throughout the refinement processes, a cross validation (R_free) test set consisting of 5% of the data was used and hydrogen atoms were included in the refinement model.

Structure validations were performed periodically during the model building/refinement process with MolProbity. The 93TH057 core_e-CH235 structure was refined to a final R_free value of 22.9% with 96% residues in the favored region of the Ramachandran plot, and 0.1% outliers. The 93TH057 core_e-CH235.9 structure was refined to a final R_free value of 22% with 97.1%) residues in the favored region of the Ramachandran plot, and 0% outliers. The

93TH057 core_e-CH235.12 structure was refined to a final R_free value of 23% with 97.0% residues in the favored region of the Ramachandran plot, and 0.1%> outliers. All figures containing representations of the protein crystal structures were made with PyMOL. Gpl20 and antibody interactions were analyzed with the PISA server.

[0409] Surface Plasmon Resonance Affinity and Kinetics Measurements.

[0410] For kinetic measurement, each antibody was captured on an anti-human IgFc immobilized sensor surface (200-500RU) and gpl20 proteins at varying concentrations were injected to monitor association and dissociation phases. Buffer reference and non-specific binding to a control antibody (palivizumab) captured surface were used to derive specific binding signals. Kinetic rate constants and dissociation constant (Kd) were derived from global curve fitting analysis using a Langmuir 1 : 1 interaction model using the BIAevaluation 4.1 software (GE Healthcare).

[0411] Electron microscopy data collection and processing.

[0412] BG505 SOSIP.664 and B41 SOSIP.664 gpl40 trimers were expressed in HEK293F cells and purified by 2G12-affinity and gel filtration chromatography as described elsewhere (Pugach et al., 2015; Sanders et al., 2013). Trimers were incubated with a 10 molar excess of Fab (CH235, CH235.9, or CH235.12) overnight at room temperature and the complexes were diluted to -0.03 mg/mL prior to application onto a carbon-coated 400 Cu mesh grid that had been glow discharged at 20 mA for 30 s. The grids were stained with 2% (w/v) uranyl formate for 60 seconds. Samples were imaged using a FEI Tecnai T12 electron microscope operating at 120 keV, with an electron dose of - 25 eVA² and a magnification of 52,000x that resulted in a pixel size of 2.05 A at the specimen plane. Images were acquired with Leginon (Suloway et al., 2005) using a Tietz TemCam-F416 CMOS camera with a nominal defocus range of 1000- 1500 nm. Automated particle picking, stack creation, and initial 2D classification was performed in the Appion software suite (Lander et al., 2009). Noise and junk particles were discarded and the remaining stack was subjected to 3D classification using Relion (Scheres, 2012) with an EM volume created from the x-ray structure of ligand-free BG505 SOSIP.664 (PDB: 4zmj) low pass filtered to 60 A as the reference model. While both CH235.9 and CH235.12 bound to either B41 or BG505 at predominantly full stoichiometry (3 Fabs per trimer), CH235 bound to either trimer at sub-stoichiometric ratios (1 Fab per BG505 trimer and 2 Fabs per B41 trimer). The 3D classes representing the predominant stoichiometry for each complex were used as the initial models (low pass filtered to 40 A) for further refinement using Relion, with C3 symmetry imposed for complexes with CH235.9 or CH235.12. The total number of particles used in refinement and final resolution of the map using a Fourier shell correlation of 0.5 are as follows: BG505 in complex with CH235 - 3,467 particles (-25 A); B41 in complex with CH235 - 4,248 particles (-24 A); BG505 in complex with CH235.9 - 2,567 particles (25 A); B41 in complex with CH235.9 - 8,061 particles (19 A); BG505 in complex with CH235.12 - 15,565 particles (17 A); B41 in complex with CH235.12 - 17,023 particles (16 A).

[0413] To create figures of each Fab in complex with a representative trimer, the 3D reconstructions for each complex were fit into an EM volume created from the x-ray structure of unliganded BG505 SOSIP.664 (PDB: 4ZMJ) low pass filtered to 30 A in UCSF Chimera (Pettersen et al., 2004) and using the "segment map" option to isolate the density of the Fab components alone. Two-dimensional back projections of the final 3D models were generated using EMAN (Tang et al., 2007).

[0414] Epitope visualization.

[0415] The HIV-1 gpl20 epitopes targeted by donor CH235 antibodies were visualized using PyMOL (Schrodinger, 2010). In this graphic program, we used 5.5-A distance for selection of epitope atom sets which were virtually identical to those defined by protein interface analysis program PISA.

[0416] Monoclonal antibody CH235.9 amino acid reversion.

[0417] Primers were designed with the online Agilent Quikchange primer designer tool (www.thermofisher.com) and were as follows (SEQ ID NOs: 184-191, in order of appearance):

CH235.9N3OT: CGTGGCGTCTGGATACAACTTCACCGACTACTATATAC;

CH235.9D3 IT: CGTCTGGATACAACTTCAACACCTACTATATACACTGGGTGC;

CH235 9_G62Q: GGTCGCAC AGATTACGC AC AGGCGTTTGGGGA;

CH235 9_G65Q: GATTACGCAGGGGCGTTTCAGGACAGAGTGTCCA;

CH235 9AIO3E: GTTAGAAATGTGGGAACGGAGGGC AGCTTGCTCCACTATG;

CH235 9G62Q/G65Q: GGTCGCAC AGATTACGC AC AGGCGTTTCAGGACAGAGTGTCCA;

CH235 9_S54R: GGATCGACCCTAGGGGTGGTCGC AC AG;

CH235.9A6IS: GTGGTCGCACAGATTACTCAGGGGCGTTTG. [0418] Presence of mutations in plasmid products was confirmed by single-colony sequencing.

[0419] Structural bioinformatics.

[0420] Average buried surface area (BSA) on gpl20 was calculated for residues with BSA > 1 A² for the gpl20-antibody complexes, and the corresponding antibody neutralization potencies were averaged for each of those residues based on data from neutralization assays. Spearman correlation between BSA on gpl20 and antibody potencies was calculated for BSA cutoffs = 0 to 85 A² and potency loglCso cutoffs = 0.60 to 1.62 μg/ml.

[0421] Sample preparation for 5' RACE method and 454 pyrosequencing.

[0422] Human PBMCs (6 l0⁷) were obtained from three HIV-1 and hepatitis C negative individuals (LP32647, LP08248 and LP23810). A 5 'RACE approach was developed to amplify immunoglobulin genes based on previously described methods (Venturi et al., 2011). Briefly, the PBMCs were pelleted at 1200 rpm for 8 min. mRNA was then extracted and eluted in 50 μΐ elution buffer using μΜΑ08 mRNA isolation kit (Miltenyi Biotec) according to manufacturer's instructions. To synthesize cDNA, 10 μΐ mRNA was mixed with 1 μΐ 5'CDS Oligo dT primers (12 μΜ) and incubated at 70 °C for 1 min and then -20 °C for 1 min. Then 1 μΐ SMARTER Oligo Primer (12 μΜ) (Clontech), 4 μΐ 5X RT buffer, 1 μΐ DTT 20 (20mM), 1 μΐ dNTP (lOmM), 1 μΐ RNAse out and 1 μΐ Superscript II reverse transcriptase (Invitrogen) were added to the reaction. After 2 hours incubation at 42 °C, the cDNA products were purified using Nucleospin II kit (Macherey-Nagel) and eluted in 50 μΐ water. 454 pyrosequencing was performed as described previously (Wu et al., 2011). The first PCR amplification was performed with a common 5' primer II A (Clontech) and an Ig gene specific 3' primer (5 ' GGGGAAGACCGATGGGCCCTTGGTGG3 ' ) (SEQ ID NO: 192) using KAPA HIFI qPCR kit (Kapa Biosystems). The PCR products were purified with 2% Size Select Clonewell E-gel (Invitrogen) and Agencourt AMPure XP beads (Beckman Coulter). The second PCR amplification was performed with primers with 454 sequencing adapters (454-RACE-F:

5'CCATCTCATCCCTGCGTGTCTCCGACTCAGAAGCAGTGGTATCAACGCAGAGT3' (SEQ ID NO: 193); 454-IgG-R:

5'CCTATCCCCTGTGTGCCTTGGCAGTCTCAGGGGGAAGACCGATGGGCCCTTGGT GG3' (SEQ ID NO: 194)). The PCR products were again purified with 2% Size Select Clonewell E-gel and Agencourt AMPure XP beads. [0423] Germline V gene specific substitution profile.

[0424] The raw reads from three healthy donors shorter than 300 nucleotides or longer than 600 nucleotides in length were not analyzed. Germline V gene was then assigned to each read using an in-house bioinformatics pipeline (Wu et al., 2015). We removed reads containing stop codons. Functional reads were then clustered using Usearch at 97% sequence identity, and one unique sequence was selected from each cluster to derive a curated dataset. To further reduce reads containing sequencing errors in the curated dataset, unique sequences having only one read in the clustering step were excluded. Finally, the curated dataset of the three donors were pooled for substitution frequency analyses.

[0425] Reads from the curated dataset that were assigned to germline V genes of interest were extracted, and were aligned using MUSCLE (Edgar, 2004). The amino acid substitution frequency or mutability of a V gene position was calculated by counting how many reads contain amino acids that are different from the germline V gene, and normalized by the total number of reads. We further calculated the frequency of the 19 types of amino acid substitutions at a position, which was used to generate positional substitution logo. The similarity of positional substitution frequency profiles between V genes of interest was measured by Pearson correlation coefficient.

[0426] Conformity analysis.

[0427] The positional conformity of a conforming antibody sequence A to a reference sequence B is defined as the number of mutated positions shared by both sequences divided by the total number of mutations in the conforming sequence. Thus:

[0428] c_p{A

[0429] where , represents the set of amino acid positions in sequence i which are mutated from the germline V residue. Insertions and missing data are ignored, but deletions relative to the germline V are counted as mutations. For 8ANC131 and CH235 (Figure 38A):

M_8ANCi₃i = {2, 9, 10, 11, 16, 19, 20, 23, 26, 30, 31, 32, 33, 34, 37, 45, 46, 48, 50, 52, 53, 55, 57, 58, 59, 60, 62, 63, 66, 68, 69, 70, 71, 74, 77, 80, 84, 85, 88, 89}

M_CH235 = { 19, 23, 31, 34, 46, 47, 50, 52, 55, 57, 59, 63, 68, 83, 84}

M_8ANCi₃i P M_CH235 = { 19, 23, 31, 34, 46, 50, 52, 55, 57, 59, 63, 68, 84}

C 8ANC131, CH235j = 13 / 15 = 86.7% [0430] Identity conformity was defined the number of positionally conforming sites in conforming antibody A which were also mutated to the same residue as in the reference antibody B. Thus:

where δ is the Kronecker delta function and A_x is the identity of the residue at position x of sequence d. For 8ANC131 and CH235 (Figure 38B): c 8ANC131, CH235j = 4 / 15 = 26.7%

[0431] Targeting precision of CD4bs-directed antibodies.

[0432] The targeting precision of the CD4bs-directed antibodies was defined as the buried surface area inside of the CD4 binding site minus the buried surface area outside of the CD4 binding site. The buried surface area of each antigen residue was determined by

NACCESS. The buried surface area from the following residue numbers were considered inside of the CD4 binding site: 257, 279, 280, 281, 282, 283, 365, 366, 367, 368, 370, 371, 455, 456, 457, 458, 459, 460, 469, 472, 473, 474, 475, 476, and 477 (Zhou et al., 2007). The buried surface areas from the rest of the residues were considered outside of the CD4 binding site. Somatic hypermutation was defined using nucleotide sequences and P values were calculated based on linear regression.

[0433] Antibody binding orientation calculation.

[0434] To calculate the relative rotation angles and translation to gpl20-bound CD4 for gpl20-bound CD4-binding site antibodies, all antibody-gpl20 complexes to be analyzed were first superposed over the outer domain of gpl20 (residue ranges: 252-392, 412-422, 437-476) with gpl20 in its CD4 complex (PDB ID: 2NXY). The calculations of rotation angles and translation were then carried out with the gpl20-aligned structures. For comparison of position of heavy chain variable domain relative to gpl20-bound CD4, the frame work regions (residues 46-52, 56-59, 66-71 and 76-82) were superimposed to regions of CD4 domain 1 (residues 34-40, 43-46, 54-59, 65-71). The superposition procedures were performed with the Superpose Molecules module in CCP4 (Collaborative Computational Project, 1994). The Chi angle and distance between centroids in the Superpose output was taken as the rotation angle and translation distance between CD4 and a CD4-binding site antibody.

[0435] Supplemental References

[0436] Collaborative Computational Project (1994). The CCP4 suite: programs for protein crystallography. Acta Crystallogr., Sect D: Biol. Crystallogr. 50, 760-763. [0437] Edgar, R.C. (2004). MUSCLE: multiple sequence alignment with high accuracy and high throughput. Nucleic Acids Res. 32, 1792-1797.

[0438] Giorgi, E.E., Funkhouser, B., Athreya, G., Perelson, A.S., Korber, B.T., Bhattacharya, T. (2010). Estimating time since infection in early homogeneous HIV-1 samples using a poisson model. BMC Bioinformatics 11, 532.

[0439] Gray, E.S., Moody, M.A., Wibmer, C.K., Chen, X., Marshall, D., Amos, J., Moore, P.L., Foulger, A., Yu, J.S., Lambson, B., et al. (2011). Isolation of a monoclonal antibody that targets the alpha-2 helix of gpl20 and represents the initial autologous neutralizing-antibody response in an HIV-1 subtype C-infected individual. J. Virol. 85, 7719-7729.

[0440] Lander, G.C., Stagg, S.M., Voss, N.R., Cheng, A., Fellmann, D., Pulokas, J.,

Yoshioka, C, Irving, C, Mulder, A., Lau, P.W., et al. (2009). Appion: an integrated, database-driven pipeline to facilitate EM image processing. J. Struct. Biol. 166, 95-102.

[0441] Montefiori, D.C. (2005). Evaluating neutralizing antibodies against HIV, SIV, and SHIV in luciferase reporter gene assays. Current protocols in immunology / edited by John E Coligan [et al] Chapter 12, Unit 12 11.

[0442] Pettersen, E.F., Goddard, T.D., Huang, C.C., Couch, G.S., Greenblatt, D.M., Meng, E.C., and Ferrin, T.E. (2004). UCSF Chimera—a visualization system for exploratory research and analysis. J. Comput. Chem. 25, 1605-1612.

[0443] Scheres, S.H. (2012). A Bayesian view on cryo-EM structure determination. J. Mol. Biol. 415, 406-418.

[0444] Seaman, M.S., Janes, H., Hawkins, N., Grandpre, L.E., Devoy, C, Giri, A., Coffey, R.T., Harris, L., Wood, B., Daniels, M.G., et al. (2010). Tiered categorization of a diverse panel of HIV-1 Env pseudoviruses for assessment of neutralizing antibodies. J. Virol. 84, 1439-1452.

[0445] Schrodinger, LLC (2010). The PyMOL Molecular Graphics System, Version 1.3rl .

[0446] Suloway, C, Pulokas, J., Fellmann, D., Cheng, A., Guerra, F., Quispe, J., Stagg, S., Potter, C.S., and Carragher, B. (2005). Automated molecular microscopy: the new Leginon system. J. Struct. Biol. 151, 41-60.

[0447] Tang, G, Peng, L., Baldwin, P.R., Mann, D.S., Jiang, W., Rees, L, and Ludtke, S.J. (2007). EMAN2: an extensible image processing suite for electron microscopy. J. Struct. Biol. 157, 38-46.

[0448] Tomaras, G,D„ Yates, N.L., Liu, P., Qin, L., Fouda, G.G., Chavez, L.L., Decamp, A.C., Parks, R.J., Ashley, V.C., Lucas, J.T., et al. (2008). Initial B-cell responses to transmitted human immunodeficiency virus type 1 : virion-binding immunoglobulin M (IgM) and IgG antibodies followed by plasma anti-gp41 antibodies with ineffective control of initial viremia. J. Virol. 82, 12449-12463

[0449] van Dongen, J. J., Langerak, A.W., Bruggemann, M., Evans, P. A., Hummel, M., Lavender, F.L., Delabesse, E., Davi, F., Schuuring, E., Garcia-Sanz, R., et al. (2003). Design and standardization of PCR primers and protocols for detection of clonal immunoglobulin and T-cell receptor gene recombinations in suspect lymphoproliferations: report of the BIOMED- 2 Concerted Action BMH4-CT98-3936. Leukemia 17, 2257-2317.

[0450] Venturi, V., Quigley, M.F., Greenaway, H.Y., Ng, P.C., Ende, Z.S., Mcintosh, T., Asher, T.E., Almeida, J.R., Levy, S., Price, D.A., et al. (2011). A mechanism for TCR sharing between T cell subsets and individuals revealed by pyrosequencing. J Immunol. 186, 4285- 4294.

[0451] Wu, X., Yang, Z.Y., Li, Y., Hogerkorp, CM., Schief, W.R., Seaman, M.S., Zhou, T.,

Schmidt, S.D., Wu, L., Xu, L., et al. (2010). Rational design of envelope identifies broadly neutralizing human monoclonal antibodies to HIV-1. Science 329, 856-861.

[0452] Wu, X., Zhang, Z., Schramm, C.A., Joyce, M.G., Kwon, Y.D., Zhou, T., Sheng, Z.,

Zhang, B., O'Dell, S., McKee, K., et al. (2015). Maturation and Diversity of the VRC01-

Antibody Lineage over 15 Years of Chronic HIV-1 Infection. Cell 161, 470-485.

[0453] Zhou, T., Xu, L., Dey, B., Hessell, A.J., Van Ryk, D., Xiang, S.H., Yang, X., Zhang,

M.Y., Zwick, M.B., Arthos, J., et al. (2007). Structural definition of a conserved

neutralization epitope on HIV-1 gpl20. Nature 445, 732-737.

[0454] ACCESSION NUMBERS

[0455] Coordinates and structure factors for CH235, CH235.9 and CH235.12 in complex with

HIV-1 gpl20 have been deposited with the Protein Data Bank (PDB ID 5F9W, 5F90 and

5F96). Next-generation sequencing data have been deposited with the NCBI Sequence Reads

Archive (SRP067168). Antibody heavy and light chains have been deposited with GenBank

(KU570032-KU570053).

[0456] Antibodies Names correlation

[0457] Various antibodies names are used throughout the application. Antibodies names correlation is as follows: CH490=CH235.6; CH491=CH235.7; CH492=CH235.8;

CH493=CH235.9; CH555=CH235.10; CH556=CH235.11; CH557=CH235.12.

[0458] REFERENCES FOR EXAMPLE 8 [0459] Alam, S.M., McAdams, M., Boren, D., Rak, M., Scearce, R.M., Gao, F., Camacho, Z.T., Gewirth, D., Kelsoe, G., Chen, P., et al. (2007). The role of antibody polyspecificity and lipid reactivity in binding of broadly neutralizing anti-HIV-1 envelope human monoclonal antibodies 2F5 and 4E10 to glycoprotein 41 membrane proximal envelope epitopes. J.

Immunol. 775, 4424-4435.

[0460] Alam, S.M., Morelli, M., Dennison, S.M., Liao, H.X., Zhang, R., Xia, S.M., Rits- Volloch, S., Sun, L., Harrison, S.C., Haynes, B.F., et al. (2009). Role of HIV membrane in neutralization by two broadly neutralizing antibodies. Proc. Natl. Acad. Sci. U. S. A. 106, 20234-20239.

[0461] Bonsignori, M., Alam, S.M., Liao, H.X., Verkoczy, L., Tomaras, G.D., Haynes, B.F., and Moody, M.A. (2012). HIV-1 antibodies from infection and vaccination: insights for guiding vaccine design. Trends Microbiol. 20, 532-539.

[0462] Bonsignori, M., Hwang, K.K., Chen, X., Tsao, C.Y., Morris, L., Gray, E., Marshall, D.J., Crump, J. A., Kapiga, S.H., Sam, N.E., et al. (2011). Analysis of a clonal lineage of HIV- 1 envelope V2/V3 conformational epitope-specific broadly neutralizing antibodies and their inferred unmutated common ancestors. J. Virol. 85, 9998-10009.

[0463] Bonsignori, M., Wiehe, K., Grimm, S.K., Lynch, R., Yang, G., Kozink, D.M., Perrin, F., Cooper, A. J., Hwang, K.K., Chen, X., et al. (2014). An autoreactive antibody from an SLE/HIV-1 individual broadly neutralizes HIV-1. J. Clin. Invest. 124, 1835-1843.

[0464] Boyd, S.D., Marshall, EX., Merker, J.D., Maniar, J.M., Zhang, L.N., Sahaf, B., Jones, CD., Simen, B.B., Hanczaruk, B., Nguyen, K.D., et al. (2009). Measurement and Clinical Monitoring of Human Lymphocyte Clonality by Massively Parallel V-D-J Pyrosequencing. Sci. Transl. Med. 1.

[0465] Dosenovic, P., von Boehmer, L., Escolano, A., Jardine, J., Freund, N.T., Gitlin, A.D., McGuire, A.T., Kulp, D.W., Oliveira, T., Scharf, L., et al. (2015). Immunization for HIV-1 Broadly Neutralizing Antibodies in Human Ig Knockin Mice. Cell 161, 1505-1515.

[0466] Felsenstein, J. (2005). PHYLIP (Phylogeny Inference Package), 3.6a3 edn (Seattle, WA: distributed by the author: Department of Genome Sciences, University of Washington).

[0467] Foote, J., and Milstein, C. (1991). Kinetic maturation of an immune response. Nature 352, 530-532.

[0468] Gao, F., Bonsignori, M., Liao, H.X., Kumar, A., Xia, S.M., Lu, X., Cai, F., Hwang, K.K., Song, H., Zhou, T., et al. (2014). Cooperation of B cell lineages in induction of HIV-1- broadly neutralizing antibodies. Cell 158, 481-491. [0469] Georgiev, I S., Doria-Rose, N.A., Zhou, T.Q., Do Kwon, Y., Staupe, R.P., Moquin, S., Chuang, G.Y., Louder, M.K., Schmidt, S.D., Altae-Tran, H.R., et al. (2013). Delineating Antibody Recognition in Polyclonal Sera from Patterns of HIV- 1 Isolate Neutralization. Science 340, 751-756.

[0470] Haynes, B.F. (2015). New approaches to HIV vaccine development. Curr. Opin.

Immunol. 35, 39-47.

[0471] Haynes, B.F., and Bradley, T. (2015). Broadly Neutralizing Antibodies and the Development of Vaccines. JAMA 313, 2419-2420.

[0472] Haynes, B.F., Fleming, J., St Clair, E.W., Katinger, H., Stiegler, G, Kunert, R., Robinson, J., Scearce, R.M., Plonk, K., Staats, H.F., et al. (2005). Cardiolipin polyspecific autoreactivity in two broadly neutralizing HIV-1 antibodies. Science 308, 1906-1908.

[0473] Haynes, B.F., Kelsoe, G, Harrison, S.C., and Kepler, T.B. (2012). B-cell-lineage immunogen design in vaccine development with HIV-1 as a case study. Nat. Biotechnol. 30, 423-433.

[0474] Haynes, B.F., and Verkoczy, L. (2014). AIDS/HIV. Host controls of HIV neutralizing antibodies. Science 344, 588-589.

[0475] Hraber, P., Korber, B., Wagh, K., Giorgi, E.E., Bhattacharya, T., Gnanakaran, S., Lapedes, A.S., Learn, G.H., Kreider, E.F., Li, Y., et al. (2015). Longitudinal Antigenic Sequences and Sites from Intra-Host Evolution (LASSIE) Identifies Immune- Selected HIV Variants. Viruses 7, 5443-5475.

[0476] Hraber, P., Seaman, M.S., Bailer, R.T., Mascola, J.R., Montefion, D.C., and Korber, B.T. (2014). Prevalence of broadly neutralizing antibody responses during chronic HIV-1 infection. AIDS 28, 163-169.

[0477] Jardine, J., Julien, J.P., Menis, S., Ota, T., Kalyuzhniy, O., McGuire, A., Sok, D., Huang, P.S., MacPherson, S., Jones, M., et al. (2013). Rational HIV immunogen design to target specific germline B cell receptors. Science 340, 711-716.

[0478] Jardine, J.G., Ota, T., Sok, D., Pauthner, M., Kulp, D.W., Kalyuzhniy, O., Skog, P.D., Thinnes, T.C., Bhullar, D., Briney, B., et al. (2015). HIV-1 VACCINES. Priming a broadly neutralizing antibody response to HIV-1 using a germline-targeting immunogen. Science 349, 156-161.

[0479] Keele, B.F., Giorgi, E.E., Salazar-Gonzalez, J.F., Decker, J.M., Pham, K.T., Salazar, M.G., Sun, C, Grayson, T., Wang, S., Li, H., et al. (2008). Identification and characterization of transmitted and early founder virus envelopes in primary HIV-1 infection. Proc. Natl. Acad. Sci. U. S. A. 105, 7552-7557.

[0480] Kepler, T.B. (2013). Reconstructing a B-cell clonal lineage. I. Statistical inference of unobserved ancestors. FlOOORes 2, 103.

[0481] Kepler, T.B., Munshaw, S., Wiehe, K., Zhang, R., Yu, J.S., Woods, C.W., Denny, T.N., Tomaras, G.D., Alam, S.M., Moody, M.A., et al. (2014). Reconstructing a B-Cell Clonal Lineage. II. Mutation, Selection, and Affinity Maturation. Front. Immunol. 5, 170.

[0482] Klein, F., Diskin, R., Scheid, J.F., Gaebler, C, Mouquet, H., Georgiev, I S., Pancera, M., Zhou, T., Incesu, R.B., Fu, B.Z., et al. (2013). Somatic mutations of the immunoglobulin framework are generally required for broad and potent HIV-1 neutralization. Cell 153, 126- 138.

[0483] Liao, H.X., Lynch, R., Zhou, T., Gao, F., Alam, S.M., Boyd, S.D., Fire, A.Z., Roskin, K.M., Schramm, C.A., Zhang, Z., et al. (2013). Co-evolution of a broadly neutralizing HIV-1 antibody and founder virus. Nature 496, 469-476.

[0484] Liu, M., Yang, G, Wiehe, K., Nicely, N.I., Vandergrift, N.A., Rountree, W.,

Bonsignori, M., Alam, S.M., Gao, J., Haynes, B.F., et al. (2015). Polyreactivity and autoreactivity among HIV-1 antibodies. J. Virol. 89, 784-798.

[0485] Mascola, J.R., and Haynes, B.F. (2013). HIV-1 neutralizing antibodies: understanding nature's pathways. Immunol. Rev. 254, 225-244.

[0486] McDonald, I.K., and Thornton, J.M. (1994). Satisfying hydrogen bonding potential in proteins. J. Mol. Biol. 238, 777-793.

[0487] McGuire, A.T., Hoot, S., Dreyer, A.M., Lippy, A., Stuart, A., Cohen, K.W., Jardine, J., Menis, S., Scheid, J.F., West, A.P., et al. (2013). Engineering HIV envelope protein to activate germline B cell receptors of broadly neutralizing anti-CD4 binding site antibodies. J. Exp. Med. 210, 655-663.

[0488] Otwinowski, Z., and Minor, W. (1997). Processing of X-ray diffraction data collected in oscillation mode. Method Enzymol. 276, 307-326.

[0489] Pugach, P., Ozorowski, G., Cupo, A., Ringe, R., Yasmeen, A., de Val, N., Derking, R., Kim, H.J., Korzun, J., Golabek, M., et al. (2015). A native-like SOSIP.664 trimer based on an HIV-1 subtype B env gene. J. Virol. 89, 3380-3395.

[0490] Rudicell, R.S., Do Kwon, Y., Ko, S.Y., Pegu, A., Louder, M.K., Georgiev, I S., Wu, X.L., Zhu, J., Boyington, J.C., Chen, X.J., et al. (2014). Enhanced Potency of a Broadly Neutralizing HIV-1 Antibody In Vitro Improves Protection against Lentiviral Infection In Vivo. J. Virol. 88, 12669-12682.

[0491] Sanders, R.W., Derking, R., Cupo, A., ien, J.P., Yasmeen, A., de Val, N., Kim, H.J., Blattner, C, de la Pena, A.T., Korzun, J., et al. (2013). A next-generation cleaved, soluble HIV-1 Env trimer, BG505 SOSIP.664 gpl40, expresses multiple epitopes for broadly neutralizing but not non-neutralizing antibodies. PLoS Pathog. 9, el 003618.

[0492] Scheid, J.F., Mouquet, H., Feldhahn, N., Seaman, M.S., Velinzon, K., Pietzsch, J., Ott, R.G., Anthony, R.M., Zebroski, H., Hurley, A., et al. (2009). Broad diversity of neutralizing antibodies isolated from memory B cells in HIV-infected individuals. Nature 458, 636-640.

[0493] Scheid, J.F., Mouquet, H., Ueberheide, B., Diskin, R., Klein, F., Oliveira, T.Y., Pietzsch, J., Fenyo, D., Abadir, A., Velinzon, K., et al. (2011). Sequence and structural convergence of broad and potent HIV antibodies that mimic CD4 binding. Science 333, 1633- 1637.

[0494] Sui, Z.W., Chen, Q.J., Wu, R., Zhang, H.B., Zheng, M., Wang, H.Z., and Chen, Z. (2010). Cross-protection against influenza virus infection by intranasal administration of M2- based vaccine with chitosan as an adjuvant. Arch. Virol. 155, 535-544.

[0495] Verkoczy, L., Chen, Y., Zhang, J., Bouton-Verville, H., Newman, A., Lockwood, B., Scearce, R.M., Montefion, D.C., Dennison, S.M., Xia, S.M., et al. (2013). Induction of HIV-1 broad neutralizing antibodies in 2F5 knock-in mice: selection against membrane proximal external region-associated autoreactivity limits T-dependent responses. J. Immunol. 191, 2538-2550.

[0496] Verkoczy, L., Diaz, M., Holl, T.M., Ouyang, Y.B., Bouton-Verville, H., Alam, S.M., Liao, H.X., Kelsoe, G., and Haynes, B.F. (2010). Autoreactivity in an HIV-1 broadly reactive neutralizing antibody variable region heavy chain induces immunologic tolerance. Proc. Natl. Acad. Sci. U. S. A. 107, 181-186.

[0497] Verkoczy, L., Kelsoe, G., Moody, M.A., and Haynes, B.F. (2011). Role of immune mechanisms in induction of HIV-1 broadly neutralizing antibodies. Curr. Opin. Immunol. 23, 383-390.

[0498] Wu, X., Yang, Z.Y., Li, Y., Hogerkorp, CM., Schief, W.R., Seaman, M.S., Zhou, T., Schmidt, S.D., Wu, L., Xu, L., et al. (2010). Rational design of envelope identifies broadly neutralizing human monoclonal antibodies to HIV-1. Science 329, 856-861. [0499] Wu, X., Zhou, T., Zhu, J., Zhang, B., Georgiev, L, Wang, C, Chen, X., Longo, N.S.,

Louder, M., McKee, K., et al. (2011). Focused evolution of HIV-1 neutralizing antibodies revealed by structures and deep sequencing. Science 333, 1593-1602.

[0500] Yang, G., Holl, T.M., Liu, Y., Li, Y., Lu, X., Nicely, N.I., Kepler, T.B., Alam, S.M.,

Liao, H.X., Cain, D.W., et al. (2013). Identification of autoantigens recognized by the 2F5 and 4E10 broadly neutralizing HIV-1 antibodies. J. Exp. Med. 210, 241-256.

[0501] Zhou, T., Georgiev, L, Wu, X., Yang, Z.Y., Dai, K., Finzi, A., Kwon, Y.D., Scheid,

J.F., Shi, W., Xu, L., et al. (2010). Structural basis for broad and potent neutralization of HIV-

1 by antibody VRC01. Science 329, 811-817.

[0502] Zhou, T., Lynch, R.M., Chen, L., Acharya, P., Wu, X., Doria-Rose, N.A., Joyce, M.G., Lingwood, D., Soto, C, Bailer, R.T., et al. (2015). Structural Repertoire of HIV-1- Neutralizing Antibodies Targeting the CD4 Supersite in 14 Donors. Cell 161, 1280-1292.

[0503]

[0504]

Example 10: DH542 antibodies

[0505] The nucleotide and amino acid sequences of the VH and VL of DH542 QSA are shown below. DH542 QSA antibody has the VH of DH542 and the VL called DH542-QSA

[0506] >DH542_HC_nt (SEQ ID NO: 178)

CAGGTGCAGCTGGTGCAGTCTGGGGCTCAAATGAAGAACCCTGGGGCCTCAGTGAAGGTCTCCTGCGC GCCTTCTGGATATACCTTCACCGACTTTTACATACATTGGTTGCGCCAGGCCCCTGGCCAGGGGCTTC AGTGGATGGGATGGATGAACCCTCAGACTGGTCGCACAAACACTGCACGAAACTTTCAGGGGAGGGTC ACCATGACCAGGGACACGTCCATCGGCACAGCCTACATGGAGTTGAGAAGCCTGACATCTGACGACAC GGCCATATATTACTGTACGACAGGGGGATGGATCAGTCTTTACTATGATAGTAGTTATTACCCCAACT TTGACCACTGGGGTCAGGGAACCCTGCTCACCGTCTCCTCAG

[0507] >DH542_HC_aa (SEQ ID NO: 179)

QVQLVQSGAQMK PGASVKVSCAPSGYT FTDFY IHWLRQAPGQGLQWMGWMNPQTGRTNTARNFQGRV TMTRDTS IGTAYMELRSLT SDDTAIYYCTTGGWI SLYYDSSYY PNFDHWGQGTLLTVSS

[0508] >DH542_LC_nt_corrected (DH542 QSA) (SEQ ID NO: 180)

CAGTCTGCCCTGACTCAGCCTGCCTCCGTGTCTGGGTCTCCTGGACAGTCGATCACCATCTCCTGCAC TGGAACCAAGTATGATGTTGGGAGTCATGACCTTGTCTCCTGGTACCAACAGTACCCAGGCAAAGTCC CCAAATACATGATTTATGAAGTCAATAAACGGCCCTCAGGAGTTTCTAATCGCTTCTCTGGCTCCAAA TCTGGCAACACGGCCTCCCTGACAATCTCTGGGCTCCGGGCTGAGGACGAGGCTGACTATTATTGCTG TTCATTTGGAGGGAGTGCCACCGTGGTCTGCGGCGGCGGGACCAAGGTGACCGTCCTAg [0509] >DH542_LC_aa_corrected (DH542 QSA) (SEQ ID NO: 181)

QSALTQPASVSGSPGQSI ISCTGTKYDVGSHDLVSWYQQYPGKVPKYMIYEVNKRPSGVSNRFSGSK SGNTASLTISGLRAEDEADYYCCSFGGSATVVCGGGTKVTVL

[0510] DH542-L4 is an antibody that has a VH of DH542 and VL of DH429 (Figure 2).

Example 10: Exemplary HIVxCD3 Bispecific Diabodies with Fc

Domains

[0511] Provided herein are stable, structurally compact, bispecific diabodies that have interchain disulfide bonds that may be engineered with an Fc Domain to extend serum half-life. The antigen binding arms of such bispecific diabodies are advantageously selected to co- engage immune effector cells (e.g. , T cells, NK cells, etc.) with antigen-expressing target cells (e.g. , HIV-1 infected cells) and activate and redirect the cytolytic activity of immune effector cell against the antigen expressing target cells.

[0512] An HIVxCD3 bispecific diabody with an Fc Domain designated "CH557xCD3 Fc" was designed and expressed. This molecule comprises three polypeptide chains and includes an HIV-1 binding arm derived from the CH557 antibody described herein, a CD3 effector cell binding arm derived from a humanized anti-CD38 mAb (hXRCD3), and CH2-CH3 IgGl Fc Domains. The general structure of the polypeptide chains and a schematic of the assembled chains are shown in Figures 5A-5B. The amino acid sequences and a representative polynucleotide encoding each polypeptide chain are presented in Figures 5C-5D. A corresponding negative control bispecific diabody with an irrelevant binding arm [aRSV derived from palivizumab] instead of the CH557 HIV-1 arm (designated "RSVxCD3 Fc,") was also generated. In addition, a comparator bispecific diabody (designated "A32xCD3 Fc") having an HIV-1 binding arm derived from the A32 antibody (Protein Data Bank (PBD) ID Code 3TNM) instead of CH557, and additional negative control bispecific diabodies having the A32 HIV-1 arm with an irrelevant binding arm [aRSV or afluorescein] instead of the CD3 arm (designated "A32xRSV Fc," and "A32x4420," respectively) were generated. The HIVxCD3 Fc bispecific diabodies are capable of simultaneously binding to HIV-1 and CD3. The control RSVxCD3 Fc bispecific diabody is capable of simultaneously binding to RSV and CD3 and the control A32xRSV Fc bispecific diabody is capable of simultaneously binding to HIV-1 and RSV. Each of the generated bispecific diabodies is a heterotrimer of polypeptide Chains 1, 2 and 3 have the general structure provided in Figure 3A (also see, e.g., Figures 5A-5B). Chains 1 and 2 comprise the VH and VL Domains of CH557, A32, XRCD3, or palivizumab, as detailed above, while Chain 3 is common to each diabody molecule.

[0513] Methods for forming bispecific diabodies, including such diabodies comprising an Fc Domain, are provided in WO 2006/113665, WO 2008/157379, WO 2010/080538, WO 2012/018687, WO 2012/162068, WO 2012/162067, WO 2014/159940, WO 2015/021089, WO 2015/026892 and WO 2015/026894.

[0514] Table 5 provides the SEQ ID NOs for each of the polypeptide chains present in CH557xCD3 Fc molecule. Table 4 provides SEQ ID NOs for the VL and VH Domains of CH557 (as provided herein), A32 (additional HIV-1 Env antibodies are known in the art, see, e.g., WO 2016/054101), hXRCD3 (additional anti-CD3 antibodies are known in the art, see, e.g., WO 2012/162067), and palivizumab (additional anti-RSV antibodies are known in the art, see, e.g., US 6,818,216), the SEQ ID NOs for the VL and VH Domains of the exemplary anti-CD16 mAb h3G8, anti-CD8 mAb OKT8 and anti-CD8 mAb TRX2 are also provided. In addition, the SEQ ID NOs for representative nucleic acid encoding sequences are provided.

[0515] For additional sequences related to this example see Table 4.

Example 10A: Binding Properties of HIVxCD3 Bispecific Diabodies with Fc Domains

[0516] The binding of CH557xCD3 Fc to recombinant HIV-1 Env and human CD3 was examined by ELISA assay. Briefly, microtiter plates were coated with recombinant proteins (human CD3s/5 heterodimer, or M.CONS gpl40 (Liao HX, et al. A group M consensus envelope glycoprotein induces antibodies that neutralize subsets of subtype B and C HIV-1 primary viruses. Virology. 2006;353(2):268-282)) in buffer and blocked. Serial dilutions of the bispecific diabodies (CH557xCD3 Fc, the comparator A32xCD3 Fc, or the controls RSVxCD3 Fc and A32xRSV Fc) were applied followed by sequential addition of biotinylated anti-EK coil antibody and streptavidin-HRP. For bispecific binding assays, the plate was coated with M.CONS gpl40, and diabody application was followed by sequential addition of biotinylated CD3s/5 and streptavidin-HRP. HRP activity was detected with a

chemiluminescent substrate. For bispecific binding assays, the plate was coated with

M.CONS gpl40, and diabody application was followed by sequential addition of biotinylated CD3s/5 and streptavidin-HRP. CH557xCD3 Fc and A32xCD3 Fc both exhibited binding to recombinant human CD3 and HIV-1 Env, individually and simultaneously, while the control molecules, RSVxCD3 Fc and A32xRSV Fc, only exhibited binding to human CD3 or HIV-1 Env, respectively, as shown by ELISA (Figures 6A-6C).

[0517] Cell surface binding of CH557xCD3 Fc to cells expressing HIV-1 Env (HEK293- D371 cells, expressing HIV-1 CM244 (subtype AE) gpl40) and human CD3 (human primary T cells) was examined by flow cytometric analysis. Briefly, Serial dilutions of the bispecific diabodies: CH557xCD3 Fc; the comparator A32xCD3 Fc; or the control RSVxCD3 Fc, were incubated with HEK293-D371 cells or human primary T cells (Pan T cells) in FACS buffer containing a blocking agent (e.g., human albumin serum). After washing, cells were resuspended in buffer containing biotin-conjugated mouse anti-EK antibody (recognizes the E/K heterodimerization region of diabody proteins), mixed with streptavidin-PE and incubated in the dark. Cells were washed, resuspended with FACS buffer, and analyzed by flow cytometry. CH557xCD3 Fc and A32xCD3 Fc exhibited binding to both HEK293-D375 and Pan T cells, while the control, RSVxCD3 Fc, only exhibited binding to Pan T cells (Figures 7A-7B). CH557xCD3 Fc exhibited stronger binding to HEK293-D375 cells in these studies (Figure 7A).

Example 10B: Cell Killing Activity of HIVxCD3 Bispecific Diabodies with Fc Domains

[0518] The ability of CH557xCD3 Fc to mediate redirected cell killing of target cells expressing HIV-1 Env was examined using a cytotoxic T lymphocyte (CTL) assay. Briefly, the HIV-1 Env expressing cell line HEK293-D371 was treated with serial dilutions of the bispecific monovalent diabodies: CH557xCD3 Fc; the comparator A32xCD3 Fc; or the control RSVxCD3 Fc, together with effector cells (human PBMCs) from two different donors at an E/T ratio of 30: 1 for 24 hours. The percentage cytotoxicity (i.e., cell killing) was determined by measuring the release of lactate dehydrogenase (LDH) into the media by damaged cells as described previously (Moore PA et al. Application of dual affinity retargeting molecules to achieve optimal redirected T-cell killing of B-cell lymphoma. Blood. 2011;117(17):4542— 4551). As measured by LDH release assays, CH557xCD3 Fc and A32xCD3 Fc both mediated redirected human immune cells derived from healthy donors to kill the HEK293-D371 cells in a concentration dependent manner at an E:T ratio of 30: 1. The average EC50 values were 13.8 ng/mL and 12.3 ng/mL for CH557xCD3 Fc and A32xCD3 Fc, respectively. In contrast, no diabody-mediated redirected T-cell killing occurred with the RSVxCD3 control diabody in which the HIV-1 arm was replaced by an irrelevant one

(Figures 8A-8B)

Example IOC: Binding Properties of HIVxCD3 Bispecific Diabodies with Fc Domains to autologous HIV-1 CH557 envelope

Methods.

[0519] The 491 (control) and 598 (CH505 Transmitted/Founder Envelope- and GFP- transfected cell lines were used to detect the ability of the diabodies of interest to bind specifically to the HIV-1 envelope of the CH505 transmitted/founder isolate. The cell lines were incubated for 24 hours with Doxycycline at concentrations of 0.5, 0.0125, and 0.0031, and Oug/ml to achieve different levels of expression of the HIV-1 CH505 Envelope as determined in preliminary experiments. The GFP expression was used to determine the level of expression as of the transduced proteins. The A32x4420 and CH557xhXR32 diabodies were tested for staining at the concentration of 10 and 1 μg/ml. Palivizumabx4420

(PAlix4420) and PalivizumabxhXR32 (PalixhXR32) diabodies based on the anti-RSV mAb were used as control for each of the two anti -HIV-1 Env diabodies, respectively, using the same concentrations. In addition the A32 (anti-Cl/C2 Env mAb; positive control), Synagis (anti-RSV mAb, negative control), and CH65 (anti-Flu HA mAb; negative control) were used as controls for the specificity of the staining.

[0520] After 24hr. incubation with the different doses of Doxycycline, each cells culture was plated with and without Palix4420, PalixhXR32, A32 x 4420, and CH557 x hXR32 DARTS at lug/ml and lOug/ml. In addition, the same cell culture were plated with and without Synagis, A32, and CH65 human mAbs at lug/ml and lOug/ml. [0521] All the different cell cultures were incubated with diabodies and human mAbs for 2hrs at 37 degrees Celsius.

[0522] After 2hr incubation cells were washed 2x in PBS and stained with a LIVE/DEAD cell marker (Aqua LIVE/DEAD) for 20 minutes at RT.

[0523] After viability staining, cells were washed 2x in Wash Buffer and appropriate secondary mixes (anti-EK + IgG2b PE for DARTS or IgG PE for human mAbs) were added to wells for 30min at 4 degrees Celsius. The anti-EK and anti-EK+IgG2 conditions were used to stain the cells as additional control conditions.

[0524] Cells were washed 2x in Wash Buffer and fixed in 1% Formalin Solution until acquiring on BD LSR Fortessa.

[0525] The analysis of data was performed by gating on the viable (Aqua Live/Dead negative) and GFP positive cells for the 598 cell line and on viable cells for the 491 control cell line. Each frequency reported in the figure refers to the percentage of diabodies or mAb positive cells referred to the Live+/GFP+ cells.

[0526] Results: The levels of GFP expression detected with each Doxycycline concentration in each staining condition are reported in Figure 48 and 49 for the diabodies and control mAbs, respectively.

[0527] As expected, we observed three different levels of GFP expression ranging from >80% of cells to less than 20%. A minimal constitutive expression of GFP was also observed in absence of Doxycycline.

[0528] The analysis of the staining with the A32x4420 and CH557xhXR32 to the 491 control cell line did not reveal any non specific staining (Figure 50, top panels). In contrast, the 598 cell lines were stained proportionally to the level of GFP observed due to the stimulation with Doxycycline, ranging from 82% to 11.7% with the CH557xhXR32 DART, and

CH557xhXR32 diabody binding was always better than the A32x4420 diabody binding at each concentration (Figure 50 bottom panels). A minimum background of 10% staining with the CH557xhXR32 was observed when the 598 cell line was used without incubation with Doxycycline. This is due to a minimum constitutive expression of the transduced genes as indicated in Figure 48 and 49 for GFP expression. The staining with the A32 mAb indicated the expression of HIV-1 Env on the membrane of the cells according to a similar profile and the specificity of the staining of the 598 cell line (Figure 51). [0529] The analysis of the staining with diabody and mAb controls represented by the anti- EK, anti-EK+IgG2, Palix4420, PalixhXR32, Synagis, and CH65 mAbs combination did not reveal non-specific staining of the 598 cell line.

Example 11A: Additional multispecific molecules

[0530] Any one of the antibodies from the CH235 lineage or DH270 lineage could be constructed in any one of the multispecific molecule formats described herein. The effector arm could target be any one of the non-limiting examples of CD2, CD3, CD8, CD 16, T-Cell Receptor (TCR), NKG2D, etc. epitopes. Non-limiting examples are provided in the Table below. The information in the specification can be readily used for alternative designs of the listed molecules, and for design of other bispecific molecules, for example CH235 lineage members or DH270 lineage members using CDRs, or VH and VL chains from these HIV-1 antibodies.

[0531] Table 6 shows a summary of some additional non limiting embodiments of bispecific diabody (with and without Fc Domains) that may be generated comprising alternative HIV-1 and/or alternative effector cell binding specificities. The information in specification can be readily used for alternative design of the listed bispecific diabodies, and for design of other bispecific diabodies, for example bispecific diabodies comprising a CD 16 binding specificity instead of CD3 and/or incorporating the HIV-1 binding specificity of CH235 lineage antibodies or DH270 lineage antibodies using CDRs, or VH and VL chains from these antibodies.

E-coil 518or519 K-coil 520 or 521

E-coil 519or518 K-coil 520 or 521

VL(HIV-l) SEQ ID NOs: NH₂-VL(CD3) SEQ ID NO:

553 502

565 or

567 or

NH₂-

570

VL(CD16) SEQ ID NO:

574

506

578

582

586

590

594

Peptide Linker SEQ ID NO: Peptide Linker SEQ ID NO: 1 508 1 508

VH(CD3) SEQ ID NO: VH(HIV-l) SEQ ID NOs:

500 551 or 564

or 566

VH(CD16)

568

SEQ ID NO:

572

504

576

580

584

588

592

Peptide Linker SEQ ID NO: Peptide Linker SEQ ID NO: 2 509 or 510 2 509 or 510

K-coil SEQ ID NO: E-coil SEQ ID NO:

520 or 521 518 or 519 or or or or

SEQ ID NO: SEQ ID NO:

E-coil 518 or 519 K-coil 520 or 521

E-coil 518or519 K-coil 520 or 521 Peptide Linker SEQ ID NO:

3 or 523 or 522

Spacer Linker

CH2-CH3 SEQ ID NO:

527, 528 or

529

[0532] Wherein SEQ ID NOs: 553, 565, 567, 570, 574, 578, and 582 correspond to the VL chain of antibody CH557, CH556, CH555, CH493, CH492, CH491, and CH490, respectively and as described in Table 4. Wherein SEQ ID NOs: 586, 590, and 594 correspond to the VL chain of antibody DH542, DH542 QSA, and D542 L4, respectively and as described in Table 4.

[0533] Wherein SEQ ID NOs: 551, 564, 566, 568, 572, 576, and 580 correspond to the VH chain of antibody CH557, CH556, CH555, CH493, CH492, CH491, and CH490, respectively and as described in Table 4. Wherein SEQ ID NOs: 584, 588, and 592 correspond to the VH chain of antibody DH542, DH542 QSA, and D542 L4, respectively and as described in Table 4.

Example 11B: Exemplary Trivalent and Control Molecules with Fc Domains

[0534] As provided herein trivalent binding molecules having three Antigen-Binding Domains provide additional functionality as they can co-engage multiple epitopes present on the surface of an effector cell, such as a T lymphocyte (e.g., CD3 and CD8), or they can bind multiple epitopes of HIV-1 (e.g., epitopes of different HIV-1 isolates or different epitopes of HIV-1 Env). The antigen binding arms of such trivalent binding molecules are advantageously selected to co-engage immune effector cells (e.g., T cells, NK cells, etc) with antigen-expressing target cells (e.g., HIV-1 infected cells) and activate and redirect the cytolytic activity of immune effector cell against the antigen expressing target cells.

[0535] Table 7 shows a summary of some non-limiting embodiments of trivalent binding molecules (having 3 or 4 polypeptide chain) that may be generated. The information in specification can be readily used for alternative design of the listed trivalent binding molecules, and for design of other trivalent binding molecules, for example trivalent binding molecules comprising alternative CD8 binding specificities and/or incorporating the HIV-1 binding specificity of CH235 lineage antibodies or DH270 lineage antibodies using CDRs, or VH and VL chains from these antibodies.

CL lambda 516or517 CL lambda 516or517

[0536] Wherein SEQ ID NOs: 553, 565, 567, 570, 574, 578, and 582 correspond to the VL chain of antibody CH557, CH556, CH555, CH493, CH492, CH491, and CH490, respectively and as described in Table 4. Wherein SEQ ID NOs: 586, 590, and 594 correspond to the VL chain of antibody DH542, DH542 QSA, and D542 L4, respectively and as described in Table 4.

[0537] Wherein SEQ ID NOs: 551, 564, 566, 568, 572, 576, and 580 correspond to the VH chain of antibody CH557, CH556, CH555, CH493, CH492, CH491, and CH490, respectively and as described in Table 4. Wherein SEQ ID NOs: 584, 588, and 592 correspond to the VH chain of antibody DH542, DH542 QSA, and D542 L4, respectively and as described in Table 4.

[0538] Table 4 provides the amino acid sequences of one such exemplary trivalent binding molecule having four polypeptide chain which may be generated are provided in SEQ ID NOs: 555, 557, 561, and 562. Chain 1 of this exemplary molecule comprises: an N-terminus, the VL Domain of CH557 (SEQ ID NO: 553), a Peptide Linker 1 (SEQ ID NO: 508), the VH Domain of hXR32 (SEQ ID NO: 500), a Peptide Linker 2 (SEQ ID NO: 510), a cysteine engineered E-coil, a Spacer Linker 3 (SEQ ID NO: 522), a knob bearing CH2-CH3 (SEQ ID NO: 531), and a C-terminus. Chain 2 of this exemplary molecule comprises: an N-terminus, the VL Domain of hXR32 (SEQ ID NO: 503), a Peptide Linker 1 (SEQ ID NO: 508), the VH Domain of CH557 (SEQ ID NO: 551), a Peptide Linker 2 (SEQ ID NO: 510), a cysteine engineered K-coil, a Spacer Linker 3 (SEQ ID NO: 521), and a C-terminus. Chain 3 of this exemplary molecule comprises: an N-terminus, the VH Domain of OKT8 (SEQ ID NO: 543), a CHI-Hinge Domain (SEQ ID NO: 515), a hole bearing CH2-CH3 (SEQ ID NO: 533), and a C-terminus. Chain 4 of this exemplary molecule comprises: an N-terminus, the VL Domain of OKT8 (SEQ ID NO: 545), a CL Kappa Domain (SEQ ID NO: 516), and a C- terminus. It will be noted that polypeptide chains 1 and 2 of such an exemplary trivalent binding molecule are the same as those present in the bispecific Fc bearing diabody

CH557xCD3 Fc described above.

[0539] Table 4 provides the amino acid sequences of another such exemplary trivalent binding molecule having three polypeptide chain which may be generated are provided in SEQ ID NOs: 555, 557, and 563. Chain 1 of this exemplary molecule comprises: an N- terminus, the VL Domain of CH557 (SEQ ID NO: 553), a Peptide Linker 1 (SEQ ID NO: 508), the VH Domain of hXR32 (SEQ ID NO: 500), a Peptide Linker 2 (SEQ ID NO: 510), a cysteine engineered E-coil, a Spacer Linker 3 (SEQ ID NO: 522), a knob bearing CH2-CH3 (SEQ ID NO: 531), and a C-terminus. Chain 2 of this exemplary molecule comprises: an N- terminus, the VL Domain of hXR32 (SEQ ID NO: 503), a Peptide Linker 1 (SEQ ID NO: 508), the VH Domain of CH557 (SEQ ID NO: 551), a Peptide Linker 2 (SEQ ID NO: 510), a cysteine engineered K-coil, a Spacer Linker 3 (SEQ ID NO: 521), and a C-terminus. Chain 3 of this exemplary molecule comprises: an N-terminus, the VL Domain of OKT8 (SEQ ID NO: 545), a Peptide Linker 5 (SEQ ID NO: 526), the VH Domain of OKT8 (SEQ ID NO: 543), a CHI-Hinge Domain (SEQ ID NO: 515), a hole bearing CH2-CH3 (SEQ ID NO: 533), and a C-terminus. It will be noted that polypeptide chains 1 and 2 of such an exemplary trivalent binding molecule are the same as those present in the bispecific Fc bearing diabody CH557xCD3 Fc described above.

[0540] Exemplary trivalent binding molecules having three or four polypeptide chains, comprising two HIV-1 Env binding sites and one CD3 binding site are readily generated, for example by replace the VL and VH Domains of the anti-CD antibody in the above described molecules with the VL and VH Domains of an anti-HIV-1 Env antibody {e.g., CH557).

[0541] The ability of such exemplary trivalent binding molecules to bind to HIV-1 Env, CD3, and/or CD8 may be evaluated using the methods describe above, or well known in the art (see, e.g., WO2016/054101). The activity of such trivalent binding molecules to mediate redirected cell killing of target cells expressing HIV-1 Env may be examined using the cytotoxic T lymphocyte (CTL) assay described above, or similar assays known in the art (see, e.g., Sung, JAM et al. Dual-Affinity Re-Targeting proteins direct T cell-mediated cytolysis of latently HIV-1 -infected cells. J Clin Invest. 2015; 125(11):4077-4090;

and WO2016/054101).

Example 12: Combinations of multispecific anitbodies

[0542] Multispecific antibodies with different HIV-1 specificity, e.g CD4 binding site and V3 glycan binding, could also be tested in combination.

[0543] Various combinations of HIV-1 multispecific antibodies to mediate redirected cell killing of target cells expressing HIV-1 Env may be examined using the cytotoxic T lymphocyte (CTL) assay described above, or similar assays known in the art (see, e.g., Sung, JAM et al. Dual-Affinity Re-Targeting proteins direct T cell-mediated cytolysis of latently HIV-1 -infected cells. J Clin Invest. 2015;125(11):4077-4090; and WO2016/054101).

Ccombinations of multispecific molecules, either with different HIV-1 specificity and/or different effector cell specificity, will be tested whether they provide enhanced benefits. Example 13: Exemplary V3 glycan Specific HIVxCD3 Bispecific Diabody

[0544] As provided herein, the binding specificity of the anti-HIV-1 antibody DH542 differs from the other anti-HIV-1 antibodies provide herein, in that it is specific for the V3 glycan of HIV-1 Env, According, the VH and VL Domains of DH542 are utilized in the generation of HIVxCD3 bispecific diabodies (and/or trivalent binding molecule as provided herein) having binding specificity for the V3 glycan of HIV-1 Env. Exemplary, HIVxCD3 bispecific diabodies, having three polypeptide chains, an HIV-1 binding arm derived from the DH542 antibody described herein, a CD3 effector cell binding arm derived from a humanized anti- CD3s mAb (hXRCD3), and CH2-CH3 IgGl Fc Domains may have the general structure shown in Figures 5A-5B. Table 8 provides the SEQ ID NOs for each of the polypeptide chains present in one non-limiting embodiment of such a molecule, designated herein as "DH542xCD3 Fc". It will be appreciated based on the instant disclosure that the third polypeptide chain of DH542xCD3 Fc may be identical to the HIVxCD3 bivalent diabodies having an Fc Domain provide above.

[0545] Chain 1 of this exemplary molecule comprises: an N-terminus, the VL Domain of DH542 (SEQ ID NO: 586), a Peptide Linker 1 (SEQ ID NO: 508), the VH Domain of hXR32 (SEQ ID NO: 500), a Peptide Linker 2 (SEQ ID NO: 510), a cysteine engineered E-coil, a Spacer Linker 3 (SEQ ID NO: 522), a knob bearing CH2-CH3 (SEQ ID NO: 531), and a C- terminus. Chain 2 of this exemplary molecule comprises: an N-terminus, the VL Domain of hXR32 (SEQ ID NO: 503), a Peptide Linker 1 (SEQ ID NO: 508), the VH Domain of DH542 (SEQ ID NO: 584), a Peptide Linker 2 (SEQ ID NO: 510), a cysteine engineered K-coil, a Spacer Linker 3 (SEQ ID NO: 521), and a C-terminus. Chain 3 of this exemplary molecule comprises: an N-terminus, a Peptide Linker 3 (SEQ ID NO: 523), a hole bearing CH2-CH3 (SEQ ID NO: 533), and a C-terminus. Such bispecific diabodies, may be generated using the methods as provided above.

[0546] The ability of such exemplary HIV-1 Env V3 glycan binding bispecific diabodies to bind to HIV-1 Env V3 glycan and CD3 may be evaluated using the methods describe above, or well known in the art (see, e.g., WO2016/054101). The activity of such exemplary bispecific diabodies to mediate redirected cell killing of target cells expressing HIV-1 Env may be examined using the cytotoxic T lymphocyte (CTL) assay described above, or similar assays known in the art (see, e.g., Sung, JAM et al. Dual- Affinity Re-Targeting proteins direct T cell-mediated cytolysis of latently HIV-1 -infected cells. J Clin

Invest. 2015; 125(l l):4077-4090; and WO2016/054101).

Example 14: Prophetic examples

[0547] All multispecific molecules described herein could be tested in any other suitable assay. For non-limiting examples of further studies and characterization See Sung et al. J Clin Invest. 2015;125(ll):4077-4090 ;Sloan DD, Lam C-YK, Irrinki A, Liu L, Tsai A, Pace CS, et al. (2015) Targeting HIV Reservoir in Infected CD4 T Cells by Dual-Affinity Retargeting Molecules (DARTs) that Bind HIV Envelope and Recruit Cytotoxic T Cells. PLoS Pathog 11(11): el005233. doi: 10.1371/journal.ppat.1005233

[0548] Non-limiting examples of these assays include: determination of binding properties of the multispecific molecules to HIV-1 envelope expressing cells and/or HIV-1 infected cells; determining whether the multispecific molecules of the invention induce redirected T-cell killing of cell lines expressing various Envelopes and concomitant T-cell activation;

determining whether multispecific molecules bind to the surface of HIV-1- infected CD4⁺T cells and redirect CD8⁺ T-cells to kill HIV-1 infected CD4⁺ cells using lymphocytes from HIV-1 seronegative donors; determining whether multispecific molecules redirect CD8⁺ T- cells to clear HIV-1 (e.g. JR-CSF, or any other suitable HIV-1 type)-superinfected CD4⁺ cells using lymphocytes from patients on suppressive ART; determining whether multispecific molecules redirect T cells from HIV-1 -infected individuals on suppressive ART to clear virus from resting CD4⁺ T cells following induction of latent virus expression.

[0549] Methods and Reagents [0550] Generation of Infectious Molecular Clones (IMCs). HIV-1 IMCs for any subtype B BaL, subtype AE CM235 and subtype C 1086.C were generated with the backbone derived from HL4-3 isolate as previously described See Edmonds TG et al. Replication competent molecular clones of HIV-1 expressing Renilla luciferase facilitate the analysis of antibody inhibition in PBMC. Virology. 2010;408(1): 1-13; Adachi A et al. Production of acquired immunodeficiency syndrome-associated retrovirus in human and nonhuman cells transfected with an infectious molecular clone. J Virol. 1986;59(2):284-291 ). All IMCs expressed the Renilla luciferase reporter gene and preserved all nine viral open reading frames. The Renilla luciferase reporter gene was expressed under the control of the HIV-1 Tat gene. Upon HIV-1 infection of CD4+ T cells, expression of Tat during HIV-1 replication will induce luciferase expression, which allows quantitation of infected cells by measuring relative luminescence units (RLU).

[0551] Cell Lines. Jurkat-522 F/Y GF cells, which constitutively express a fusion protein of Copepod Green Fluorescent Protein (copGFP) and Firefly Luciferase (System Biosciences), were generated at Macrogenics from Jurkat-522 F/Y cells by transduction and clone selection. HEK293-D371 cells, which have doxycycline-inducible expression of HIV-1 CM244

(subtype AE) gpl40 or CH505 T/F envelope, were obtained from Dr. John Kappes

(University of Alabama at Birmingham).

[0552] Flow Cytometric Analysis of diabody or mAb Binding to Cells. Diabodies at a desired concentration, e.g. 4 μg/mL are incubated with 10⁵ cells in 200 μΙ_, FACS buffer containing 10% human AB serum for 30 minutes at room temperature. After washing, cells are resuspended in 100 μΙ_, of 1 μg/mL biotin-conjugated mouse anti-EK antibody (recognizes the E/K heterodimerization region of diabody proteins and/or trivalent binding molecules), are mixed with 1 :500 diluted streptavidin-PE and incubated in the dark for 45 minutes at 2-8°C. Cells are washed, resuspended with FACS buffer, and analyzed with a BD Calibur flow cytometer and FlowJo software (TreeStar, Ashland OR). Binding to IMC-infected CD4⁺ T cells from normal human donors is conducted as previously described (54) for the A32 and 7B2 mAbs, and with biotin-conjugated mouse anti-EK antibody and 1 :500 diluted

streptavidin-PE for the HIVx4420 control diabodies.

[0553] The ability of CH557xCD3 Fc (or any of the bispecific diabodies and/or trivalent binding molecules) to mediate redirected cell killing of HIV-1 IMC-Infected CD4⁺ Cells may be examined using assays well known in the art (see, e.g., Sung, JAM et al. Dual -Affinity Re-Targeting proteins direct T cell-mediated cytolysis of latently HIV-infected cells. J Clin Invest. 2015; 125(l l):4077-4090; and WO2016/054101). Briefly, resting PBMC from normal healthy HIV-1 seronegative donors are activated {e.g., with anti-human CD3 and anti-human CD28 antibodies). A CD4+ enriched cell population is obtained by depletion of CD8+ T cells {e.g., using commercially available magnetic beads) and is spinoculated in the presence of the luciferase-expressing IMC-representing an HIV-1 subtype {e.g., AE (CM235); B (BaL); or C (1086. C)) and cultured for -72 hours. The CD4+-infected target cells are then incubated with resting CD8+ effector cells (which may be isolated by negative selection from autologous PBMC, using a commercial CD8+ T cell isolation kit) at different E/T ratios {e.g., 33 : 1, 11 : 1, 3 : 1, and 0: 1) in the absence or presence of bispecific diabodies for 6-48 hours at

concentrations ranging from 0.0001-1,000 ng/ml. Uninfected and infected target cells alone are included as additional controls. Each condition is tested in duplicate. After incubation, the percentage of specific lysis (%SL) of target cells is determined, for example by adding ViviRen Live Cell Substrate (Promega) and measuring the RLU on a luminometer. The %SL of target cells may be determined as described previously (Pollara J et al. HIV-1 Vaccine- Induced CI and V2 Env-Specific Antibodies Synergize for Increased Antiviral Activities. J Virol. 2014;88(14):7715-7726).

[0554] Redirected T-Cell Cytotoxicity Assay Against HIV-1 Env-expressing Cell Lines and Assessment of T-Cell Activation. Pan T cells are isolated from healthy human PBMCs with the Dynabeads® Untouched™ Human T Cells Kit (Invitrogen). HIV-1 Env expressing cell lines (1-4 xlO⁵ cells/mL) are treated with serial dilutions of diabodies (or trivalent binding molecules), together with human T cells at an effectontarget (E:T) ratio = 10: 1, or otherwise at varying E:T ratios as indicated, and incubated at 37°C, 5% C0₂ overnight. Cytotoxicity is measured by lactate dehydrogenase (LDH) release (CytoTox 96® Non-Radioactive

Cytotoxicity Assay, Promega) as described previously (32). With the Jurkat-522 F/Y GF cell line, cytotoxicity is also measured by luminescence using Luciferase-Glo substrate

(Promega). Specific lysis is calculated from luminiscence counts (RLU): cytotoxicity (%) = 100 x (1-(RLU of Sample ÷ RLU of Control)), where Control = average RLU of target cells incubated with effector cells in the absence of DART. Data are fit to a sigmoidal dose- response function to obtain 50% effective concentration (EC₅₀) and percent maximum specific lysis values. T-cell activation is measured by FACS analysis after cells in the assay plate are labeled with CD8-FITC, CD4-APC, and CD25-PE antibodies (BD Biosciences), followed by cell collection by FACS Calibur flow cytometer equipped with acquisition software CellQuest Pro Version 5.2.1 (BD Biosciences). Data analysis is performed using FlowJo software (Treestar, Inc).

[0555] Redirected T-Cell Cytotoxicity Assay Against HIV-1 IMC-Infected CD4⁺ Cells.

Cryopreserved resting PBMC from normal healthy HIV-1 seronegative donors are activated for 72 hours with anti-human CD3 (clone OKT3; eBioscience) and anti-human CD28 (clone CD28.2; BD Pharmingen). Subsequently, a CD4⁺ enriched cell population (purity >92.3%; average±standard deviation 95.73±2.6%) is obtained by depletion of CD8⁺ T cells using magnetic beads (Miltenyi Biosciences), spinoculated in presence of the luciferase-expressing IMC representing HIV-1 subtype AE (CM235), B (BaL) or C (1086.C) and cultured for 72 hours. CD4⁺ infected target cells are incubated with resting CD8⁺ effector cells (isolated by negative selection from autologous PBMC, CD8⁺ T cell Isolation Kit, Miltenyi Biosciences) at 33 : 1, 11 : 1, 3 : 1, and 0: 1 E:T ratios in the absence or presence of diabodies (and/or trivalent binding molecules) for 6-48 hours at concentration ranging from 1,000 to 0.0001 ng/mL. Uninfected and infected target cells alone are included as additional controls. Each condition is tested in duplicate. After incubation, ViviRen™ Live Cell Substrate (Promega) is added and RLU measured on a luminometer; percentage specific lysis (%SL) of target cells is determined as described previously. SeePollara J et al. HIV-1 Vaccine-Induced CI and V2 Env-Specific Antibodies Synergize for Increased Antiviral Activities. J Virol.

2014;88(14):7715-7726.

[0556] T-Cell Degranulation (CD107) Assay. As described for the cytotoxicity assay with HIV-1 FMC-infected cells as targets, activated CD4⁺ cells infected with HIV-1 BaL IMC are plated with resting CD8⁺ effector cells at a 33: 1 E:T ratio in the absence or presence of 1 ng/mL diabodies and incubated for 6 hour. For the CD4 T cell degranulation, activated CD4⁺ T cells are either infected with JR-CSF and labeled with the viability (NFLl) and target specific (TFL4) markers utilized in an ADCC assay or added to targets as effectors at a 10: 1 ratio prior to addition of diabodies. Each condition is tested in duplicate. CD 107 PE-Cy5 (clone H4A3; eBioscience) is titered and added during the last six hours of the incubation along with Monensin solution (BD GolgiStop). A panel of antibodies consisting of

LIVE/DEAD Aqua stain, anti-CD3 APC-H7 (clone SK7; BD Pharmingen), anti-CD4 BV605 (clone OKT4; Biolegend), anti-CD8 BV650 (clone RPA-T8; Biolegend) are used to detect CD107⁺ CD8⁺ T cells. After washing and fixation, samples are acquired on a custom made LSRII (BD Bioscience, San Jose, CA) within the next 24 hours. A minimum of 300,000 total viable events is acquired for each test. The analysis of the data is performed using the Flow- Jo software (Treestar, Ashland, OR).

[0557] T-Cell Viability and Activation Assays. CD8⁺ T cells and CD8 depleted PBMCs obtained from HIV infected ART suppressed patients are plated at 5xl0⁴ cells per well in 96 well plates with lOOng/mL of the indicated DART. Cells are cultured in 0.2mL of cIMDM media supplemented with 10% FBS, 1% Penicillin/Streptomycin and 5U/mL IL-2 for 7 days, and then are stained with the following antibodies: HLA-DR-PerCP (clone L243), CD25-PE (clone M-A251), CD8-FITC (clone HIT 8 a), CD8-PE (clone HIT8a), CD4-FITC (clone RPA- T4), and Annexin V-PE and 7-AAD (all BD biosciences, San Jose, CA).

[0558] Redirected T-Cell Viral Clearance Assay. CD8⁺ T-cells are isolated from PBMCs by positive selection (EasySep human CD8⁺ Selection Kit, Stem Cell). CD8-depleted PBMCs are first activated with 2μg/mL of PHA (Remel, Lenexa, KS) and 60U/mL of IL-2, and then infected by spinoculation at 1200xg for 90 minutes with either JR-CSF or autologous reservoir virus (AR) at an MOI of 0.0 I). AR virus is obtained from pooled supernatants of replicate wells from outgrowth assays of resting CD4+ T-cells for each patient performed. Fifty-thousand (5xl0⁴) targets/well are co-cultured with CD8⁺ T cells in triplicate at the indicated E:T ratio in the absence or presence of 100 ng/mL of diabody in 0.2m of cIMDM media supplemented with 10% FBS, 1% Penicillin/Streptomycin and 5 U/mL IL-2. For experiments performed in the presence of antiretrovirals (ARVs), 24 hours after spinoculation cells are washed and l uM of raltegravir and 4μΜ of abacavir are added, and then diabodies and CD8⁺ T-cells are added to cultures. Supernatant is assayed on day 7 by p24 ELISA (ABL, Rockville, MD). Results are calculated as the log (p24 of infected target cells only control divided by p24 of the test condition).

[0559] Latency Clearance Assay (LCA). The reduction of virus recovery from CD4⁺ infected cells is assessed by a standard quantitative viral outgrowth assay using the resting CD4⁺ T cells of aviremic, ART-treated patients, following the addition of antiviral effector cells and/or molecules, as previously described. See Sung JA et al. Expanded Cytotoxic T-Cell Lymphocytes Target the Latent Hiv Reservoir. J Infect Dis. 2015;: 1-15. . In this case the LCA is used to model the ability of diabodies to clear virus emerging from the latent reservoir under clinically and pharmacologically relevant conditions. Resting CD4⁺ T-cells are isolated from a leukapheresis product as previously described (72) and exposed to PHA (4μg/mL) and IL-2 (60U/mL) for 24 hours or vorinostat (VOR) (335nM, 6 hours) (Merck Research

Laboratories), and plated at 0.5 to 1 x 10⁶ cells/well in 12 to 36 replicate wells depending on the size of the reservoir. The VOR is then washed off and CD8s added at an E:T of 1 : 10 as well as 100 ng/mL of the indicated DART. Cells are co-cultured for 24 hours (unless specified otherwise) following which the diabody (and/or trivalent binding molecule) proteins are washed off and allogeneic CD8-depleted PBMCs from an HIV negative donor are added to amplify residual virus. Supernatant is assayed for the presence of p24 antigen on day 15 for each well. Results are calculated as % viral recovery [(# of positive wells/total number plated)xl00], normalized to a control in which no CD8⁺ T cells are added.

[0560] Additional animal studies will be conducted to evaluate toxicity, safety, PK and PD profiles, and efficacy in preventing or treating HIV infection of the multispecific molecules of the invention.

[0561] All documents and other information sources cited herein are hereby incorporated in their entirety by reference.

Claims

What is claimed is:

1. A bispecific molecule comprising a first polypeptide chain and a second polypeptide chain, covalently bonded to one another, wherein:

(I) the first polypeptide chain comprises in the N- to C-terminal direction:

(i) a domain (A) comprising a binding region of the light chain variable domain of a first immunoglobulin (VL1) having the binding specificity of a CD4 binding site HIV-1 antibody (1);

(iii) a domain (C) comprising a heterodimer promoting domain;

(II) the second polypeptide chain comprises in the N- to C-terminal direction:

(ii) a domain (E) comprising a binding region of a heavy chain variable domain of the first immunoglobulin (VHl) having the binding specificity of a CD4 binding site HIV-1 antibody HIV-1 antibody (1); and

(iii) a domain (F) comprising a heterodimer promoting domain; and wherein:

the domains (A) and (E) associate to form a binding site that binds the HIV-1 envelope like a CD4 binding site HIV-1 antibody (1); and the domains (B) and (D) associate to form a binding site that binds the epitope (2).

2. The bispecific molecule of claim 1, wherein:

(i) domains (A) and (B) are separated by a peptide linker 1;

(ii) domains (C) and (B) are separated by a peptide linker 2;

(iii) domains (D) and (E) are separated by a peptide linker 1; and

(iv) domains (F) and (E) are separated by a peptide linker 2.

3. A bispecific molecule comprising a first polypeptide chain, a second polypeptide chain, and a third polypeptide chain, wherein the first and second polypeptide chains are covalently bonded and the second and third polypeptide chains are covalently bonded, and wherein:

(I) the first polypeptide chain comprises in the N- to C-terminal direction:

(iii) a domain (C) comprising a heterodimer promoting domain; and

(iv) a CH2-CH3 domain;

(II) the second polypeptide chain comprises in the N- to C-terminal direction:

(ii) a domain (E) comprising a binding region of a heavy chain variable domain of the first immunoglobulin (VH1) having the binding specificity of a CD4 binding site HIV-1 antibody (1); and

(iii) a domain (F) comprising a heterodimer promoting domain;

(III) the third polypeptide chain comprises in the N- to C-terminal direction:

(i) a CH2-CH3 domain, and wherein:

the domains (A) and (E) associate to form a binding site that binds the HIV-1 envelope like a CD4 binding site HIV-1 antibody (1);

4. The bispecific molecule of claim 3, wherein:

(i) the third polypeptide chain further comprises a peptide linker 3 N-terminal to the CH2-CH3 domain; (ii) domains (A) and (B) are separated by a peptide linker 1;

(iii) domains (C) and (B) are separated by a peptide linker 2;

(v) domains (D) and (E) are separated by a peptide linker 1; and

(vi) domains (F) and (E) are separated by a peptide linker 2.

5. A bispecific molecule comprising a first polypeptide chain, a second polypeptide chain and a third polypeptide chain, wherein the first and second polypeptide chains are covalently bonded and the second and third polypeptide chains are covalently bonded, and wherein:

(I) the first polypeptide chain comprises in the N- to C-terminal direction:

(i) a CH2-CH3 domain;

(ii) a domain (A) comprising a binding region of the light chain variable domain of a first immunoglobulin (VL1) having the binding specificity of a CD4 binding site HIV-1 antibody (1);

(iii) a domain (B) comprising a binding region of a heavy chain variable domain of a second immunoglobulin (VH2) specific for an epitope (2); and

(iv) a domain (C) comprising a heterodimer promoting domain;

(II) the second polypeptide chain comprises in the N- to C-terminal direction:

(iii) a domain (F) comprising a heterodimer promoting domain;

(III) the third polypeptide chain comprises in the N- to C-terminal direction:

(i) a CH2-CH3 domain, and wherein:

the domains (D) and (E) are linked so that they do not associate with one another to form an epitope binding site; the domains (A) and (E) associate to form a binding site that binds the HIV-1 envelope like a CD4 binding site HIV-1 antibody (1);

6. The bispecific molecule of claim 5, wherein:

(i) the CH2-CH3 domain and domain (A) are separated by a peptide linker 4;

(ii) domains (A) and (B) are separated by a peptide linker 1;

(iii) domains (C) and (B) are separated by a peptide linker 2;

(iv) domains (D) and (E) are separated by a peptide linker 1;

(v) domains (F) and (E) are separated by a peptide linker 2;

(vi) the first polypeptide chain further comprises a peptide linker 3 N-terminal to the CH2-CH3 domain; and

(vii) the third polypeptide chain further comprises a peptide linker 3 N-terminal to the CH2-CH3 domain.

7. A bispecific molecule comprising a first polypeptide chain, a second polypeptide chain, a third polypeptide chain, and a fourth polypeptide chain, wherein the first and second polypeptide chains are covalently bonded and the first and third polypeptide chains are covalently bonded, and wherein:

(i) a domain (A) comprising a binding region of the light chain variable domain of a first immunoglobulin (VLl) having the binding specificity of a CD4 binding site HIV-1 antibody (1);

(iii) a domain (C) comprising a heterodimer promoting domain; and

(iv) a CH2-CH3 domain;

(i) a domain (D) comprising a binding region of a light chain variable domain of the second immunoglobulin (VL2) specific for the epitope (2); (ii) a domain (E) comprising a binding region of a heavy chain variable domain of the first immunoglobulin (VHl) having the binding specificity of a CD4 binding site HIV-1 antibody (1);

(iii) a domain (F) comprising a heterodimer promoting domain; and wherein the domains (A) and (B) are linked so that they do not associate with one another to form an epitope binding site;

8. The bispecific molecule of claim 7, wherein:

(i) domains (A) and (B) are separated by a peptide linker 1;

(ii) domains (C) and (B) are separated by a peptide linker 2;

(iv) domains (D) and (E) are separated by a peptide linker 1; and

(v) domains (F) and (E) are separated by a peptide linker 2.

9. A trivalent binding molecule comprising a first, second, third and fourth polypeptide chain wherein:

(iii) a domain (C) comprising a heterodimer promoting domain; and

(iv) a CH2-CH3 Domain;

(i) a domain (D) comprising a binding region of a light chain variable domain of the first immunoglobulin (VLl) specific for the epitope (2); (ii) a domain (E) comprising a binding region of a heavy chain variable domain of the second immunoglobulin (VH1) specific for the epitope (1);

(iii) a domain (F) comprising a heterodimer promoting domain; and

(ii) a CHI -Hinge Domain, and a CH2-CH3 Domain; and

(ii) CL Kappa Domain or a CL Lambda Domain; and wherein

at least one of epitope (1), epitope (2), and epitope (3) is an epitope bound by an CD4 binding site HIV- 1 antibody, and at least one of epitope (1), epitope (2), and epitope (3) is an epitope of CD3, CD 8 or CD16;

the CH2-CH3 domains of the first and third polypeptide form an Fc Domain;

said first and second polypeptide chains are covalently bonded to one another;

said third and fourth polypeptide chains are covalently bonded to one another.

10. The trivalent binding molecule of claim 9, wherein:

(i) domains (A) and (B) are separated by a peptide linker 1;

(iv) domains (D) and (E) are separated by a peptide linker 1; and

11. A trivalent binding molecule comprising a first, second, and third polypeptide chain wherein:

(iii) a domain (C) comprising a heterodimer promoting domain; and

(iv) a CH2-CH3 Domain;

(iii) a domain (F) comprising a heterodimer promoting domain; and

(iii) a CH2-CH3 Domain; and wherein

at least one of epitope (1), epitope (2), and epitope (3) is an epitope bound by a CD4 binding site HIV-1 antibody, and at least one of epitope (1), epitope (2), and epitope (3) is and epitope of CD3, CD8, or CD16;

the CH2-CH3 domains of the first and third polypeptide form an Fc Domain; said first and second polypeptide chains are covalently bonded to one another; said first and third polypeptide chains are covalently bonded to one another; and

said third and fourth polypeptide chains are covalently bonded to one another.

12. The tnvalent binding molecule of claim 11, wherein:

(i) domains (A) and (B) are separated by a peptide linker 1;

(iv) domains (D) and (E) are separated by a peptide linker 1;

(vi) domains (H) and (G) are separated by a peptide linker 5; and

(vii) the CH2-CH3 domain and domain (G) are separated by a peptide linker 3.

13. The tnvalent binding molecule of claims 9-12, wherein one of epitope (1), epitope (2), or epitope (3) is an epitope of CD8.

14. The molecules of claims 3-8 and 9-12, wherein the CH2-CH3 domain of the first

polypeptide chain is the of the "knob" design (SEQ ID NOs: 531 or 532) and the CH2- CH3 domain of the third polypeptide chain is of the "hole" design (SEQ ID NOs: 533 or 534).

15. The molecules of claim 14, wherein the CH2-CH3 domain of the first polypeptide

comprises SEQ ID NO: 531 and the CH2-CH3 domain of the third polypeptide chain comprises SEQ ID NO: 533.

16. The molecules of claims 3-8 and 9-12, wherein the CH2-CH3 domain of the third

polypeptide chain is the of the "knob" design (SEQ ID NOs: 531 or 532) and the CH2- CH3 domain of the first polypeptide chain is of the "hole" design (SEQ ID NOs: 533 or 534).

17. The molecules of claim 16, wherein the CH2-CH3 domain of the third polypeptide

comprises SEQ ID NO: 531 and the CH2-CH3 domain of the first polypeptide chain comprises SEQ ID NO: 533.

18. The molecules of claims 1-12 wherein the epitope (2) is a CD3 epitope, CD8 epitope, or a CD 16 epitope.

19. The molecules of claims 1-12, wherein the CD4 binding site HIV-1 antibody is CH557 (also referred to as CH235.12) or any of the antibodies of the CH235 lineage.

20. The molecules of claims 1-12, wherein the molecule binds HIV-1 envelope with the specificity of CH557 antibody and binds CD3, CD8, or CD16.

21. The molecules of claims 1-12, wherein domain (A) comprises the CDR1, CDR2, and CDR3 of the light chain variable domain of immunoglobulin CH557, CH556, CH555, CH493, CH492, CH491, or CH490.

22. The molecules of claims 1-12, wherein domain (E) comprises the CDR1, CDR2, and CDR3 of the heavy chain variable domain of immunoglobulin CH557, CH556, CH555, CH493, CH492, CH491, or CH490.

23. The molecules of claims 1-12, wherein domain (A) comprises the light chain variable domain of immunoglobulin CH557, CH556, CH555, CH493, CH492, CH491, or CH490.

24. The molecules of claims 1-12, wherein domain (E) comprises the heavy chain variable domain of immunoglobulin CH557, CH556, CH555, CH493, CH492, CH491, or CH490.

25. The molecules of claims 1-12, wherein domain (B) comprises the heavy chain variable domain of an anti-CD3 antibody.

26. The molecules of claims 1-12, wherein the domain (D) comprises the light chain variable domain of an anti-CD3 antibody.

27. The trivalent binding molecule of claims 9-12, wherein domain (G) comprises heavy chain variable domain of an anti-CD8 antibody.

28. The trivalent binding molecule of claims 9-12, wherein domain (H) comprises light chain variable domain of an anti-CD8 antibody.

29. The bispecific molecule of claim 3, wherein the first polypeptide comprises SEQ ID NO: 555, the second polypeptide comprises SEQ ID NO: 557, and the third polypeptide comprises SEQ ID NO: 559.

30. The tnvalent binding molecule of claim 9, wherein the first polypeptide comprises SEQ ID NO: 555, the second polypeptide comprises SEQ ID NO: 557, the third polypeptide comprises SEQ ID NO: 561, and the fourth polypeptide comprises SEQ ID NO: 562.

31. The trivalent binding molecule of claim 11, wherein the first polypeptide comprises SEQ ID NO: 555, the second polypeptide comprises SEQ ID NO: 557, and the third

polypeptide comprises SEQ ID NO: 563.

32. A composition comprising any one of the molecules of claims 1-3 lor any combination thereof and a carrier.

33. A composition comprising a bispecific molecule or trivalent molecule which comprises at least one arm with the binding specificity of HIV-1 antibody CH557, and a second arm targeting CD3, CD8 or CD 16.

34. The composition of claim 33, further comprising a second bispecific molecule or trivalent molecule comprising a first arm with the binding specificity of HIV-1 antibody different from the binding specificity of the first bispecific molecule or trivalent molecule and a second arm targeting CD3, CD8 or CD 16, wherein the first and secondmolecules are different.

35. A method to treat or prevent HIV-1 infection in a subject in need thereof comprising

administering to the subject a composition comprising any one of the molecules of claim 1-31 or a combination of any one of these molecules in a therapeutically effective amount.

36. The method of claim 35, further comprising administering a latency activating agent.

37. The method of claim 36, wherein the latency activating agent is vorinostat, romidepsin, panobinostat, disulfiram, JQ1, bryostatin, PMA, inonomycin, or any combination thereof.

38. A vector comprising nucleic acids comprising nucleotides encoding the molecules of any one of claim 1-31.

39. A composition comprising a vector comprising a nucleic acid encoding the molecules of any one of claim 1-31.