EA046760B1 - Canine distemper virus hemagglutinin and fusion polypeptides - Google Patents

Description

Cross reference to related applications

This application claims priority to U.S. Provisional Patent Application No. 62/913,111, filed October 9, 2019. The description of the prior application is deemed to be part of the specification of (and is incorporated by reference) the present application.

Field of technology to which the invention relates

This document relates to canine distemper virus (CDV) hemagglutinin (H) and fusion (F) polypeptides. For example, this document relates to CDV H polypeptides, CDV F polypeptides, recombinant viruses (eg, vesicular stomatitis viruses (VSV)) containing CDV H polypeptides and/or CDV F polypeptides, nucleic acid molecules encoding CDV H polypeptide and/or F CDV polypeptide, methods of producing recombinant viruses (eg, VSV) containing CDV H polypeptides and/or CDV F polypeptides, and methods of using recombinant viruses (eg, VSV) containing CDV H polypeptides and/or CDV F polypeptides for the treatment of cancer or infectious diseases.

State of the Art Information

Viruses such as VSV, measles viruses (MeV) and adenoviruses can be used as oncolytic viruses to treat cancer. Vesicular stomatitis virus (VSV) is a member of the Rhabdoviridae family. The VSV genome is a single negative RNA molecule encoding five major polypeptides: nucleocapsid polypeptide (N), phosphoprotein polypeptide (P), matrix (M) polypeptide, glycoprotein polypeptide (G), and viral polymerase polypeptide (L).

The essence of the invention

Provided herein are methods and materials related to CDV H and/or F polypeptides. For example, provided herein are CDV H polypeptides, CDV F polypeptides, recombinant viruses (eg, vesicular stomatitis viruses (VSV)) containing CDV H polypeptides and/or CDV F polypeptides, nucleic acid molecules encoding CDV H polypeptide and/or F CDV polypeptide, methods for producing recombinant viruses (eg, VSV) containing CDV H polypeptides and/or CDV F polypeptides, and methods of using recombinant viruses (eg, VSV) containing CDV H polypeptides and/or CDV F polypeptides for the treatment of cancer or infectious diseases.

As described herein, CDV F polypeptides can be engineered to have increased fusogenic activity when expressed by cells in combination with a CDV H polypeptide compared to the level of fusogenic activity of a wild-type CDV F polypeptide expressed by such cells in combination with such a CDV H polypeptide. For example, engineered CDV F polypeptides having a truncated signal peptide sequence may exhibit increased fusogenic activity when expressed by cells in combination with a CDV H polypeptide (e.g., a wild-type CDV H polypeptide or a CDV H polypeptide lacking specificity) compared to the level of fusogenic activity of F a wild-type CDV polypeptide containing the full-length signal peptide sequence expressed by such cells in combination with such a CDV H polypeptide. Such CDV F polypeptides can be incorporated into a virus to produce a recombinant virus that has the ability to enhance the fusogenic activity observed in cells infected by such a virus.

As also described herein, CDV H polypeptides can be engineered to lack specificity such that they do not have the ability, when used in combination with an F polypeptide (e.g., CDV F polypeptide), to enter cells or fuse with cells through the polypeptide Nectin 4, a SLAMF1 polypeptide or viral receptor present on wild-type Vero cells. Such CDV H polypeptides may provide a platform for the generation of H polypeptides having the ability to redirect specificity to one or more targets of interest. For example, an H polypeptide provided herein may be further modified to include a binding sequence (e.g., a single chain antibody (scFv) sequence) having binding specificity for a target of interest, resulting in a recombinant virus containing such an H polypeptide with a redirected specificity and F polypeptide will be able to infect cells expressing such a target.

In addition, VSV viruses can be engineered to contain a nucleic acid molecule encoding a VSV N polypeptide, a VSV P polypeptide, a VSV M polypeptide, a CDV F polypeptide (e.g., wild-type CDV F polypeptide or the recombinant CDV F polypeptide described herein ), CDV H polypeptide (e.g., wild-type CDV H polypeptide or recombinant CDV H polypeptide described herein), and VSV L polypeptide. Such a nucleic acid molecule may not contain a functional VSV G polypeptide and/or may not contain a nucleic acid sequence that encodes a full-length VSV G polypeptide. For example, the VSV provided herein can be engineered to contain a nucleic acid molecule that encodes a VSV N polypeptide, a VSV P polypeptide,

- 1 046760

VSV M polypeptide, CDV F polypeptide (e.g., wild-type CDV F polypeptide or recombinant CDV F polypeptide described herein), CDV H polypeptide (e.g., wild-type CDV H polypeptide or recombinant CDV H polypeptide described herein), and VSV L polypeptide, and does not have the ability to encode a functional VSV G polypeptide. In some cases, the VSV provided herein may be engineered to contain a nucleic acid molecule encoding a VSV N polypeptide, a VSV P polypeptide, a VSV M polypeptide, a CDV F polypeptide (e.g., a wild-type CDV F polypeptide or a recombinant F polypeptide CDV H polypeptide described herein), a CDV H polypeptide (e.g., a wild-type CDV H polypeptide or a recombinant CDV H polypeptide described herein), and a VSV L polypeptide, with a nucleic acid sequence encoding a CDV F polypeptide and a CDV H polypeptide that is located at the position at which the nucleic acid sequence encoding the full-length VSV G polypeptide is normally located in wild-type VSV. In some cases, the VSV provided herein may be engineered to contain a nucleic acid molecule wherein the nucleic acid sequence encoding the VSV G polypeptide is replaced with a nucleic acid encoding the CDV F polypeptide (e.g., wild-type or recombinant CDV F polypeptide CDV F polypeptide described herein) and CDV H polypeptide (eg, wild-type CDV H polypeptide or recombinant CDV H polypeptide described herein).

As described herein, VSV/CDV hybrids can be engineered to exhibit CDV selectivity and rapid replication, as observed with wild-type or parental VSV. In some cases, the VSV/CDV hybrid presented herein can be engineered to have a preselected tropism. For example, CDV F and/or H polypeptides with knockout specificity for nectin-4 and/or SLAMF1 can be used. In such cases, the scFv or polypeptide ligand may be attached, for example, to the C-terminal H of the CDV polypeptide. In such cases, the scFv or polypeptide ligand may determine the tropism of the VSV/CDV hybrid. Examples of scFvs that can be used to target VSV/CDV hybrids to cellular receptors (eg, tumor-associated cell receptors) include, but are not limited to, anti-EGFR, anti-oFR, anti-CD46, anti-CD38, anti-HER2/neu, anti-EpCAM, anti- CEA, anti-CD20, anti-CD133, anti-CD117 (c-kit) and anti-CD138 and anti-PSMA. Examples of polypeptide ligands that can be used to target VSV/CDV hybrids include, but are not limited to, urokinase plasminogen activator (UPA) polypeptides, cytokines such as IL-13 or IL-6, single chain T cell receptors (scTCR), echistatin polypeptides , stem cell factor (SCF), EGF and integrin-binding polypeptides.

In some cases, the VSV/CDV hybrid provided herein may have a nucleic acid molecule that includes a sequence encoding an interferon (IFN) polypeptide (e.g., a human IFN-β polypeptide), a sodium iodide symporter (NIS) polypeptide (e.g., a human NIS), fluorescent polypeptide (eg, GFP polypeptide), any suitable therapeutic transgene (eg, HSV-TK or cytosine deaminase), polypeptide that weakens host immunity (eg, influenza virus NS1, HSV γ34.5, or SOCS1), or tumor antigen (for example, components of a cancer vaccine). The nucleic acid encoding the IFN polypeptide may be located between the nucleic acid encoding the VSV M polypeptide and the nucleic acid encoding the VSV L polypeptide. This arrangement may allow viruses to express an amount of IFN polypeptide that is effective in activating antiviral innate immune responses in noncancerous tissues and thus attenuating potential viral toxicity without interfering with efficient viral replication in cancer cells. The nucleic acid encoding the NIS polypeptide may be located between the nucleic acid encoding the VSV M polypeptide and the VSV L polypeptide. This arrangement may allow viruses to express an amount of NIS polypeptide that is: (a) effective in selectively accumulating iodide in infected cells, allowing viral spread to be visualized using radioisotopes and radiation therapy directed at infected cancer cells, and (b) not so high to be toxic to infected cells. The location of the nucleic acid encoding the IFN polypeptide between the nucleic acid encoding the VSV M polypeptide and the nucleic acid encoding the VSV L polypeptide, and the location of the nucleic acid encoding the NIS polypeptide between the nucleic acid encoding the VSV M polypeptide and the VSV L polypeptide in the genome VSV can result in VSVs that are viable, capable of replication and propagation, express appropriate levels of functional IFN polypeptides, and express appropriate levels of functional NIS polypeptides to capture radioiodine for both imaging and radiation virotherapy.

- 2 046760

Generally, one aspect of this document describes a CDV F polypeptide having a signal peptide sequence that is less than 75 amino acid residues in length. The signal peptide sequence may include no more than 75 amino acid residues of SEQ ID NO: 6. The CDV F polypeptide may include SEQ ID NO: 4 provided that the CDV F polypeptide is absent at least amino acid residues 1-60, or is absent, according to at least amino acid residues 1-105 SEQ ID NO: 4. A recombinant virus comprising CDV F polypeptide and CDV H polypeptide may exhibit increased fusogenic activity compared to a comparable control recombinant virus comprising full-length CDV F polypeptide and wild-type CDV H polypeptide .

In yet another embodiment, provided herein is a nucleic acid molecule encoding a CDV F polypeptide. The CDV F polypeptide may have a signal peptide sequence that is less than 75 amino acid residues in length. The signal peptide sequence may include no more than 75 amino acid residues of SEQ ID NO: 6. The CDV F polypeptide may include SEQ ID NO: 4 provided that the CDV F polypeptide is absent at least amino acid residues 1-60, or is absent, according to at least amino acid residues 1-105 SEQ ID NO: 4. A recombinant virus comprising CDV F polypeptide and CDV H polypeptide may exhibit increased fusogenic activity compared to a comparable control recombinant virus comprising full-length CDV F polypeptide and wild-type CDV H polypeptide .

In another embodiment, provided herein is a recombinant virus comprising a CDV F polypeptide. The CDV F polypeptide may have a signal peptide sequence that is less than 75 amino acid residues in length. The signal peptide sequence may include no more than 75 amino acid residues of SEQ ID NO: 6. The CDV F polypeptide may include SEQ ID NO: 4 provided that the CDV F polypeptide is absent at least amino acid residues 1-60, or is absent, according to at least amino acid residues 1-105 SEQ ID NO: 4. A recombinant virus comprising CDV F polypeptide and CDV H polypeptide may exhibit increased fusogenic activity compared to a comparable control recombinant virus comprising full-length CDV F polypeptide and wild-type CDV H polypeptide .

In another embodiment, provided herein is a recombinant virus comprising a nucleic acid molecule. The nucleic acid molecule may encode a CDV F polypeptide. The CDV F polypeptide may have a signal peptide sequence that is less than 75 amino acid residues in length. The signal peptide sequence may include no more than 75 amino acid residues of SEQ ID NO: 6. The CDV F polypeptide may include SEQ ID NO: 4 provided that the CDV F polypeptide is absent at least amino acid residues 1-60, or is absent, according to at least amino acid residues 1-105 SEQ ID NO: 4. A recombinant virus comprising CDV F polypeptide and CDV H polypeptide may exhibit increased fusogenic activity compared to a comparable control recombinant virus comprising full-length CDV F polypeptide and wild-type CDV H polypeptide .

In another embodiment, provided herein is a CDV H polypeptide comprising 454A, 464A, 479A, 494A, 510A, 520A, 525A, 526A, 527S, 528A, 529A, 537A, 539A, 547A, 548A, or a combination thereof according to amino acid numbering in SEQ ID NO: 2. The CDV H polypeptide may include a combination of two, three, four, five or six of 454A, 464A, 479A, 494A, 510A, 520A, 525A, 526A, 527S, 528A, 529A, 537A, 539A, 547A and 548A. The CDV H polypeptide may include a combination of seven, eight, nine, ten, or eleven of 454A, 464A, 479A, 494A, 510A, 520A, 525A, 526A, 527S, 528A, 529A, 537A, 539A, 547A, and 548A. The CDV H polypeptide may include a combination of 12, 13, or 14 of 454A, 464A, 479A, 494A, 510A, 520A, 525A, 526A, 527S, 528A, 529A, 537A, 539A, 547A, and 548A. The CDV H polypeptide may include 454A, 464A, 479A, 494A, 510A, 520A, 525A, 526A, 527S, 528A, 529A, 537A, 539A, 547A, and 548A. The CDV H polypeptide may include M437 according to amino acid numbering in SEQ ID NO: 5.

In yet another embodiment, provided herein is a CDV H polypeptide comprising the sequence shown in FIG. 11, except that said sequence includes a mutation of the represented amino acid residue selected from the group consisting of P454, V/L/F460, L/F/W479, I494, I/L/V510, Y520, Y/N525, D /G526, I/V527, S/T528, R529, Y/D537, Y539, Y/F547 and T/M548, in accordance with the amino acid numbering in SEQ ID NO: 5. The CDV H polypeptide may include a mutation of two, three, four , five or six amino acid residues represented, selected from the group. The CDV H polypeptide may include a mutation of seven, eight, nine, ten or eleven amino acid residues present, selected from a group. The CDV H polypeptide may include a mutation at 12, 13 or 14 of the present amino acid residues selected from the group. The CDV H polypeptide may include mutation of the presented amino acid residues from the group. The CDV H polypeptide may include M437 according to amino acid numbering in SEQ ID NO: 5.

- 3 046760

In another embodiment, provided herein is a nucleic acid molecule encoding a CDV H polypeptide. The CDV H polypeptide may include the sequence shown in FIG. 11, except that said sequence includes a mutation of the represented amino acid residue selected from the group consisting of P454, V/L/F460, L/F/W479, I494, I/L/V510, Y520, Y/N525, D /G526, I/V527, S/T528, R529, Y/D537, Y539, Y/F547 and T/M548, in accordance with the amino acid numbering in SEQ ID NO: 5. The CDV H polypeptide may include a mutation of two, three, four , five or six amino acid residues represented, selected from the group. The CDV H polypeptide may include a mutation of seven, eight, nine, ten or eleven amino acid residues present, selected from a group. The CDV H polypeptide may include a mutation at 12, 13 or 14 of the present amino acid residues selected from the group. The CDV H polypeptide may include mutation of the presented amino acid residues from the group. The CDV H polypeptide may include M437 according to amino acid numbering in SEQ ID NO: 5.

In another embodiment, provided herein is a recombinant virus comprising a CDV H polypeptide. CDV H polypeptide comprising the sequence shown in FIG. 11, except that said sequence includes a mutation of the represented amino acid residue selected from the group consisting of P454, V/L/F460, L/F/W479, I494, I/L/V510, Y520, Y/N525, D /G526, I/V527, S/T528, R529, Y/D537, Y539, Y/F547 and T/M548, in accordance with the amino acid numbering in SEQ ID NO: 5. The CDV H polypeptide may include a mutation of two, three, four , five or six amino acid residues represented, selected from the group. The CDV H polypeptide may include a mutation of seven, eight, nine, ten or eleven amino acid residues present, selected from a group. The CDV H polypeptide may include a mutation at 12, 13 or 14 of the present amino acid residues selected from the group. The CDV H polypeptide may include mutation of the presented amino acid residues from the group. The CDV H polypeptide may include M437 in accordance with the amino acid numbering in SEQ ID NO: 5. The virus may include a CDV F polypeptide, wherein the CDV F polypeptide includes a signal peptide sequence that is less than 75 amino acid residues in length. The signal peptide sequence may include no more than 75 amino acid residues of SEQ ID NO: 6. The CDV F polypeptide may include SEQ ID NO: 4 provided that the CDV F polypeptide is absent at least amino acid residues 1-60, or is absent, according to at least amino acid residues 1-105 SEQ ID NO: 4. A recombinant virus comprising CDV F polypeptide and CDV H polypeptide may exhibit increased fusogenic activity compared to a comparable control recombinant virus comprising full-length CDV F polypeptide and wild-type CDV H polypeptide .

In another embodiment, provided herein is a recombinant virus comprising a nucleic acid molecule encoding a CDV H polypeptide. The CDV H polypeptide may include the sequence shown in FIG. 11, except that said sequence includes a mutation of the represented amino acid residue selected from the group consisting of P454, V/L/F460, L/F/W479, I494, I/L/V510, Y520, Y/N525, D /G526, I/V527, S/T528, R529, Y/D537, Y539, Y/F547 and T/M548, in accordance with the amino acid numbering in SEQ ID NO: 5. The CDV H polypeptide may include a mutation of two, three, four , five or six amino acid residues represented, selected from the group. The CDV H polypeptide may include a mutation of seven, eight, nine, ten or eleven amino acid residues present, selected from a group. The CDV H polypeptide may include a mutation at 12, 13 or 14 of the present amino acid residues selected from the group. The CDV H polypeptide may include mutation of the presented amino acid residues from the group. The CDV H polypeptide may include M437 in accordance with the amino acid numbering in SEQ ID NO: 5. The virus may include a nucleic acid molecule encoding the CDV F polypeptide. The CDV F polypeptide may have a signal peptide sequence that is less than 75 amino acid residues in length. The signal peptide sequence may include no more than 75 amino acid residues of SEQ ID NO: 6. The CDV F polypeptide may include SEQ ID NO: 4 provided that the CDV F polypeptide is absent at least amino acid residues 1-60, or is absent, according to at least amino acid residues 1-105 SEQ ID NO: 4. A recombinant virus comprising CDV F polypeptide and CDV H polypeptide may exhibit increased fusogenic activity compared to a comparable control recombinant virus comprising full-length CDV F polypeptide and wild-type CDV H polypeptide .

In another embodiment, provided herein is a recombinant virus described herein that is a hybrid virus of (a) CDV and (b) VSV, MeV, or adenovirus.

In another embodiment, provided herein is a replication-competent vesicular stomatitis virus comprising an RNA molecule, wherein the RNA molecule includes a nucleic acid sequence that is a template for a positive sense transcript encoding a VSV N polypeptide, a nucleic acid sequence that

- 4 046760 is a template for a positive sense transcript encoding a VSV P polypeptide, a nucleic acid sequence that is a template for a positive sense transcript encoding a VSV M polypeptide, a nucleic acid sequence that is a template for a positive sense transcript encoding an F polypeptide CDV, a nucleic acid sequence that is a template for a positive sense transcript encoding a CDV H polypeptide, and a nucleic acid sequence that is a template for a positive sense transcript encoding a CDV L polypeptide, wherein the RNA molecule lacks a nucleic acid sequence that is template for a positive-sense transcript encoding a functional VSV G polypeptide. The CDV F polypeptide may have a signal peptide sequence that is less than 75 amino acid residues in length. The signal peptide sequence may comprise no more than 75 amino acid residues of SEQ ID NO: 6. The CDV F polypeptide may comprise SEQ ID NO: 4 provided that the CDV F polypeptide lacks at least amino acid residues 1-60 or is absent at least At least amino acid residues 1-105 SEQ ID NO: 4. A recombinant virus comprising a CDV F polypeptide and a CDV H polypeptide may exhibit increased fusogenic activity compared to a comparable control recombinant virus comprising a full-length CDV F polypeptide and a wild-type CDV H polypeptide. The CDV H polypeptide may be the CDV H polypeptide described in one of the previous paragraphs. The CDV H polypeptide may comprise the amino acid sequence of a single chain antibody. The single chain antibody may be a single chain antibody directed against CD19, CD20, CD38, CD46, EGFR, oFR, HER2/neu or PSMA. The RNA molecule may include a nucleic acid sequence that is a template for a positive sense transcript encoding an NIS polypeptide.

In another embodiment, provided herein is a composition comprising a virus according to any of the preceding paragraphs.

In another embodiment, provided herein is a nucleic acid molecule comprising a nucleic acid strand including a nucleic acid sequence that is a template for a positive sense transcript encoding a VSV N polypeptide, a nucleic acid sequence that is a template for a positive sense transcript encoding VSV P polypeptide, a nucleic acid sequence that is the template for the positive sense transcript encoding the VSV M polypeptide, a nucleic acid sequence that is the template for the positive sense transcript encoding the CDV F polypeptide, a nucleic acid sequence that is the template for the positive a sense transcript encoding a CDV H polypeptide, and a nucleic acid sequence that is a template for a positive sense transcript encoding a VSV L polypeptide, wherein the nucleic acid chain does not contain a nucleic acid sequence that is a template for a positive sense transcript encoding a functional G polypeptide VSV. The CDV F polypeptide may have a signal peptide sequence that is less than 75 amino acid residues in length. The signal peptide sequence may include no more than 75 amino acid residues of SEQ ID NO: 6. The CDV F polypeptide may include SEQ ID NO: 4 provided that the CDV F polypeptide is absent at least amino acid residues 1-60, or is absent, according to at least amino acid residues 1-105 SEQ ID NO: 4. A recombinant virus comprising CDV F polypeptide and CDV H polypeptide may exhibit increased fusogenic activity compared to a comparable control recombinant virus comprising full-length CDV F polypeptide and wild-type CDV H polypeptide . The CDV H polypeptide may be the CDV H polypeptide described in one of the previous paragraphs. The CDV H polypeptide may comprise the amino acid sequence of a single chain antibody. The single chain antibody may be a single chain antibody directed against CD19, CD20, CD38, CD46, EGFR, oFR, HER2/neu or PSMA. The RNA molecule may include a nucleic acid sequence that is a template for a positive sense transcript encoding an NIS polypeptide.

In another embodiment, provided herein is a composition comprising a nucleic acid molecule according to any of the preceding paragraphs.

In another embodiment, a method for treating cancer is provided herein. The method includes administering a composition described herein (eg, a composition containing a virus described herein) to a mammal comprising cancer cells, wherein the number of cancer cells in the mammal is reduced following administration. A mammal can be a human. Cancer can be myeloma, melanoma, glioma, lymphoma, mesothelioma, lung cancer, brain cancer, stomach cancer, colon cancer, rectal cancer, kidney cancer, prostate cancer, ovarian cancer, breast cancer, pancreatic cancer, cancer liver or head and neck cancer.

- 5 046760

In another embodiment, provided herein is a method for inducing tumor regression in a mammal. The method includes administering a composition described herein (eg, a composition containing a virus described herein) to a mammal comprising a tumor, wherein the size of the tumor decreases upon administration. A mammal can be a human. Cancer can be myeloma, melanoma, glioma, lymphoma, mesothelioma, lung cancer, brain cancer, stomach cancer, colon cancer, rectal cancer, kidney cancer, prostate cancer, ovarian cancer, breast cancer, pancreatic cancer, cancer liver or head and neck cancer.

In another embodiment, this document provides a method for isolating replication-competent vesicular stomatitis viruses from cells. Vesicular stomatitis viruses include an RNA molecule comprising a nucleic acid sequence that is a template for a positive sense transcript encoding a VSV N polypeptide, a nucleic acid sequence that is a template for a positive sense transcript encoding a VSV P polypeptide, a nucleic acid sequence that is a template for a positive sense transcript encoding a VSV M polypeptide, a nucleic acid sequence that is a template for a positive sense transcript encoding a CDV F polypeptide, a nucleic acid sequence that is a template for a positive sense transcript encoding a CDV H polypeptide, and a sequence a nucleic acid that is a template for a positive sense transcript encoding a CDV L polypeptide, wherein the RNA molecule does not contain a nucleic acid sequence that is a template for a positive sense transcript encoding a functional VSV G polypeptide. The method includes: (a) introducing a nucleic acid encoding an RNA molecule into cells under conditions under which replication-competent vesicular stomatitis viruses are produced, and (b) collecting replication-competent vesicular stomatitis viruses.

Unless otherwise specified, all technical and scientific terms used in this description have the same meaning as commonly known to one skilled in the art to which this invention relates. Although methods and materials similar or equivalent to those described herein may be used to practice the invention, suitable methods and materials are described below. All publications, patent applications, patents and other sources cited herein are incorporated by reference in their entirety. In the event of any conflict, this specification, including definitions, will control. In addition, the materials, methods, and examples are for illustrative purposes only and should not be construed as limitation.

Details of one or more embodiments of the invention are set forth in the accompanying drawings and description below. Other features, objects and advantages of the invention will be apparent from the description and drawings, as well as from the claims.

Description of drawings

Fig. 1. Molecular phylogenetic analysis of the hemagglutinin gene of CDV genotypes using the maximum likelihood method. The tree was constructed using the maximum likelihood method based on the Generalized Time-Invertible Model. The analysis included 119 complete CDV H sequences. Strains 22458/15 and 5804 were classified as Arctic-like and Europe-1/South America 1 genotypes, respectively. Evolutionary studies were carried out in MEGA7.

Fig. 2. Preparation of H/F CDV complexes for targeted cell fusion. (A) Syncytia formation assay: Monolayers of Vero cells and Vero cells expressing human nectin-4 polypeptides or canine SLAMF1 polypeptides were cotransfected with expression plasmids encoding the H and F polypeptide from CDV5804, CDV22458/16, or CDVOnderstepoort (forming large plaques) (CDV _OL ) as stated. The formation of syncytia was recorded 24 hours after Giemsa staining. (B) Cells were further transfected with a plasmid expressing GFP to enhance sensitivity. DAPI was used to stain nuclei. (C) Schematic representation of the receptor-exposed morbillivirus attachment protein. The receptor-binding protein contained a cytoplasmic tail (C), a transmembrane domain (T), and an ectodomain fused to a CD38-specific single chain variable fragment (scFv) followed by a His tag. (D) Surface expression of receptor-exposed morbillivirus attachment proteins. HEK293T cells were transfected with the indicated attachment proteins with redirected specificity against the CD38 receptor, and surface expression was examined by FACS using PE-conjugated anti-HIS antibody. (E) Schematic representation of a quantitative fusion assay based on a self-associating split-luciferase assay. Effector cells were transfected with plasmids encoding H and F polypeptides and plasmids encoding half of Renilla luciferase (RL) and green fluorescent proteins (GFP). Target cells carrying receptors were transfected with plasmids encoding the other half

- 6,046,760 split reporter genes into two parts and mixed with effector cells. When the cell contents were mixed, the nonfunctional halves of the reporter genes restored enzyme activity, which was measured in real time. (F) Cell-targeted fusion activity of CD38 receptor-exposed morbillivirus attachment proteins. Effector cells were cotransfected with the indicated addition/fusion polypeptide pairs. After co-cultivation with target cells, the luminescent signal was recorded. Values and error bars (SD) were taken from one representative experiment performed in at least three replicates. It should be noted that the CDV F signal peptide was replaced by a homologous form from MEV (SP MeV). Statistical significance was calculated using two-way analysis of variance with Tukey's multiple comparison test. *, p<0.02; ***, p<0.0005; ****, p<0.0001. (G) Integrity of CDV H polypeptides. HEK293T cells were transfected with the indicated CDV H polypeptides, and cell lysates were immunoblotted with anti-HIS (H CDV) or anti-β-actin (application control) antibodies. (H) Cell-targeted fusion of H/F complexes with redirected specificity. Cell fusion was determined as described in (C). Effector cells were cotransfected with receptor-nonselective pairs of MeV H/F and CDV H5804/F22458/16 with redirected CD38 specificity and layered onto CHO cells expressing the corresponding receptors. Values and error bars (SD) were taken from one representative experiment performed in at least three replicates. (I) Redirection of CDV H/F complex specificity to Her2/neu. Cell-cell fusion assays were performed using CDV H/F complexes exhibiting Her2/neu-specific scFvs or affibodies (ZX). Binding affinities for the exposed ligand are shown. (J) Relationship between receptor expression and receptor binding affinity. The fusion activity of CDV H polypeptides exhibiting Her2/neu-specific affibodies with different affinities was assessed on Her2/neu-positive cell lines: HT1080 (3.4/103 molecules/cell), Sko3pi (4.19/103 molecules/cell) , TET67L (1.5/10 ⁵ molecules/cell). Cells transfected with CDV F alone were used as a negative control.

Fig. 3. H/F complexes from CDV can replace measles virus envelope glycoproteins, leading to cell targeting and resistance to measles virus antibodies. (A) Cell surface expression. CHO cells were transfected with the indicated H polypeptide expression plasmid (HIS-tagged), together with one of the DSP plasmids and an F polypeptide expression plasmid (MeV or CDV F224568/16). After 24 h, H expression was determined by CELISA with anti-HIS antibodies. (B) CHO cells transfected as in (A) were cocultured with the indicated resulting CHO cells, and luminescence values were obtained over a 9-hour period. Values and error bars (SD) were taken from one representative experiment performed in at least three replicates. (C) Protein composition of measles virus encoding CDV H/F complexes directed against CD46 (scFv A09, Stealth 2.0). 1.6E4 TCID ₅₀ particles were subjected to SDS-PAGE and immunoblotting with the appropriate antibody. Measles virus was used as a control. (D) Replication kinetics. Growth kinetics of Stealth 2.0 virus were determined on Vero and Vero-HIS (multiplicity of infection (MOI)=0.03) at the indicated time points. MeV growth kinetics on Vero/hSLAMF1 cells were included for comparison purposes. (E) Neutralization assay. Fluorescence neutralization assay for focus suppression was performed with antisera to MeV and CDV. Different dilutions of antiserum were preincubated for 1 h at 37°C with a fixed amount of virus. The mixture was then used to infect Vero-HIS cells, and the amount of virus in control wells without antibodies was set to 100%. All neutralization curves represent the mean and standard deviation of four replicate curves obtained on a single 96-well plate. (F) Receptor selectivity of recombinant viruses. CHO cells expressing the appropriate receptors were infected with recombinant viruses at the indicated multiplicity of infection. GFP autofluorescence was recorded after 2 days.

Fig. 4. Oncolytic activity of MeV with CD46-targeted chimeric envelope proteins. CDB17 SCID mice were subcutaneously implanted with the human myeloma cell line U266.B1. When the tumor volume reached 500 mm3, mice were randomized and left without treatment (PBS control) or received a single dose of 1/10 ⁵ TCID ⁵⁰ particles intravenously. Tumor volumes (B) and survival (C) were then recorded.

Fig. 5. Schematic of illustrative recombinant VSVs in accordance with some embodiments. VSV-hIFNe-NIS: VSV Indiana was engineered to express human interferon beta (ΜΡΝβ) in the M/G intergenic region and human sodium iodide symporter (NIS) in the G/L intergenic region, and isolated as described elsewhere (Naik et al., Leukemia, 26:1870-78 (2012)). VSVs expressing CDV-F _22458/16 (with the CDV F polypeptide signal peptide replaced by the MeV F polypeptide signal peptide) and CDV-H ₅₈₀₄ were generated using the pVSV-smart platform. Point mutations in the CDV-H ₅₈₀₄ polypeptide, Y539A and R529A, were introduced using site-directed mutagenesis to eliminate natural tropism for nectin-4 and SLAMF1 polypeptides

- 7 046760 dogs respectively. Targeting viruses were generated by displaying scFvs targeting EGFR or CD38 at the C terminus of CDV-H ₅₈₀₄ with an IGES linker peptide and an H6 polyhistidine tag. Specificity-redirected VSV-CDV F/H constructs were maintained in Vero-anti-H6 to allow infection, virus amplification, and target cell fusion. The titers for each recombinant virus are indicated.

Fig. 6. Monolayers of the indicated CHO cells (wild type or stably overexpressing the indicated receptors) were mock infected or infected (MOI=0.2) with VSV-CDVFH-GFP or VSVCDVF/H _aa -oEGFR-GFP. Fluorescence micrographs were taken at 100× magnification at the indicated time points. GFP expression (green) correlates with virus infection and virus spread in monolayers.

Fig. 7. Monolayers of the indicated CHO cells (wild type or stably expressing EGFR or CD38 receptors) were infected (MOI=0.1) with VSV-CDVFHaa-oEGFR or VSV-CDVFH _aa -αCD38. After 42 h, cell monolayers were fixed with paraformaldehyde and stained with crystal violet. Photographs were taken at 40x magnification.

Fig. 8. Therapeutic effects of chimeric VSV-CDVFHaa-oEGFR against human ovarian cancer xenograft in the abdominal cavity. 5-6 week old female nude mice (Envigo, Indianapolis, IN) were intraperitoneally implanted with ²⁰⁶ SKOV3ip.1-Fluc cells (200 μl/mouse) (day -7). At 7 days postimplantation (day 0), tumor-bearing mice were randomized based on firefly luciferase signals using the IVIS Spectrum system (Perkin Elmer, Hopkinton, MA). Mice were identified by microchip and ear plucking. After randomization, mice received a single dose of 1*107 TCID50 virus or control saline (250 μl/mouse) by intraperitoneal injection. Mice were euthanized if they developed ascites, tumors at the subcutaneous injection site exceeded 10% of body weight, or if animals lost >20% of body weight. All surviving mice were euthanized at the end of the experiment (day 92 after virus treatment). Kaplan-Meier survival curves were generated and compared using the log-rank test. Clinical observations and weight were recorded three times a week until the end of the study or the mouse was euthanized.

Fig. 9 - nucleic acid sequence (SEQ ID NO: 1) of the CDV H open reading frame encoding the CDV H polypeptide (SEQ ID NO: 2).

Fig. 10 is the nucleic acid sequence (SEQ ID NO: 3) of the CDV F open reading frame encoding the CDV F polypeptide (SEQ ID NO: 4).

Fig. 11. Redirection of wild-type CDV envelope specificity to EGFR and CD38. (A) Schematic representation of the cloning strategy to generate wild-type CDV H polypeptides with redirected specificity (top). Standard one-letter abbreviations for amino acids were used to indicate changes introduced to eliminate the use of native receptors (SLAMF1 and nectin-4) (bottom). Amino acid numbering corresponds to SEQ ID NO: 5. Single-chain antibody fragments are represented as a C-terminal extension of glycoprotein H using the factor Xa (Fxa) cleavage site (amino acid sequence IEGR). All constructs optionally contain a hexahistidine tag to facilitate virus rescue on Vero-His cells. (B) Cotransfection experiments demonstrating the targeting competence of a receptor-blind H polypeptide fusion targeting CD38. CHO-CD38 cells in 12-well plates were cotransfected with either CMV-directed CDV F plasmid, CMV-directed wild-type CDV H-CD38, or CMV-directed receptor-blind CDV H-CD38, and after 24 h the cells were fixed, stained, and imaged. (C) Targeted cell fusion mediated by CDV H constructs was resistant to mixed human measles immune serum. CHO-CD38 cells were cotransfected with CMV-targeted plasmids H and F along with a CMV-targeted GFP plasmid for imaging and incubated with the indicated dilutions. Cells were photographed 24 h after transfection. (D) Chimeric measles viruses carrying CDV H-directed polypeptides retain specificity on a panel of CHO cells expressing the required receptors or human tumor cell lines with the required receptor. Cell lines were infected with the respective viruses at an MOI of 0.5 and photographed 48 h later. (E) Experimental design to study the oncolytic efficacy and specificity of in vivo CDV envelope-redirected chimeric measles viruses. Athymic mice were implanted with 5*106 SKOV3ip-fluc cells subcutaneously (SC) or intraperitoneally (i.p.), followed by 6 doses of 1*106 TSH ₅₀ /ml intratumorally (i.p.) for subcutaneous tumors or 2*106 TSH ₅₀ /ml i.p. for i.p. tumors every other day, starting on day 10. (F) Individual volumes of subcutaneous SKOV3ip tumors treated with the corresponding viruses (top), and survival of animals with intraperitoneal tumors after treatment with the corresponding viruses (bottom).

Fig. 12 - Alignment of representative H number of CDV polypeptides. The top sequence (designated AF378705.1_America1) is SEQ ID NO: 5 and is used for numbering purposes where indicated.

- 8 046760

Fig. 13 - alignment of representative F number of CDV polypeptides. The signal peptide sequence extends from amino acid position 1 to amino acid position 135. The top sequence (designated AF378705.1_America1) is SEQ ID NO: 7 and is used for numbering purposes where indicated.

Fig. 14. CDV OL can infect cells lacking SLAMF1 and nectin-4 receptors. (A) Assessment of infection of Vero cells by CDV isolates compared with the OL strain. Cells were infected at an MOI of 0.1 (determined on Vero-dogSLAMF1 cells) and 48 h later stained with HemaQuick for visualization. (B) A panel of CHO cells expressing different relevant receptors was infected with an eGFP MeV reporter incorporating CDV H/F OL glycoproteins. Infectivity was recorded using a fluorescence microscope. 40 ^x magnification.

Fig. 15. Heterologous combination of wild-type CDV H/F with a truncated signal peptide results in increased receptor-dependent fusion as a result of weaker H-F interactions. (A) Syncytia formation in cells cotransfected with CDV-F, CDV-H, and eGFP. The CDV-F signal peptides were replaced by the MeV-F homologue, as indicated by the black boxes in the diagram. 24 h after cotransfection, fusion was assessed using the GFP channel. (B) Quantitative fusion analysis. BHK effector cells were transfected with the indicated combination of attachment proteins (CDV-H or Nipah-G) and fusion proteins (F) with one of the plasmids containing one of the two parts of the split reporter. Target CHO cells and CHO cells expressing CD38 (CHO-CD38) were transfected with a different plasmid containing one of the two parts of the split reporter. At 16 h posttransfection, cells were layered and Renilla luciferase (RLU) activity was determined 8 h later. Values represent the mean ± standard deviation (SD) of one representative experiment performed in triplicate. Statistical significance was determined using one-way analysis of variance with the Holm-Šidák multiple comparison test (ns, not significant *, p<0.05; **, p<0.002; ***, p<0.0001). (C) Coimmunoprecipitation of CDV-H/F. HEK293T cells transiently expressing wild-type HIS-tagged CDV-H proteins or mutant HIS-tagged CDV-H proteins together with FLAG-tagged CDV-F proteins were lysed and immunoprecipitated (IP) with anti-FLAG antibody. Signal intensity was determined using anti-HIS antibody. (D) Quantitative fusion analysis of the fully retargeted CDV-H and MeV-H proteins on CHO cells and CHO-derived cells. MeV-H/F and CDV-H/F complexes, HIS-tagged or HIS-tagged and CD38-redirected, were transfected into effector cells and the luminescent signal was determined over time. MeV-Haals=MeV-H, blinded for CD46, nectin-4 and SLAMF1 by mutations Y481A, R533A, S548L and F549S.

Fig. 16. Conservatism of the amino acid residue M437 in the CDV-H protein in various genetic groups. Sequence alignments were performed with CDV-H sequences obtained from GenBank, including the CDV-H sequence determined here for isolate SPA.Madrid/22458/16. Registration numbers are indicated.

Fig. 17. Integrity and expression of chimeric ligand-exposing receptor-binding proteins. (A) Western blot analysis of HEK293T cells transfected with the indicated proteins, fused or not, to an anti-CD38 scFv. Proteins were labeled with anti-HIS antibody or anti-β-actin antibody (application control). (B) Expression of attachment proteins and mutants on HEK293T cells fixed with or without permeabilization was analyzed by flow cytometry. Histograms are from one representative experiment with two biological replicates. The geometric mean intensity ± standard deviation of two biological replicates is shown in the upper right corner of each histogram. The shaded curves indicate cells transfected with empty plasmids.

Fig. 18. Insertion of a FLAG tag into the F ectodomain and its effect on protein bioreactivity. (A) Schematic representations of uncleaved MeV-F and CDV-F. The NH2 and COOH termini, signal peptide (SP), fusion peptide (FP), transmembrane (TM) and cytoplasmic regions are indicated. The sequence surrounding the cleavage site (in bold) and the sequence of the fusion peptide are shown. Numbering takes into account homotypic signal peptides. (B) Syncytia formation in Vero cells after cotransfection of homologous expression plasmids H and F with FLAG insertions at different positions. Cells were stained 16 h after transfection and micrographs were taken for quantification. (C) Quantification of syncytium formation. Data are presented as mean ± standard deviation (n=20). Significance was determined using one-way analysis of variance with the Holm-Šidák multiple comparison test (ns, not significant; ***, p < 0.001). (D) Split protein fusion assay for cotransfection of CDV-H/F SPA with or without insertion of a FLAG tag at amino acid 216. The luciferase signal was measured after 8 h. The experiment was performed in duplicate.

Fig. 19. CD46 specificity of scFv. (A) Analysis of target proteins by SDSPAG. MW: molecular weight marker, C: Coomassie stain; WB: Western blot using anti-CD46 antibody. (B) Size exclusion chromatography curve for CD46 used

- 9 046760 in experiments. The calculated MW value is given from the calibration curve. (C) Binding of scFv-Fc-tagged fusion proteins to CD46 or nectin-4 determined by ELISA. Detection was carried out using the Fc fragment used as a control for the amount of protein. The experiments were carried out in double technical repetition. Data are presented as mean±standard deviation (n=2). Significance was determined using one-way analysis of variance with the Holm-Šidák multiple comparison test. *, p<0.05; **, p<0.005.

Fig. 20. The CD46 binding affinity of exposed scFv determines CD46-dependent cell fusion of the CDV H/F complex with redirected specificity. (A) Representative sensorgram (in resonance units, RU) of CD46 binding to biosensor surfaces containing (solid line) or not (broken line) single-chain antibody fragments (scFv). Experimental data represent injection of scFv K2 for 300 s followed by injection of buffer. Then, 1 μM CD46 was passed over both surfaces of the biosensor and the signal was recorded during (association) and after administration (dissociation). The surfaces were finally regenerated at the end of the cycle as described herein. (B) Binding of CD46 to scFv as assessed by surface plasmon resonance. Sensograms show response units (black lines) of different concentrations of CD46 per scFv. The best-fit 1:1 binding model is shown as broken red lines. Binding affinity (Kd) was determined based on the rates of association and dissociation (Table 1). (C) Quantitative fusion assay for MeV-H and CDV-H variants on CHO cells. The experiment was performed in duplicate and repeated twice with similar results (see Figure 21). Data are presented as mean ± standard deviation.

Fig. 21. The binding affinity of scFv exposed on the CDV-H/F complex directs enhanced cell-cell fusion. (A) Cellular enzyme-linked immunosorbent assay (CELISA) to determine the amount of cellular protein used in the quantitative fusion assay in FIG. 20C. CELISA was performed on CHO cells transfected with the indicated attachment protein using a monoclonal antibody against the 6*H^ tag (n=5). (B) Quantitative fusion analysis of the CD46-redirected CDV-H/F complex using affinity-adjusted scFv (same data as Fig. 20C). Y539A indicates a substitution in CDV-H to eliminate natural tropism for nectin-4.

Fig. 22. CD46-redirected αβ envelope glycoproteins of CDV determine the tropism of the virus. (A) Schematic representation of Stealth: vaccine-derived measles virus pseudotyped with CD46-redirected CDV H/F envelope proteins. Created using BioRender.com. (B) Role of CD46 binding affinity in viral entry. Cells were infected at the indicated multiplicity of infection with Stealth viruses exhibiting scFvs with different affinities for CD46. eGFP expression was monitored 48 h postinfection. (C) CHO cells and derivative cells expressing HIS pseudoreceptor or CD46 were infected with Fluc-expressing Stealthviruses (K1 and A09) at a multiplicity of infection of 0.5. Luciferase expression was measured 48 h postinfection. n=2, excluding CHO-CD46, *, p-value <0.05 (two-tailed t-test), (D) Multistep growth kinetics of Stealth-A09 in Vero or Vero-aHIS cells. At the indicated time points, the supernatant and cell pellet were collected, and virus titers were determined on Vero-aHIS cells. Values and error bars (SD) were determined for a representative experiment performed in triplicate. (E) Protein composition of viruses. Western blot analysis was performed using similar amounts of viral particles and labeled with appropriate antibodies. The molecular weight of the standard is indicated. (F) Viral tropism. CHO cell derivatives were infected with eGFP-expressing viruses as indicated. After 48 h, eGFP autofluorescence was determined. Scale bar, 200 µm. (G) Genetic stability of Stealth. Vero-hSLAMF1 cells were infected with Stealth and passaged several times. After 8 passages, the collected virus was used to infect Vero cells expressing human or canine SLAMF1. Representative micrographs are shown after three or six days postinfection.

Fig. 23. Assessment of interactions with receptors for the engineered CDV fusion apparatus complex. Cells were cotransfected with MeV-F and MeV-H or redirected variants CDV-F and CDV-H with a CD46-specific scFv. For visualization purposes, an expression plasmid encoding eGFP was cotransfected, and eGFP autofluorescence was visualized 24 h after transfection. Y539A indicates a substitution in CDV-H to reduce natural tropism for nectin-4. For semiquantitative determination, the symbols + and - were used (same as in Fig. 15A). Unexposed means no scFv.

Fig. 24. High binding affinity of CD46 determines the oncolytic activity of the CD46-targeted Stealth virus in a mouse model of ovarian cancer. (A) Study design diagram. SKVOv3ip.1 tumor cells encoding the firefly luciferase gene (SKOV3ip.Fluc) were implanted intraperitoneally into athymic mice. On day 10, 1^106 TCID ₅₀ Stealth particles were administered using the same route. Tumor burden was then monitored using bioluminescence imaging (BLI) at 7-day intervals. (B) Kaplan survival curves of carrier mice.

- 10 046760

SKOV3ip.Fluc treated with Stealth-N1E and Stealth-A09 viruses (n=5). Statistical significance was determined using the log-rank test. (C) Representative BLI images showing the dorsal view of treated animals. Radiation intensities (photons per second per cm per steradian, f/s/cm ² /sr) were converted to colors indicating tumor burden in mice according to the legend shown on the right. (D) Quantification of total body luminescence in photons per second per cm per steradian (f/s/cm ² /sr). n=5. Ns, not significant; *, p-value <0.05; **, p<0.005.

Fig. 25. Stealth-A09 virus causes oncolysis indistinguishable from the parent MeV in a mouse model of multiple myeloma. (A) SCID mice bearing subcutaneous U266.B1 cell tumors received a suboptimal dose of virus intravenously. Tumor growth was measured with calipers (n=5) and animals were euthanized when tumors ulcerated or when tumor size reached 20% of body weight. (B) Kaplan-Meier survival curves (n=5). Significant differences between groups were determined using the log-rank test (*, p<0.05). (C) Migration of virus into subcutaneous tumor cells after systemic administration. eGFP expression was assessed by immunohistochemistry on two representative samples from each group collected at euthanasia. Scale, 200 nm.

Fig. 26. Increased binding affinity to CD46 improves entry of CD46-specific virus. Fluc-expressing Stealth viruses were used to infect the indicated cells at a reduced MOI. Luciferase expression was measured 48 h postinfection. n=2 for all except CHO-CD46 and Stealth-A09 (n=3).

Fig. 27. Stealth virus remains oncolytic in the presence of MeV immune serum. (A) SKOV3ip.Fluc cells were injected into athymic nude mice and allowed to engraft for 10 days. Mice in the respective groups then received 600 mIU IgG anti-MeV antibody intraperitoneally, three hours before virus administration by the same route. (B) Kaplan-Meier survival curves (n=5 mice per group). Significant differences between groups were determined using the log-rank test (ns, not significant; *, p<0.05; **, p<0.002). (C) Representative BLI images showing the dorsal view of treated animals. Radiation intensity (photons per second per cm per steradian, photon/s/cm ² /sr) was converted into colors to indicate tumor burden in mice according to the legend shown on the right. (D) Quantification of total body luminescence in photons per second per cm per steradian (f/s/cm ² /sr). n=5. Statistical significance was determined using one-way analysis of variance with Dunnett's multiple comparison test. Ns, not significant.

Fig. 28. No cross-neutralization between measles virus and Stealth. (A) Virus neutralization assay for MeV and Stealth. Mixed human AB serum (left panel) or ferret anti-CDV serum (right panel) were used. Relative infection refers to the amount of infection in the presence of serum compared with that in the absence of serum. Values were calculated from two or three biological replicates performed in four technical replicates. (B) Antiserum from infected HuCD46Ge-IFNARKO mice was also used to determine cross-neutralization between viruses, n=8 (note that some data points overlap). ND50 titers were converted to mIU/mL based on the ND50 value obtained for MeV when assessed using the WHO International Serum Standard 3 (3 IU/mL).

Detailed description

Presented herein are CDV F polypeptides. As described herein, the CDV F polypeptide can be engineered such that viral particles containing the CDV F polypeptide together with the CDV H polypeptide exhibit increased fusogenic activity. For example, the CDV F polypeptide can be designed to contain a signal peptide sequence that is no more than 75 amino acids in length. Typically, wild-type CDV F polypeptides contain a signal peptide sequence approximately 135 amino acids in length. An example 135 amino acid signal peptide sequence of a wild type CDV F polypeptide is provided in SEQ ID NO: 6 (MHKEIPEKSRTRTHTQQDLPQQKSTEYTEIKTSRARHGITPAQRSTHYGPRTLDRLVCYIMNRAMSCKQ ASYRSDNIPAHGDHEGVVHHTPESVSQGARSQLKRRTSNAINSGFQYIWLVLWCIGIASLFLCSKA).

As described herein, truncation of the signal peptide sequence of CDV F polypeptides so that it does not exceed 75 amino acids in length can result in CDV F polypeptides, which, when part of viruses together with CDV H polypeptides, provide increased fusogenic activity of the viruses by compared with the level of fusogenic activity exhibited by comparable control viruses containing the CDV F polypeptide having the full-length wild-type signal peptide sequence (eg, SEQ ID NO: 6).

The CDV F polypeptide provided herein may contain a signal peptide sequence that ranges from 7 amino acids to 75 amino acids in length. For example, the CDV F polypeptide provided herein may contain a signal peptide sequence that is 7 to 75 in length (e.g., 7 to 70, 7 to 65, 7 to 60, 7 to

- 11 046760

55, from 7 to 50, from 7 to 45, from 7 to 40, from 7 to 35, from 7 to 30, from 7 to 25, from 10 to 75, from 15 to 75, from 20 to 75, from 25 to 75, 35 to 75, 45 to 75, 50 to 75, 55 to 75, 65 to 75, 20 to 60, 25 to 50, 30 to 60 or 30 to 40) amino acids. The CDV F polypeptide provided herein can be prepared by truncating the wild-type signal peptide sequence at its N-terminus, at its C-terminus, or at both its N-terminus and C-terminus, or by removing amino acids from the region between the N-terminus and the C-terminus. terminal and C-terminal regions of the wild-type signal peptide sequence. In some cases, the measles viral signal peptide sequence may be used as the signal peptide of the CDV F polypeptide described herein. Examples of CDV F polypeptide signal peptide sequences provided herein include, but are not limited to, the sequences presented in Table 1. 1.

Table 1

Examples of signal peptide sequences

Example Subsequence SEQ ID NO: 1 MNRAMSCKQASYRSDNIPAHGDHEGVVHHTPESVSQG ARSQLKRRTSNAINSGFQYIWLVLWCIGIASLFLCSKA 8 2 MTSNAINSGFQYIWLVLWCIGIASLFLCSKA 9 3 MSIMGLKVNVSAIFMAVLLTLQTPTG 10 4 MAINSGSQCTWLVLWCLGIASLFLCSKA eleven 5 MAINSGSQCTWLVLWCLGMASLFLCSKA 12 6 MAINSGSQCTWLVLWCFGTASLFLCSKA 13 7 MATSPGPQCTWLVLWCIGIASLFLCSEA 14 8 MAVKSGSQCTWLVLWCIGVASLFLCSKA 15 9 MASNPGSQCTWLVLWCIGIASLFLCSKA 16 10 MATNAGSQYTWLVLWCIGIASLFLCSKA 17 eleven MATSSGSQCTWLVLWCIGVASLFLCSKA 18 12 MAINSGSQCTWLVLWCIGVASLFLCSKA 19 13 MAINSGFQYIWLVLWCIGIASLFLCSKA 20

In some cases, the CDV F polypeptide provided herein may be engineered to lack the entire signal peptide sequence. For example, the CDV F polypeptide provided herein may have one of the amino acid sequences shown in FIG. 13, which starts with amino acid at position 140.

The CDV F polypeptide provided herein may have any suitable amino acid sequence, provided that the CDV F polypeptide does not contain a signal peptide sequence longer than 75 amino acid residues. Examples of CDV F polypeptide amino acid sequences that may be used as described herein include, but are not limited to, those amino acid sequences set forth in FIG. 13.

CDV H polypeptides are also presented herein. As described herein, CDV H polypeptide can be engineered such that viruses containing the CDV H polypeptide together with the CDV F polypeptide exhibit reduced or completely eliminated tropism for SLAMF1 polypeptides and/or nectin-4 polypeptides compared to viruses containing Wild-type CDV H polypeptides. For example, a CDV H polypeptide may be engineered to contain a mutation in one or more (e.g., one, two, three, four, five, six, seven, eight, nine, ten, 11, 12, 13, 14, or 15 ) from amino acid positions 454, 460, 479, 494, 510, 520, 525, 526, 527, 528, 529, 537, 539, 547 and 548. Typically, viruses containing wild-type CDV H polypeptides (together with F polypeptides CDV) exhibit tropism for SLAMF1 polypeptides and nectin-4 polypeptides, resulting in the viruses infecting SLAMF1-positive cells and nectin-4-positive cells. As described herein, mutation of one or more amino acid positions P/S454, V/L/F460, L/F/W479, I494, I/L/V510, Y520, Y/N525, D/G526, I/V527, S/T528, R529, Y/D537, Y539, Y/F547, and Y/M548 CDV H polypeptide substituted with another amino acid (e.g., alanine) may reduce or eliminate the ability of viruses containing such CDV H polypeptide (together with CDV F polypeptide ), infect SLAMF1-positive cells and/or nectin-4-positive cells. Examples of CDV H polypeptides presented herein with reduced or eliminated tropism for SLAMF1 polypeptides and/or nectin-4 polypeptides include, but are not limited to, CDV H polypeptides presented at

- 12 046760 fig. 12, provided that the CDV H polypeptide contains a mutation in one or more (for example, one, two, three, four, five, six, seven, eight, nine, ten, 11, 12, 13, 14 or 15) of P /S454, V/L/F460, L/F/W479, I494, I/L/V510, Y520, Y/N525, D/G526, I/V527, S/T528, R529, Y/D537, Y539, Y /F547 and Y/M548. Examples of mutations that can be used to produce a CDV H polypeptide having reduced or eliminated tropism for SLAMF1 polypeptides and/or nectin-4 polypeptides include, but are not limited to, the mutations presented in table. 2. Examples of combinations of mutations presented in table. 2, which can be used to produce a CDV H polypeptide having reduced or eliminated tropism for SLAMF1 polypeptides and/or nectin-4 polypeptides include, but are not limited to, the polypeptides listed in table. 3.

table 2

Examples of mutations that may be introduced into the CDV H polypeptide (e.g., CDV H polypeptide shown in FIG. 12)

Example Position Amino acid substitution Specific examples 1 P454 Any amino acid other than P and S A, I, L, M, F, W, Y, V, R, H, K, D, E, S, T, N, Q, C, G or P 2 V/L/F460 Any amino acid other than V, L and F A, I, M, F, W, Y, S, T, N, Q, C, G, P, R, H, K, D or E 3 L/F/W479 Any amino acid except L, F and W A, I, M, Y, V, S, T, N, Q, C, G, P, R, H, K, D, E 4 I494 Any amino acid except I A, L M, F, W, Y, V, S, T, N, Q, C, G, P, R, H, K, D or E 5 I/L/V510 Any amino acid other than I, L and V A, M, F, W, Y, S, T, N, Q, C, G, P, R, H, K, D or E 6 Y520 Any amino acid other than Y A, I, L M, F, W, V, S, T, N, Q, C, G, P, R, H, K, D or E 7 Y/N525 Any amino acid other than Y and N A, I, L M, F, W, V, S, T, Q, C, G, P, R, H, K, D or E 8 D/G526 Any amino acid other than D and G A, I, L M, F, W, Y, V, S, T, N, Q, C, P, R, H, K or E 9 I/V527 Any amino acid except I and V A, L M, F, W, Y, S, T, N, Q, C, G, P, R, H, K, D or E 10 S/T528 Any amino acid other than S and T A, I, L M, F, W, Y, V, N, Q, C, G, P, R, H, K, D or E eleven R529 Any amino acid other than R A, I, L M, F, W, Y, V, S, T, N, Q, C, G, P, H, K, D or E 12 Y/D537 Any amino acid other than Y and D A, S, T, N, Q, G, C, P, D, R, H, K or E 13 Y539 Any amino acid other than Y A, S, T, N, Q, G, C, P, D, E, R, H, K or E 14 Y/F547 Any amino acid except Y and F A, S, T, N, Q, G, C, P, D, E, R, H, K or E 15 Y/M548 Any amino acid other than Y and M A, S, T, N, Q, G, C, P, D, E, R, H, K or E

Table 3

Examples of combinations of mutations from table. 2 that may be included in the CDV H polypeptide

Combination Combination of mutations from table. 2 1 All from 1-15 2 8, 9, 10, 11, 14 and 15 3 7, 8 and 11 4 7 and 8 5 7 and 11 6 8 and 11

This document also provides recombinant viruses (e.g., VSV) containing the CDV H polypeptide provided herein and/or the CDV F polypeptide provided herein, as well as methods for producing recombinant viruses (e.g., VSV) containing the H polypeptide CDV as provided herein and/or CDV F polypeptide,

- 13 046760 presented in this document. For example, a recombinant virus (e.g., VSV) can be constructed to include: (a) a CDV H polypeptide as provided herein and a wild-type CDV F polypeptide, (b) a wild-type CDV H polypeptide and a CDV F polypeptide as represented by as herein, or (c) a CDV H polypeptide as provided herein and a CDV F polypeptide as provided herein. In some cases, a recombinant virus (eg, VSV) can be engineered to include a CDV H polypeptide having CDV H 5804 and a CDV F polypeptide having CDV F 22458/16.

Also provided herein are nucleic acid molecules encoding the CDV H polypeptide provided herein and/or nucleic acid molecules encoding the CDV F polypeptide provided herein. For example, a nucleic acid molecule (eg, a vector) encoding a CDV H polypeptide as provided herein and/or a CDV F polypeptide as provided herein can be constructed.

This document provides methods and materials related to VSV. For example, this document provides replication-competent VSVs, nucleic acid molecules encoding replication-competent VSVs, methods for producing replication-competent VSVs, and methods for using replication-competent VSVs to treat cancer or infectious diseases.

As described herein, VSV can be engineered to contain a nucleic acid molecule encoding a VSV N polypeptide, a VSV P polypeptide, a VSV M polypeptide, a CDV F polypeptide (e.g., a CDV F polypeptide as provided herein), an H polypeptide CDV (eg, CDV H polypeptide as provided herein) and VSV L polypeptide, and does not encode a functional VSV G polypeptide. It should be understood that the sequences described herein with respect to VSV are included in a plasmid encoding the positive sense cDNA of the viral genome, which provides the negative sense genome of VSV. Thus, it should be understood that a nucleic acid sequence that encodes, for example, a VSV polypeptide may refer to an RNA sequence that is the template for a positive sense transcript that encodes (eg, via forward translation) such a polypeptide.

The nucleic acid encoding the CDV F polypeptide and the CDV H polypeptide may be located at any position in the VSV genome. In some cases, the nucleic acid encoding the CDV F polypeptide and the CDV H polypeptide may be located after the nucleic acid encoding the VSV M polypeptide. For example, a nucleic acid encoding a CDV F polypeptide and a nucleic acid encoding a CDV H polypeptide may be located between a nucleic acid encoding a VSV M polypeptide and a nucleic acid encoding a VSV L polypeptide.

Any suitable nucleic acid encoding a CDV F polypeptide can be inserted into the VSV genome. For example, a nucleic acid encoding a wild-type CDV F polypeptide or a CDV F polypeptide as provided herein can be inserted into the VSV genome.

Any suitable nucleic acid encoding a CDV H polypeptide can be inserted into the VSV genome. For example, a nucleic acid encoding a wild-type H polypeptide or an H polypeptide as provided herein can be inserted into the VSV genome. In some cases, a nucleic acid encoding a CDV H polypeptide that lacks specificity for SLAMF1 and/or nectin-4 can be inserted into the VSV genome. For example, a nucleic acid encoding a CDV H polypeptide having one or more of the mutations shown in Table. 2 may be integrated into the VSV genome. In some cases, the VSV/CDV hybrid presented herein can be engineered to have preselected tropism. For example, CDV F and/or H polypeptides with specificity for SLAMF1 and/or nectin-4 knocked out can be used such that a scFv or polypeptide ligand can be attached, for example, to the C-terminus of the H CDV polypeptide. In such cases, the scFv or polypeptide ligand may determine the tropism of the VSV/CDV hybrid. Examples of scFvs that can be used to target VSV/CDV hybrids to cellular receptors (eg, tumor-associated cell receptors) include, but are not limited to, anti-EGFR, anti-CD46, anti-oFR, anti-PSMA, anti-HER-2, anti-CD19, anti- CD20 or anti-CD38. Examples of polypeptide ligands that can be used to target VSV/CDV hybrids include, but are not limited to, urokinase plasminogen activator (UPA) polypeptides, cytokines such as IL-13, single chain T cell receptors (scTCR), echistatin polypeptides, and integrin binding polypeptides.

In some cases, the VSV nucleic acid molecule provided herein may encode an IFN polypeptide, a fluorescent polypeptide (eg, a GFP polypeptide), an NIS polypeptide, a therapeutic polypeptide, an anti-innate immune polypeptide, a tumor antigen, or a combination thereof. The nucleic acid encoding the IFN polypeptide may be located after the nucleic acid encoding the VSV M polypeptide. For example, a nucleic acid encoding an IFN polypeptide may be located between a nucleic acid

- 14 046760 an acid encoding a VSV M polypeptide and a nucleic acid encoding a CDV F polypeptide or a nucleic acid encoding a CDV H polypeptide. This position may allow viruses to express an amount of IFN polypeptide that is effective in activating antiviral innate immune responses in noncancerous tissues and thus attenuating potential viral toxicity without interfering with efficient viral replication in cancer cells.

Any suitable nucleic acid encoding an IFN polypeptide can be inserted into the VSV genome. For example, a nucleic acid encoding an IFN beta polypeptide may be inserted into the VSV genome. Examples of nucleic acids encoding IFN-beta polypeptides that can be integrated into the VSV genome include, but are not limited to, a nucleic acid encoding a human IFN-beta polypeptide, the nucleic acid sequence of which is listed in GenBank® under Reg. number NM_002176.2 (GI No. 50593016), a nucleic acid encoding a murine IFN-beta polypeptide, the nucleic acid sequence of which is presented in GenBank® under reg. number NM_010510.1 (GI No. 6754303), BC119395.1 (GI No. 111601321) or BC119397.1 (GI No. 111601034), and a nucleic acid encoding a rat IFN-beta polypeptide, the nucleic acid sequence of which is reported in GenBank® under reg. number NM_019127.1 (GI No. 9506800).

The nucleic acid encoding the NIS polypeptide may be located after the nucleic acid encoding the CDV F polypeptide or the nucleic acid encoding the CDV H polypeptide. For example, a nucleic acid encoding an NIS polypeptide may be located between a nucleic acid encoding a CDV F or H polypeptide and a nucleic acid encoding a VSV L polypeptide. This position may allow viruses to express an amount of NIS polypeptide that: (a) is effective in selectively accumulating iodide in infected cells, allowing viral spread to be visualized using radioisotopes and radiation therapy directed at infected cancer cells, and (b) is not high enough to be toxic to infected cells.

Any suitable nucleic acid encoding an NIS polypeptide can be inserted into the VSV genome. For example, a nucleic acid encoding a human NIS polypeptide may be inserted into the VSV genome. Examples of nucleic acid encoding NIS polypeptides that can be inserted into the VSV genome include, but are not limited to, nucleic acid encoding human NIS polypeptide with a nucleic acid sequence submitted to GenBank ^® under Reg. numbers NM_000453.2 (GI No. 164663746), BC105049.1 (GI No. 85397913) or BC105047.1 (GI No. 85397519), a nucleic acid encoding a mouse NIS polypeptide with a nucleic acid sequence submitted to GenBank® under registration. numbers NM_053248.2 (GI No. 162138896), AF380353.1 (GI No. 14290144) or AF235001.1 (GI No. 12642413), a nucleic acid encoding a chimpanzee NIS polypeptide with a nucleic acid sequence submitted to GenBank ^® under reg. number XM_524154 (GI No. 114676080), a nucleic acid encoding a canine NIS polypeptide with a nucleic acid sequence submitted to GenBank ^® under reg. number XM_541946 (GI No. 73986161), a nucleic acid encoding a bovine NIS polypeptide with a nucleic acid sequence submitted to GenBank ^® under reg. number XM_581578 (GI No. 297466916), a nucleic acid encoding a porcine NIS polypeptide with a nucleic acid sequence submitted to GenBank® under reg. number NM_214410 (GI No. 47523871), and a nucleic acid encoding a rat NIS polypeptide with a nucleic acid sequence submitted to GenBank® under reg. number NM_052983 (GI No. 158138504).

The VSV nucleic acid sequences provided herein, which encode VSV N polypeptide, VSV P polypeptide, VSV M polypeptide and VSV L polypeptide, may be derived from the VSV Indiana strain submitted to GenBank® under Reg. number NC_001560 (GI No. 9627229), or may be derived from the VSV New Jersey strain.

In one aspect, provided herein are VSVs comprising a nucleic acid molecule (e.g., an RNA molecule) having (e.g., in the 3'^5' direction) a nucleic acid sequence that is a template for a positive sense transcript encoding a VSV N polypeptide, a nucleic acid sequence that is the template for a positive sense transcript encoding a VSV P polypeptide, a nucleic acid sequence that is a template for a positive sense transcript encoding a VSV M polypeptide, a nucleic acid sequence that is a template for a positive sense transcript encoding F a CDV polypeptide, a nucleic acid sequence that is a template for a positive sense transcript encoding a CDV H polypeptide, and a nucleic acid sequence that is a template for a positive sense transcript encoding a VSV L polypeptide, but not containing a nucleic acid sequence that is a template for a positive-sense transcript encoding a functional VSV G polypeptide. Such VSVs can infect cells (eg, cancer cells) and may be replication competent.

- 15 046760

Any suitable method can be used to insert a nucleic acid (eg, a nucleic acid encoding a CDV F polypeptide, a nucleic acid encoding a CDV H polypeptide, a nucleic acid encoding an IFN polypeptide, and/or a nucleic acid encoding an NIS polypeptide) into the genome. ENE. For example, methods described elsewhere (Schnell et al., PNAS, 93:11359-11365 (1996), Obuchi et al., J. Virol., 77(16):8843-56 (2003)); Goel et al., Blood, 110(7):2342-50 (2007)); and Kelly et al., J. Virol., 84(3):1550-62 (2010)) can be used to insert nucleic acid into the VSV genome. Any suitable method can be used to identify VSVs containing the nucleic acid molecule described herein. Such methods include, but are not limited to, PCR and nucleic acid hybridization techniques such as Northern and Southern analysis. In some cases, immunohistochemistry and biochemical techniques may be used to determine whether VSV contains a particular nucleic acid molecule by detecting the expression of the polypeptide encoded by that particular nucleic acid molecule.

In another aspect, provided herein are nucleic acid molecules that encode a VSV N polypeptide, a VSV P polypeptide, a VSV M polypeptide, a CDV F polypeptide, a CDV H polypeptide, and a VSV L polypeptide, but are not capable of encoding a functional VSV G polypeptide. For example, a nucleic acid molecule provided herein may be a single nucleic acid molecule that includes a nucleic acid sequence that encodes a VSV N polypeptide, a nucleic acid sequence that encodes a VSV P polypeptide, a nucleic acid sequence that encodes a VSV M polypeptide, a nucleic acid sequence that encodes a CDV F polypeptide, a nucleic acid sequence that encodes a CDV H polypeptide, and a nucleic acid sequence that encodes a VSV L polypeptide, and does not contain a nucleic acid sequence that encodes a functional VSV G polypeptide.

In another aspect, provided herein are nucleic acid molecules that encode VSV N polypeptide, VSV P polypeptide, VSV M polypeptide, IFN polypeptide, CDV F polypeptide, CDV H polypeptide, NIS polypeptide, and VSV L polypeptide, but are not capable of encoding a functional VSV G polypeptide. For example, a nucleic acid molecule provided herein may be a single nucleic acid molecule that includes a nucleic acid sequence that encodes a VSV N polypeptide, a nucleic acid sequence that encodes a VSV P polypeptide, a nucleic acid sequence that encodes a VSV M polypeptide, a nucleic acid sequence that encodes an IFN polypeptide, a nucleic acid sequence that encodes a CDV F polypeptide, a nucleic acid sequence that encodes a CDV H polypeptide, a nucleic acid sequence that encodes an NIS polypeptide, and a nucleic acid sequence that encodes a VSV L polypeptide, and however, it is not capable of encoding the functional VSV G polypeptide.

The term nucleic acid as used herein includes both RNA (eg, viral RNA) and DNA, including cDNA, genomic DNA, and synthetic (eg, chemically synthesized) DNA. Nucleic acid can be double-stranded or single-stranded. The single-stranded nucleic acid may be a sense strand or an antisense strand. In addition, the nucleic acid may be circular or linear.

Also provided herein are a method of treating cancer (eg, to reduce tumor size, inhibit tumor growth, or reduce the number of viable tumor cells), methods of inducing host immunity against cancer, and methods of treating an infectious disease such as HIV or measles. For example, a recombinant virus (e.g., VSV) provided herein can be administered to a mammal having cancer to reduce tumor size, inhibit cancer cell or tumor growth, reduce the number of viable cancer cells in the mammal, and/or induce an immunogenic host response against tumors. A recombinant virus (eg, VSV) provided herein can be propagated in host cells to increase the available number of copies of the virus, typically at least 2-fold (eg, 5-10-fold, 50-fold) 100 times, 500-1000 times and even up to 5000-10000 times). In some cases, the recombinant virus (eg, VSV) provided herein can be propagated to the desired concentration in standard cell culture media (eg, DMEM or RPMI-1640 supplemented with 5-10% fetal bovine serum, at 37°C in 5% CO2). The virus titer is usually determined by inoculating cells (eg Vero cells) in culture.

Recombinant viruses (eg, VSV) provided herein can be administered to a cancer patient, for example, by direct injection into a group of cancer cells (eg, a tumor) or intravenous delivery into cancer cells. The recombinant virus (eg, VSV) provided herein can be used to treat various forms of cancer, including, without limitation, myeloma (eg, multiple myeloma), melanoma, glioma, lymphoma,

- 16 046760 mesothelioma and cancer of the lung, brain, stomach, colon, rectum, kidney, prostate, ovary, breast, pancreas, liver, head and neck.

The recombinant viruses (eg, VSV) provided herein can be administered to a patient in a biocompatible solution or pharmaceutically acceptable delivery vehicle, either directly into a group of cancer cells (eg, intratumorally) or systemically (eg, intravenously). Suitable pharmaceutical formulations depend in part on the application and route of administration, eg transdermal or injection. Such forms should not prevent the composition or composition from reaching the target cell (ie, the cell into which the virus is to be delivered) or interfere with the manifestation of its effects. For example, pharmacological compositions administered into the bloodstream must be soluble.

Although the doses administered will vary depending on the individual patient (eg, depending on the size of the tumor), the effective dose can be determined by setting as a lower limit the concentration of virus that has been demonstrated to be safe and increasing it to higher doses of up to 10 ¹² PFU , while monitoring the reduction in cancer cell growth, as well as the presence of any unwanted side effects. A therapeutically effective dose usually provides at least a 10% reduction in the number of cancer cells or tumor size. Dose escalation studies can be used to obtain the desired effect of a given viral treatment (see, for example, Nies and Spielberg, Principles of Therapeutics, in Goodman &Gilman's The Pharmacological Basis of Therapeutics, eds. Hardman, et al., McGraw-Hill, NY, 1996, pp. 43-62).

Recombinant viruses (eg, VSV) provided herein can be delivered in doses that range from, for example, about 10 ³ PFU to about 10 ¹² PFU (for example, from about 10 ⁵ PFU to about 10 ¹² PFU, from about 10 ⁶ PFU to about 10 ¹¹ PFU or about 10 ⁶ PFU to about 10 ¹⁰ PFU). A therapeutically effective dose can be achieved by repeated doses. Repeat administration is appropriate in cases where observations of clinical symptoms or tumor size or monitoring tests indicate that a group of cancer cells or tumor has stopped shrinking, or that the degree of viral activity is decreasing while the tumor is still present. Repeat doses can be administered by the same route as initially or by a different route. A therapeutically effective dose may be delivered in multiple separate doses (eg, several days or weeks apart), and in one embodiment, one to about twelve doses are provided. Alternatively, a therapeutically effective dose of the recombinant viruses (eg, VSV) provided herein can be delivered via a sustained release formulation. In some cases, a recombinant virus (eg, VSV) provided herein may be delivered in combination with pharmacological agents that promote viral replication and spread in cancer cells, or agents that protect non-cancerous cells from viral toxicity. Examples of such agents are described elsewhere (Alvarez-Breckenridge et al., Chem. Rev., 109(7):3125-40 (2009)).

Recombinant viruses (eg, VSV) provided herein can be administered using a sustained release device. The formulation for sustained release of recombinant viruses (eg, VSV) provided herein may include, for example, a polymeric excipient (eg, a swelling or non-swelling gel or collagen). A therapeutically effective dose of recombinant viruses (eg, VSV) provided herein can be provided in a polymeric excipient where the excipient/virus composition is implanted into a site of cancer cells (eg, near or within a tumor). The action of body fluids leads to the gradual dissolution of the excipient and the continuous release of an effective dose of the virus over a period of time. Alternatively, the sustained release device may comprise a series of alternating active and release layers. Each active layer of such a device typically contains a dose of virus contained in an excipient, while each separating layer contains only excipient or low concentrations of virus (ie, below the effective dose). As each subsequent layer of the device dissolves, pulsed doses of the virus are delivered. The size/composition of the separation layers determines the time interval between doses and is optimized according to the therapeutic regimen used.

In some cases, recombinant viruses (eg, VSV) presented herein can be administered directly. For example, the virus can be injected directly into a tumor (eg, a breast cancer tumor) that can be felt through the skin. Ultrasound monitoring can also be used in this method. Alternatively, direct injection of the virus can be performed through a catheter or other medical access device and can be used in conjunction with an imaging system to localize a group of cancer cells. In this method, an implantable dosing device is typically placed close to a group of cancer cells using a guide wire installed in a medical access device. Effective

- 17 046760 a dose of a recombinant virus (eg, VSV) presented herein can be injected directly into a group of cancer cells that is visible in an open surgical field.

In some cases, recombinant viruses (eg, VSV) presented herein can be delivered systemically. For example, systemic delivery can be accomplished intravenously by injection or using an intravenous delivery device designed to administer multiple doses of a drug. Such devices include, but are not limited to, butterfly infusion needles, peripheral intravenous catheters, mid-venous catheters, peripherally inserted central venous catheters, and surgically inserted catheters or ports.

The course of therapy with a recombinant virus (eg, VSV) presented herein can be monitored by assessing changes in clinical symptoms or by directly observing the number of cancer cells or tumor size. For solid tumors, the effectiveness of viral treatment can be assessed by measuring the size or weight of the tumor before and after treatment. Tumor size can be measured either directly (eg, using a caliper), using imaging techniques (eg, X-ray, magnetic resonance imaging, or computed tomography), or by assessing non-imaging optical data (eg, spectral data). In the case of a group of cancer cells (eg, leukemia cells), the effectiveness of a viral treatment can be determined by measuring the absolute number of leukemia cells in the patient's circulating blood before and after treatment. The effectiveness of viral treatment can also be assessed by monitoring the levels of a specific cancer antigen. Specific cancer antigens include, for example, carcinoembryonic antigen (CEA), prostate specific antigen (PSA), prostatic acid phosphatase (PAP), CA 125, alpha fetoprotein (AFP), carbohydrate antigen 15-3 and carbohydrate antigen 19-4 .

The invention will be further described in the following examples, which do not limit the scope of the invention as described in the claims.

Examples

Example 1. CDV F and H polypeptides and recombinant viruses.

Cell lines.

African green monkey kidney cells Vero (Vero; American Type Culture Collection [ATCC], cat. no. CCL-81) and their derivatives (or expressing nectin-4 (Noyce et al., Virology, 436(1):210-20 ) 2013)), SLAMF1 (Tatsuo et al., Nature, 406(6798):893-7 (2000)), canine SLAMF1 (von Messling et al., J. Virol., 77(23):12579-91 ( 2003)) or membrane-bound single chain antibody specific for hexahistidine peptide (Nakamura et al., Nat. Biotechnol., 23(2):209-14 (2005)) were maintained in Dulbecco's modified Eagle's medium (DMEM) (GE Healthcare Life Sciences, cat. no. SH30022.01) supplemented with 5% (v/v) heat-inactivated fetal bovine serum (FBS, Gibco) and 0.5 mg/ml geneticin (G418; Corning) for Vero/NECTIN-4 and Vero/ SLAMF1 or 1 mg/ml zeocin (ThermoFisher) for Vero/dogSLAMF1. Human kidney epithelial cells (HEK293T) obtained from Dr. Cosset (University of Lyon), neonatal hamster kidney cells (BHK, ATCC, cat. no. CCL-10), human glioblastoma cells U-87 MG ((ATCC, cat. no. CCL-10). HTB-14) and human ovary tumor cells SKOV3ip.1 were maintained in DMEM with 10% FBS. Chinese hamster ovary (CHO), CHO-CD46 and CHO-EGFR cells (Nakamura et al., Nat. Biotechnol., 22(3) :331-6 (2004)), CHO-SLAMF1 (Tatsuo et al., Nature, 406(6798):893-7 (2000)), CHO-dogSLAMF1 (Seki et al., J. Virol., 77(18) ):9943-50 (2003)), CHO-NECTIN4 (Liu et al., J. Virol., 88(4):2195-204 (2014)), CHO-CD38 (Peng et al., Blood, 101: 2557-62 (2003), CHO-HER2/neu (Hasegawa et al., J. Virol., 81(23):13149-57 (2007), Ramos B-cell Burkitt lymphoma cells (ATCC, cat. no. CRL -1596) and Raji cells (ATCC, cat. no. CCL-86) were cultured in Roswell Memorial Park Institute (RPMI) 1640 medium (Corning Inc., cat. no. 10-040-CV, Corning, NY, USA), as described elsewhere.

Plasmids and genome-wide construction of rMeV.

To create CDV 22458/16 expression plasmids, total RNA was isolated from CDV 22458/16 isolate-infected Vero/dog SLAMF1 cells (passage 1) using the RNeasy Mini Kit (Qiagen, Hilden, Germany). Both CDV-H and CDV-F genes were reverse transcribed with SuperScript III reverse transcriptase (Thermo Fisher Scientific, cat. no. 11752050) and amplified by PCR using the following primers:

CDVH7050 (+): AGAAAACTTAGGGCTCAGGTAGTCC;

CDVH8949 (-): TCGTCTGTAAGGGATTTCTCACC;

CDVF4857 (+): AGGACATAGCAAGCCAACAGG;

CDVH7050 (-): GGACTACCTGAGCCCTAAGTTTTCT.

PCR products were directly sequenced by Sanger (Genewiz, Plainfield NJ, USA) and cloned into the pJET1.2 vector (Thermo Fisher). The open reading frame of CDV H (Fig. 8) was then amplified by PCR with a forward primer (5'-CCG GTA GTT AAT TAA AAC TTA GGG TGC AAG ATC ATC GAT AAT GCT CTC CTA CCA AGA TAA GGT G-3') and a reverse primer (5'CTA TTT CAC ACT AGT GGG TAT GCC TGA TGT CTG GGT GAC ATC ATG TGA TTG GTT CAC TAG

- 18 046760

CAG CCT CAA GGT TTT GAA CGG TTA CAG GAG-3'), and cloned at the PacI and SpeI sites (New England Biolabs, Iswich MA, USA) into the pCG vector (Cathomen et al., J. Virol., 72(2) :1224-34 (1998)) using the InFusion HD kit (Takara, Shinagawa, Tokyo, Japan). The primers yielded PacI and SpeI restriction sites (underlined, respectively) as well as the MeV-H untranslated region (italics) of MeV-H. Similarly, the open reading frame of CDV F (Fig. 9; amino acid residues 136-662 in SEQ ID NO: 4) was cloned into plasmid pCG-CDV-F at the HpaI/SpeI sites (von Messling et al., J. Virol., 75(14):6418-27 (2001)). The resulting plasmid pCG-CDV F 22458/16 contains the MeV-F untranslated region and the MeV-F signal peptide.

Expression plasmids for the CDV vaccine H/F Onderstepoort and isolate 5804 (von Messling et al., J. Virol., 75(14):6418-27 (2001)), as well as MeV Nse are described elsewhere (Cathomen et al. , J. Virol., 72(2):1224-34 (1998)). Specificity-redirected H protein variants were created by inserting a homologous PacI/SfiI-digested PCR product into pTNH6 vectors (Nakamura et al., Nat. Biotechnol., 23(2):209-14 (2005); and Nakamura et al. , Nat. Biotechnol., 22(3):331-6 (2004). Site-directed mutagenesis (QuickChange site-directed mutagenesis kit, Agilent Technologies, Santa Clara CA, USA) was used to eliminate tropism in H as well as to remove the SpeI site in CDV-F and introduce truncations in the endoplasmic tails.

rMeV envelope protein replacement viruses were generated by shifting the PacI/SpeI and NarI/PacI regions of the corresponding expression plasmids. rMeV rescue was performed using the START system (Nakamura et al., Nat. Biotechnol., 23(2):209-14 (2005)).

Protein expression.

Cells were transfected using Fugene HD (PROMEGA, Fitchburg WI, USA) or TransIT-LT1 transfection reagent (Mirus Bio LLC, Madison WI, USA). For quantitative fusion analysis, a split reporter system was used (Kondo et al., J. Biol. Chem., 285(19):14681-8 (2010); and Ishikawa et al., Protein Eng. Des. Sel. , 25(12):813-20 (2012)), as described elsewhere (Munoz-Alia et al., Viruses, 11(8), pii: E688, doi: 0.3390/v11080688 (2019)), using BHK cells as effector cells. For semiquantitative assessment of fusion, Vero cells and derivatives were transfected with 1 μg of each H and F expression plasmids and 1 day later stained with Hema-Quik (Thermo Fisher Scientific, cat. no. 123-745). Images were obtained using a microscope (Eclipse Ti-S; Nikon) at 4* magnification. Alternatively, a GFP expression plasmid was included to further visualize syncytia formation. To assess H polypeptide levels, transfected cells were examined by flow cytometry or CELISA using a 6 ^/ His-tagged monoclonal antibody (Miltenyi Biotec, cat. no. 130-120-787 or Thermo Fisher Scientific, cat. no. MA1-135) , as described elsewhere (Munoz-Alia et al., Viruses, 11(8), pii: E688, doi: 10.3390/v11080688 (2019); and Saw et al., Methods, 90:68-75 (2015) ). To assess total protein expression by flow cytometry, cells were treated with eBioscience Intracellular Fixation & Permeabilization buffer (Thermofisher, cat. no. 88-8823-88).

Viral protein content.

Virus preparations were heated in the presence of dithiothreitol, fractionated in a 4-12% Bis-Tris polyacrylamide gel and transferred to polyvinylidene fluoride membranes. Blots were probed with anti-MeV-Hcyt antibody (Cathomen et al., J. Virol., 72(2):1224-34 (1998)), anti-MeV-N antibody (Toth et al., J. Virol., 83 ). (2):961-8 (2009)) or His-tag antibody (Genscript, Piscataway NJ, USA, cat. no. A01857-40) and labeled with a conjugated rabbit secondary antibody (ThermoFisher, cat. no. 31642). Blots were incubated with SuperSignal Wester Pico chemiluminescent substrate (ThermoFisher) and analyzed using a ChemiDoc imaging system (Bio-Rad).

Neutralization assays.

The fluorescent focus suppression neutralization assay was used as described elsewhere (Munoz-Alia et al., J. Virol., 91(11): e00209-17 (2017)). Polyclonal serum against canine distemper virus, Lederle avirulent (ferret antiserum), was obtained through BEI resources (NR-4025). The human serum used was mixed from 60 to 80 donors who were specific blood type AB (Valley Biomedical Products & Services, Inc, Cat. No. HS1017, Lot C80553).

Results.

Construction of the fusogenic H/F CDV complex.

CDV envelope glycoproteins share 36% (H polypeptide) and 66% (F polypeptide) amino acid homology with MeV glycoproteins. The open reading frame of the H and F polypeptides was obtained from the first passage wild-type CDV isolate SPA.Madrid/22458/16 (CDV 22458/16) from postmortem tissues of a moribund dog (SPA.Madrid/22458/16). Phylogenetic analysis of full-length hemagglutinin genes by maximum similarity method showed that the H polypeptide of CDV 22458/16 is clustered in the Arctic clade (Fig. 1). Cotransfection of H/F complexes from 22458/16 revealed fusion activity only in the presence of the SLAMF1 receptor (Fig. 2A). A lack of fusion activity was observed on nectin-4-expressing cells. Coexpression of heterotypic H

- 19 046760 wild-type CDV polypeptide 5804 with CDV F polypeptide 22458/16 resulted in marked syncytia formation in Vero cells expressing nectin-4. On the other hand, H/F complexes from the large plaque forming variant of the Onderstepoort vaccine strain resulted in the formation of large syncytia independent of nectin-4 and SLAMF1 expression.

Using a reporter gene to more accurately identify syncytium formation, the above fusogenic phenotype was confirmed for the heterotypic combination of wild-type CDV H ₅₈₀₄ polypeptide and CDV _F2245 8/i ₆ polypeptide (Fig. 2B). Although the lack of syncytium formation for homotypic H/F ₅₈₀₄ may be due to the presence of a native 135 amino acid signal peptide, the distinct fusogenic phenotype observed for CDV H/F _22458/ i ₆ , which induces cell-to-cell fusion exclusively in a SLAMF'1-dependent pathway, was also examined . The clone used for the CDV H _22458/ i ₆ polypeptide was found to contain an amino acid substitution compared to the consensus sequence (M437L). This clone was used to create a clone encoding the CDV H _22458/16 polypeptide with a leucine to methionine substitution at position 437.

Expanding the tropism of the CDV fusion apparatus.

Because cotransfection of wild-type CDV H/F complexes, but not Ondertepoort vaccine-derived H/F complexes, resulted in syncytia formation in a specific receptor-dependent manner, the following was done to determine whether receptor use could be extended to alternative receptors. CD38 was chosen as the target receptor, and for this purpose a CD38-specific scFv was displayed in the C-terminal domain of the attachment protein (Fig. 2C). For comparison purposes, the MeV H polypeptide as well as the Nipah G polypeptide were included in the analysis. The results presented in Fig. 2D demonstrate that the different constructs were expressed on the surface at comparable levels. It should be noted that the L437M substitution in CDV H _22458/16 did not appear to affect cell surface expression. Fusion efficiency was then compared quantitatively using a self-associating split-luciferase assay described elsewhere (Kondo et al., J. Biol. Chem., 285:14681-14688 (2010); and Ishikawa et al., Protein Eng., Des. Sel., 25:813-820 (2012)). In this assay (Fig. 2E), effector cells were transfected with expression plasmids for the H/F complex and one half of the bisected GFP/Renilla luciferase protein (DSP1-7). Similarly, target cells expressing the corresponding receptor were transfected with the other half (DSP8-12). When the contents are mixed, the non-functional halves of the GFP/Renilla luciferase protein are combined and the activity is measured. The results of this experiment with effector cells expressing various H or G/F complexes are shown in FIG. 2F. Whereas no fusion activity was observed using the parental CHO cell line, fusion activity was evident in recombinant CHO cells with stable expression of either the HIS-specific scFv (CHO-HIS) or the CD38 molecule. Overall, activity levels were more pronounced using CD38 targeting compared to the 6*His-anti-6*HIS scFv pseudoreceptor system. Such differences were more noticeable with G ^aCD38 Hunax. The fusion ability of the former was significantly reduced compared to the heterotypic CDV H5804“ ^CD38 polypeptide/CDV F22458/16 polypeptide or homotypic H/F from the Onderstepoort strain. Surprisingly, the L437M substitution in the CDV H22458/i6 polypeptide caused a dramatic change in the fusogenic phenotype of the polypeptide, even though no difference in expression levels was observed (Fig. 2G). This new fusogenic ability was now equivalent to that obtained with the homotypic CDV H/F of 5804 only when the signal peptide of the CDV F5804 polypeptide was replaced by the shorter signal peptide of MeV F. However, fusion levels for the heterotypic combination of CDV H5804“ ^CD38 polypeptide/CDV F22458/i ₆ polypeptide were higher and were similar to those obtained for the large plaque forming variant Onderstepoort. These results demonstrate not only that the CDV fusion machinery can be modified to utilize alternative receptors, but also that a hyperfusogenic phenotype can be generated from a heterotypic combination of H/F complexes from different CDV strains in a receptor-dependent manner. This increased fusogenic capacity of the CDV H/F complex was achieved by shortening the signal peptide of the CDV F polypeptide and using a CDV polypeptide having M437.

Directed interaction with receptors.

The CDV H polypeptides described in the previous paragraph can still use human nectin-4 as a receptor. To disrupt the interaction with nectin-4, a nucleic acid encoding the CDV H polypeptide containing the Y539A mutation was generated. In fig. 2H shows that the CDV H ^aCD38 polypeptide, including the Y539A point mutation (CDV HY539A ^aCD38 ), lost fusion activity on cells with human nectin-4, but retained the ability to induce fusion on CHO-HIS cells (CDV Hy539a“ ^CD388 and CDV Hy5 ₃ 9a ) and CHO-CD38 (CDV Hy5 ₃ 9a“ ^CD388 ). This fusion activity was comparable to that obtained using the MeV H polypeptide with fully redirected specificity (MeV Haals ^aCD38 and MeV Haals “ ^CD38 ”). These results

- 20 046760 demonstrate that the CDV H/F complex can be efficiently retargeted to specific receptors.

Effect of ligand binding affinity on H/F CDV-directed cell fusion.

To evaluate the effect of differences in binding affinity of ligand exposed on the ectodomain H of the CDV polypeptide on fusogenicity, Her2/neu-specific scFv binding fragments as well as affibody molecules were exposed (Hasegawa et al., J. Virol., 81(23): 13149 -57 (2007); Wikman et al., Protein Eng. Des. Sel., 17(5):455-62 (2004) and Orlova et al., Cancer Res., 66(8):4339-48 (2006) )). In fig. 2I shows that binding affinity greater than 1 nM was required to initiate fusion in CHO cells modified to stably express Her2/neu molecules, but not in the parental cell line. This was true regardless of the nature of the binding molecule, whether it was in the form of a scFv molecule or an affibody. To investigate the relationship between receptor density and binding molecule affinity, quantitative fusion assays were repeated with a set of cancer cell lines expressing varying levels of Her2/neu molecules on the surface: HT1080 (1.2*104), Sko3pi (1.5*105) and TET67L (4.3*103). While exposure of the affibody with the highest binding affinity (Z342, 0.022 nM) allowed H CDV to initiate fusion in all cell lines tested regardless of receptor density, Z4 (50 nM) enabled cell-to-cell fusion only in Skov3pi cells, which expressed the highest receptor density. These results demonstrate that there is a relationship between binding molecule affinity and receptor density on the target cell: lower receptor density has an effect with higher binding molecule affinity.

CD46-targeted CDV H/F complexes can overcome sensitivity to oncolytic measles virus neutralization.

Various CD46-specific binding scFvs were displayed on CDV H polypeptides in an attempt to obtain scFv-CDV H polypeptides maintaining a fusion level similar to the MeV H Nse strain. Cell surface expression levels were compared (Fig. 3A). Cellular enzyme-linked immunosorbent assay (CELISA) showed that both non-targeting MeV H polypeptides and CDV H polypeptides, as well as CD46-targeting CDV H polypeptides, were similarly expressed on the cell surface. Quantitative fusion assays then showed that while only MeV H polypeptides resulted in fusion activity in CHO-nectinD cells, all but nontargeting CDV H polypeptides caused fusion in CHO-CD46 cells (Fig. 3B). Binding scFvs A10, A09, G09, and K2 exposed on CDV H polypeptides induced fusion levels similar to those observed in MeV H polypeptides, whereas scFV G101469 and K01 showed comparatively lower levels. The CDV H-scFv A09/CDV F polypeptide was selected to replace the existing MeV envelope. This virus, herein referred to as Stealth 2.0, was rescued and the H/F polypeptides were confirmed to be successfully displayed on virions. Western blot analysis confirmed that MeV H polypeptide was detected only in MeV when MeV H polypeptide antibody was used against the cytoplasmic tail. In contrast, the CDV H-scFV A09 polypeptide was detected only when the membrane was labeled with an antibody against the 6*HIS tag (Fig. 3C). The same HIS tag system allowed Stealth 2.0 virus replication in Vero cells stably expressing anti-HIS scFv, but not in the parental Vero cell line (Fig. 3D). Replication kinetics were comparable to those obtained with MeV on Vero/hSLAM. Together, these results demonstrate that the H and F envelope polypeptides of measles virus can be replaced by the H and F polypeptides from CDV without adversely affecting viral replication.

The neutralization sensitivity of Stealth 2.0 was tested using mixed sera from 20-30 US donors. Antiserum against CDV was used as a control. In fig. 3E shows that Stealth 2.0 was insensitive to the neutralizing activity of MeV antiserum. In contrast, the pattern of neutralization of MeV was essentially the opposite, being neutralized by anti-measles antibodies but not by anti-CDV antibodies.

The following was performed to confirm the virus tropism provided by the new envelope, since virus entry could occur in the absence of obvious fusion. Virus-induced GFP autofluorescence was observed when CHO cells expressed the receptors CD46, nectin-4, and canine or human SLAMF1 ( Fig. 3F ). Conversely, Stealth 2.0-directed GFP autofluorescence was observed only when scFvs were expressed against 6*HIS (CHO-HIS), CD46, and canine SLAMF1. These results demonstrate that human CD46-tropic measles virus that is resistant to neutralization by anti-measles virus antibodies can be produced using a CD46-targeted H/F CDV envelope.

Stealth 2.0 induces the same antitumor effect as the original MeV.

To evaluate the use of Stealth 2.0 as an oncolytic agent, the following was performed. SCID mice bearing U266.B1 tumors received a single intravenous dose of either MeV or Stealth 2.0 (Fig. 4A). The implanted tumors in the PBS group continued to grow exponentially (Fig. 4B), and as a result, all animals in this group had to be sacrificed due to tumor burden on day 12 (Fig. 4C). In contrast, in both treatment groups the tumor

- 21 046760 progressed more slowly, resulting in a significant increase in median survival. Since similar oncolytic activity was observed for Stealth 2.0 and MeV, and the latter could use nectin-4 and SLAMF1 receptors in addition to CD46, these results demonstrate that targeting CD46 is sufficient for tumor regression in a multiple myeloma model. Moreover, Stealth 2.0 demonstrated that it can replace the existing oncolytic MeV vaccine when patients have high levels of neutralizing antibodies against measles virus.

CDV H/F complexes may redirect the specificity of other mononegaviruses.

To investigate the suitability of CDV H/F complexes for controlling the tropism of VSVs, which are members of the genus Lysasavirus, family Rhadoviridae, the following was performed. VSV engineered to express interferon beta (IFN-β) and sodium iodide symporter (NIS), VSV-hlFNe-NIS (Naik et al., Mol. Cancer Ther., 17(1):316-326 (2018)) was produced and modified by replacing the VSV-G polypeptide with CDV H and F polypeptides using methods described elsewhere (Ayala-Breton et al., Hum. Gene Ther., 23(5):484-91 (2012)). CDV polypeptide F _22458/16 and either the original CDV polypeptide H ₅₈₀₄ (VSV-CDVFH-GFP) or CDV H polypeptide redirected against EGFR (VSV-CDVFHaal-oEGFR-GFP) or CD38 (VSV-CDVFHaal-oCD38-GFP) were used ) (Fig. 5). In addition, each of the CDV H polypeptides contained the R529A mutation (CDV Haa) in the existing CDV H _Y539A background to eliminate interaction with canine SLAMF1.

To confirm that the novel envelope protein complex directed the tropism of the virus, a panel of CHO cells expressing specific receptors was infected. As shown in FIG. 6, when the virus exposed the parent CDV F/H complex, GFP autofluorescence was observed in those cells that expressed either the nectin-4 receptor or the canine SLAMF1 receptor. In contrast, when the EGFR-specific CDV H scFv was present, only infection and GFP autofluorescence were observed in those cells that expressed the EGFR receptor. Likewise, when the CD38-specific CDV H scFv was present, only infection and GFP autofluorescence were observed in those cells that expressed the CD38 receptor. Moreover, the formation of syncytia and the cell killing effect were observed in cells expressing CD38, but not in CHO cells expressing EGFR receptors; whereas in the case of the EGFR-specific virus (VSV-CDVFHaaloEGFR-GFP), the opposite pattern was observed (Fig. 7). These results in the context of rhabdoviruses demonstrate that CDV F/H complexes can be used to direct cell entry and syncytia formation in a receptor-specific manner through scFv display using a preferential receptor to redirect viral tropism.

The use of this system as an oncolytic vector was also evaluated in vivo (Fig. 8). Nude mice bearing SKOV3ip.1 tumors received a single dose of EGFR-targeted VSV or VSV-hlFNe-NIS, which are currently in clinical trials. While VSV-hlFNeNIS did not improve survival compared with PBS-treated controls, EGFR-targeted VSV resulted in a significant improvement in survival (p<0.005). These results demonstrate that the targeted VSVs described herein can be applied in the field of oncology.

CD38 and EGFR-targeted MeV.

To confirm that targeting CD38 and EGFR in the context of MeV can be achieved using CDV F and H polypeptides (Fig. 11), the following was performed. To improve safety, additional mutations were introduced into the CDV H polypeptide to prevent potential reversion to natural receptors (Sawatsky et al., J. Virol., 92(15):e0069-18 (2018)). In addition to the amino acid substitution R529A, substitutions D526A, l527A, S528A, R529A, Y539A, Y547A, and T548 were also included (Fig. 11A). In fig. 11B shows that introduction of these mutations in the context of the CD38-directed CDV H polypeptide did not affect fusogenicity. Likewise, in FIG. 11C shows that various point mutations did not affect the sensitivity of the polypeptides to neutralization by human measles-immune serum. Although the MeV CD38-directed H polypeptide fusion activity was inhibited by mixed measles antisera up to a 1:80 dilution, the homologous CD38-directed CDV H polypeptide was not inhibited at the highest concentration tested at the 1:10 dilution. As with rhabdoviruses, measles virus comprising CDV H/F targeting complexes can induce infection and cell fusion in a CD38-specific or EGFR-specific manner (Fig. 11D). This infection specificity has also been observed in a variety of tumor cell lines. Skov3pi and U87 (EGFR-positive) cells were infected with EGFR-targeting viruses but not with CD38-targeting viruses. In contrast, Raji and Ramos cell lines (CD38 positive) were exclusively infected with CD38-targeting viruses. As a control, cells expressing the 6*HIS pseudoreceptor were infected with all specificity-redirected viruses. These results demonstrate that CDV F/H complexes can

- 22 046760 used to evade measles immunity and to direct specific penetration into cells and the formation of syncytia.

The oncolytic activity of CD38 and EGFR-targeted MeVs in vivo was also examined. SKOV3ip.1 tumors were implanted either subcutaneously or intraperitoneally into athymic nude mice, followed by virus challenge by the same route (Fig. 11E). Although CD38-targeted viruses showed some therapeutic efficacy, the efficacy shown by EGFR-targeted viruses was greater (Fig. 11F). In contrast, CD38-targeted viruses did not demonstrate antitumor activity after intraperitoneal injection, whereas EGFR-targeted viruses resulted in complete tumor regression, which was observed with 100% survival. Of note, no differences were observed between MeVs with specificity redirection by either the H/F MeV or H/F CDV complexes. These results demonstrate that CDV H/F complexes can enhance the antitumor properties of oncolytic measles virus without the problem of neutralization by measles-induced neutralizing antibodies.

Example 2. Further analysis of CD46-specific oncolytic measles viruses resistant to neutralization by human measles serum.

This example repeats some of the information and results from Example 1 in addition to presenting additional results.

Heterologous combinations of wild-type CDV glycoproteins result in enhanced cell membrane fusion.

The MeV envelope was replaced with an alternative viral envelope that may allow the virus to escape neutralization by measles antibodies. For this purpose, wild-type CDV was selected. Although there is a strain of CDV approved for use in vaccines (strain Onderstepoort), this strain may use a currently unidentified receptor in addition to SLAMF1 and nectin-4 (Fig. 14), which may make it difficult to change the tropism of the virus. Therefore, the focus was on wild-type strains since they are known to interact exclusively with SLAM and nectin-4.

To identify the most fusogenic CDV H/F glycoprotein pair, the following was performed. Various H/F combinations from isolates 5804P and SPA.Madrid/16 (hereafter referred to as 5804 and SPA, respectively) were transiently expressed in Vero cells expressing either SLAMF1 or NECTIN4, and the extent of cell fusion (formation of syncytia) induced by viral proteins was assessed qualitatively. Fusion activity was not observed for CDV-H/F pairs when the 135 amino acid signal peptide was retained in CDV-F (Fig. 15A). When replaced with a homologous sequence from MeV F, coexpression of H/F proteins from 5804P resulted in fusion of cells expressing SLAMF1 and NECTIN4, whereas coexpression of such proteins from SPA caused cell fusion exclusively in Vero cells expressing SLAMF1. On the other hand, coexpression of the heterologous combination of CDV-H 5804 and CDV-F SPA, but not CDV-H SPA and CDV-F 5804, resulted in the formation of syncytia in Vero cells expressing SLAMF1 and NECTIN4. The data above provided evidence in support of fusion failure on the CDV-H SPA. To begin investigating whether the lack of a CDV-H SPA fusion on NECTIN4-expressing cells was due to lower receptor affinity, fusion phenotypes were quantitatively compared across non-natural receptors, thereby equalizing the conditions for receptor binding affinity. This approach consisted of fusing a His-tagged CD38-specific scFv to the C-terminal domain of receptor-binding proteins and determining fusion levels in CHO cells encoding either CD38 or a pseudoreceptor for an HIS tag (CHO-oHIS). The L437M substitution was introduced into the CDV-H SPA because L437 corresponded to a lineage-specific mutation not present in any other CDV genetic group (Fig. 16). For comparison purposes, receptor binding proteins with redirected specificity from other viruses were additionally included: MeV-H- and NipahG (Bender et al., PLoS Pathog., 12(6):e1005641 (2016); and Nakamura et al., Nat. Biotechnol. ., 22(3):331-6 (2004). To make comparisons more rigorous, expression of receptor binding proteins was initially analyzed by Western blotting and flow cytometric analysis, which showed no significant effect on protein folding or surface expression (FIGS. 17A and 17B). When CDV-H/F pairs from SPA proteins were expressed in CHO-C38, only CDV-H SPA with M437L induced fusion (Fig. 15B). Notably, no significant difference in fusion ability was observed between the two homotypic CDV-H/F pairs from isolates SPA or 5804P. Surprisingly, the fusion activity initiated by the heterologous combination H/F CDV-H 5804/F SPA was superior to the fusion activity achieved by the homologous combinations CDV-H/F 5804 and CDVH/F SPA. In the case of the Nipah G/F pair with redirected specificity, although significant fusion levels were observed in CHO-CD38 cells (p<0.0001), they were not significant compared to the levels obtained using the non-redirected H/F OL pair CDV.

Based on this series of experiments, the hyperfusogenic pair CDV-H 5804/F SPA was selected for further study and modification.

- 23 046760

The strength of CDV H/F interaction is inversely proportional to the efficiency of cell-cell fusion.

The enhanced cell membrane fusion observed for the CDV-H 5804/F SPA pair may be due to lower binding avidity at the H/F contact region. This was based on the observation that H/F dissociation is required for the fusion process (Plemper et al., J. Virol., 76(10):505161 (2002) and Bradel-Tretheway et al., J. Virol., 93( 13) (2019)). To test this hypothesis, the relative strength of association of different combinations of CDV-H and F proteins was assessed using coimmunoprecipitation (co-IP) assays. To facilitate detection of CDV-F, SPA was fused to a FLAG tag, which did not affect the biological activity of the protein (Fig. 18). The results presented in Fig. 15C showed that the presence of the M437L mutation in CDV-H SPA weakened its affinity for CDV-F SPA. In addition, the affinity of CDV-F SPA for CDV-H 5804 was slightly lower than the affinity of CDV-F SPA for CDV-H SPA (Fig. 15C). Taken together, these data indicate the presence of an inverse correlation between fusogenicity levels and the strength of the CDV-H/F interaction.

Fully redirected CDV envelope glycoproteins exhibit comparable fusion activity to MeV glycoproteins.

The CDV-H proteins described above can still use nectin-4 as a receptor (Fig. 15A). To maximize the efficiency of specificity redirection, it is necessary to eliminate the CDV-H protein from this unwanted interaction with human cells. To determine whether eliminating this natural tropism would affect cell fusion caused by CDV-H/F binding to a non-natural receptor, the Y539A mutation was introduced into CDV-H, which corresponds to Y543A in MeV-H, a mutation that has previously been shown to exclude nectin-4-dependent fusion (Mateo et al., J. Virol., 87(16):9208-16 (2013)) and does not affect cell surface expression (Sawatsky et al., J. Virol., 86(7) :3658-66 (2012)). The efficiency of the CDV-H 5804 (Y539A) 5804/F SPA fusion was then compared with that of the MeV-H/F pair with fully redirected specificity (Nakamura et al., Nat. Biotechnol., 23(2):209-14 (2005)) . For this purpose, a quantitative and kinetic fusion assay was used, based on the separation of the GFP/luciferase reporter protein into two parts. The data presented in Fig. 15D indicate that CD38-directed CDV-H 5804 (Y539A) does not have fusion activity in CHO-hcktuh-L cells but induces fusion in CHO-oHIS cells (CDV H 5804 (Y539A)/F SPA and CDV H 5804 (Y539A) construct aCD38/F SPA) and CHO-CD38 cells (CDV H 5804 (Y539A)aCD38/F SPA). Since the fusion activity of the nectin-4-blinded CDV-H 5804/F SPA pair was comparable to that obtained with the fully redirected MeV-H protein, it was concluded that CDV-H 5804 (Y539A) can effectively redirect the specificity of the CDV-H complex /F to specific receptors, and this protein was selected to be incorporated into the virus with completely redirected specificity.

Binding affinity determines the efficient redirection of CDV H/F complexes specificity toward CD46.

Given that the CDV-H protein selected as described above can be efficiently redirected to CD38 by fusion with a CD38-specific scFv, this protein was then redirected to CD46 by display of a CD46-specific scFv. It was hypothesized that C-terminal display on a CDV-H scFv that recognizes CD46 with sufficiently high binding affinity would result in CD46-mediated cell-cell fusion activity similar to that induced by the MeV H/F complex. To test this, anti-CD46 scFvs with high affinity for CD46 were identified by assessing binding to purified CD46 of several different scFv variants isolated from a phage display antibody library (FIG. 19). Surface plasmon resonance technology was used using sensor chips with covalently immobilized anti-Fc antibody. Chimeric Fc-scFv fusion proteins were captured on the surface of the sensor, which was then reacted with soluble CD46 including SCR1-4 (Fig. 20A). Under these assay conditions, the results showed that the affinity constants (Kd) of chimeric Fc-scFv fusion proteins containing fragments A09 and K2 were significantly higher than those with fragments K01 and N1E (A09>K2>N1E>K2), primarily from -due to an increased rate of association (A09) or a lower rate of dissociation (K2) (Table 4, Fig. 20B).

Table 4

Affinity and kinetic rate constants for single chain variable fragment binding to CD46

Interaction Ko _n x10 ⁴ (M ^-1 s ^-1 ) KoffX10 ^-3 (s ^-1 ) Kd(hM) scFv A09-CD46 3.34±1.68 0.606±0.07 21.4±7.82 scFv K2-CD46 24.0±3.79 10.8±1.81 44.8±13.3 scFv N1E-CD46 3.69±1.28 5.16±2.21 139±22.6 scFv K1-CD46 4.13±0.77 11.5±2.58 287±97.5

- 24 046760

Association (Kon) and dissociation (Koff) reaction rates were determined using the BIACORE T100 instrument using a 1:1 Langmuir binding model.

To determine whether scFv fusion to CDV-H protein can support CD46-dependent fusion, and if so, how CD46 binding affinity influences cell fusion, the following was performed. The primary approach was to perform quantitative fusion assays for specificity-ablated CDV-H [5804 (Y539)] and specificity-redirected CDV-H [5804 (Y539)-scFv]/F pairs and compare them to the unmodified MeV/F complex on CHO cells and CHO cells expressing either nectin-4 or CD46. All proteins were expressed at comparable levels (Fig. 21). With the exception of scFv K01, all other anti-CD46 scFvs allowed the CDV-H/F complex to induce cell-to-cell fusion in CHO-CD46 cells (Fig. 20C) and in a HeLa cell line with high CD46 expression (Fig. 21). Only the MeV-H/F complex induced cell-to-cell fusion in €ΉΟ-ηοκ™η-4 cells.

From this series of experiments, it was concluded that there is a binding affinity threshold for CD46-mediated cell-cell fusion via redirected CDV H/F complexes, and above this threshold there is a positive correlation between binding affinity and cell-cell fusion.

CD46-targeted CDV envelope glycoproteins are efficiently incorporated into MeV virions and mediate viral entry according to their binding affinity.

To examine whether higher affinity for the receptor translates into higher infectivity of the virus, the following was done. To begin to address this question, a panel of isogenic MeVs was generated where the MeV envelope was replaced with CDV-F SPA along with CDV-H 5804 (Y539A) exhibiting low-affinity (K1), medium-affinity (N1E), and high-affinity (A09) scFvs specific for CD46 (Fig. 22A). These Stealth viruses were further engineered to express either eGFP or firefly luciferase as reporter genes and were rescued in Vero-oHIS cells. The ability of Stealth viruses to infect CHO-CD46 cells or a producer cell line (Vero-oHIS) was then assessed. The data presented in Fig. 22B showed that all three Stealth viruses effectively induced syncytia formation on Vero-oHIS cells, but only Stealth-A09 formed syncytia on CHO-CD46 cells. Stealth-N1E produced small clusters of unfused GFP-positive CHO-CD46 cells, whereas Stealth-K1 did not appear to infect them (Fig. 22B). Using luciferase as an infection reporter, CD46-dependent infection of Stealth-K1 was detected, but luciferase levels were significantly lower than those obtained by infection of Stealth-A09 cells (Figs. 22C and 23), indicating that binding to CD46 determines the efficiency of virus penetration.

The Stealth-A09 virus (this virus is referred to as Stealth 2.0 in Example 1) was chosen for further studies because it showed excellent CD46-dependent entry. Stealth-A09 replicated in Vero-oHIS cells but not in the parental Vero cell line, indicating efficient viral replication through the HIS pseudoreceptor and lack of interaction with African green monkey CD46 (Fig. 22D). To assess the relative particle infectivity of Stealth-A09 relative to the parent MeV, Western blot analysis of virus preparations was performed to detect the major structural protein (N), and no significant differences in expression levels were observed (Fig. 22E). Because equivalent numbers of virus particles were analyzed, the data indicate similar particle-to-plaque-forming unit (PFU) ratios for the two viruses, indicating efficient incorporation of foreign envelope protein into the virions. The receptor-specific tropism of the viruses was then fully examined by infecting a panel of CHO cells stably expressing human or canine oHIS, CD46, nectin-4, or SLAMF1. The results presented in Fig. 22F showed that MeV infected CHO cells expressing CD46 and nectin-4 receptors and either canine or human SLAMF1, whereas Stealth-A09 infected only CHO-oHIS, CHO-CD46 and CHOdogSLAM cells, indicating that Stealth-A09 is efficiently retargeted to human CD46.

Consistent with a recent report showing that CDV-H does not bind human SLAMF1 (Fukuhara et al., Viruses, 11(8) (2019)), no entry of MeV-Stealth-A09 into CHOhSLAMF1 was observed (Figure 22F). However, adaptation to human SLAMF1 by another CDV strain was observed (Bieringer et al., PLoS One, 8(3):e57488 (2013)). Therefore, to assess the likelihood of adaptation of MeVStealth to human SLAMF1, the virus was sequentially passaged in Vero-hSLAMF1 cells and the tropism of the virus was analyzed. Before tropism testing, infected cells were blindly passaged eight times with an interval of 5 days. This number of passages was previously shown to be sufficient to induce cycling adaptation of measles virus quasispecies (Donohue et al., PLoS Pathog., 15(2):e1007605 (2019)). As shown in FIG. 22G, at the end of this selective pressure, MeV-Stealth-A09 was still able to induce syncytia formation only in Vero-dogSLAMF1 cells, but not in Vero or Vero-hSLAMF1 cells, where only occasional GPF-positive cells were observed. Thus, the evidence argues against a potential adaptation allowing the use of the pathogenic human receptor SLAMF1.

- 25 046760

Taken together, these results indicate that MeV H/F glycoproteins can be replaced by CD46-redirected CDV-H/F glycoproteins and that cell entry is dependent on receptor affinity.

The oncolytic activity of MeV-Stealth depends on its binding affinity to CD46.

To determine the antitumor potential of Stealth viruses and the role of CD46 binding affinity in vivo, the following was performed. For this purpose, athymic mice carrying SKOV3ip.1 tumors disseminated in the peritoneum, expressing the firefly luciferase gene (SKOV3ip.Fluc), were intraperitoneally injected once with saline or 10 ⁶ TCID5 ₀ Stealth-N1E and Stealth-A09 (n=5). Tumor burden was then monitored using in vivo bioluminescence imaging (FIG. 24A). At day 7, animals treated with either Stealth-N1E or Stealth-A09 showed a comparable reduction in tumor burden compared to the control group (Fig. 24B). However, the statistical significance of the reduction in tumor burden was lost for the Stealth-N1E group by day 21, whereas this was not the case for the Stealth-A09 group. Comparison of survival curves showed that only MeV Stealth-A09 increased the survival of mice compared to the control group. Thus, Stealth-A09 exhibits oncolytic activity, which appears to depend on its higher affinity for CD46.

MeV-Stealth provides oncolysis and prolongs survival in mice bearing myeloma and ovarian tumors.

To evaluate the utility of MeV-Stealth-A09 versus MeV as an oncolytic agent, the following was performed. This was done by initially leaving untreated (PBS group) or treating severe combined immunodeficiency (SCID) mice bearing subcutaneous human myeloma xenografts (derived from U266.B1 cells) with a suboptimal intravenous dose of MeV-Stealth or MeV. Tumors in the PBS group continued to show exponential growth and all mice had to be sacrificed due to tumor burden on day 12 (FIG. 25A). Treatment with either MeV or MeV-Stealth-A09 slowed tumor progression, resulting in a significant increase in median survival of 7 and 5 days, respectively (Fig. 25B). To assess whether the oncolytic effect was due to viral replication, histological examination of the removed tumors was performed. The results presented in Fig. 25C clearly indicate that both viruses are capable of colonizing tumor tissue. These results strongly suggest that CD46-targeted Stealth induces oncolysis at similar levels compared to MeV, which targets both CD46 and SLAMF1, thereby inducing tumor regression in this multiple myeloma model.

The therapeutic effect of MeV-Stealth-A09 in prolonging survival in the presence of measles serum was then assessed. Before undertaking this in vivo study, the sensitivity of the recombinant viruses to in vitro neutralization was first assessed. The results presented in Fig. 26 showed that MeV-Stealth-A09 was insensitive to the neutralizing activity of human anti-measles serum, whereas it was completely neutralized by ferret or mouse anti-CDV serum. A substantially opposite picture was observed for MeV (Fig. 23). SKOV3ip.Fluc cells were then implanted into the peritoneal cavity of athymic nude mice that received either PBS or human measles serum, followed by a single intraperitoneal injection of MeV or MeV-Stealth (FIG. 27A). Animals that did not receive any viral injections, as well as animals that were injected with MeV in the presence of immune serum, showed high bioluminescent activity that continued to increase over time (Fig. 27B), indicating that MeV could not cause oncolysis in the presence of pre-existing immunity. In contrast, Kaplan-Meier survival curves showed that in the absence of immune serum, both oncolytic virus (OV) treatments significantly prolonged survival in mice, with a median survival of 37 days for MeV-treated mice and 53 days for Stealth-treated mice, according to compared to a median survival time of 18 days in control mice (Figure 27C). However, the prolongation of survival following MeV treatment was completely negated when mice received anti-measles serum, whereas MeV-Stealth treatment still provided a significant increase in survival (median survival time of 28 days) with no statistical difference when compared with treatment without immune serum .

These results demonstrate that targeting CD46 induces oncolysis in both a xenograft model of myeloma and an orthotropic model of ovarian cancer, and that replacement of the MeV envelope with a homologous CDV-H/F fusion machinery spares MeV from the action of human anti-MeV immune serum.

Methods and materials.

Cell lines.

Newborn hamster kidney cells (BHK, cat. no. CCL-10, ATCC, Manassas, VA, USA), human epithelial kidney cells (HEK293T), obtained from Dr. Francois-Loic Cosset, Universite de Lyon ), and the human ovarian cancer cell line SKOV3ip.1-Fluc (Mader

- 26 046760 et al., Clin. Cancer Res., 15(23):7246-55 (2009)) were maintained in Dulbecco's modified Eagle's medium (DMEM; cat. no. SH30022.01, GE Healthcare Life, Pittsburg, PA, USA) supplemented with 5% fetal bovine serum (FBS; cat. 10437-028; Thermo Fisher Scientific, Waltham, MA, USA). African green monkey kidney cells Vero (Vero, ATCC, cat. no. CCL-81) and their derivatives (expressing human nectin-4 (Noyce et al., Virology, 436(1):210-20 (2013)), SLAMF1 human (Ono et al., J. Virol., 75(9):4399-401 (2001)) or a membrane-bound single chain variable fragment (scFv) specific for hexahistidine peptide (6*HK tags) (Nakamura et al. , Nat. Biotechnol., 23(2):209-14 (2005)) were cultured in DMEM (cat. no. SH30022.01, GE Healthcare Life Sciences) as described elsewhere (Munoz-Alia et al. , Viruses, 11(8) (2019). Vero cells constitutively expressing the canine SLAMF1 molecule (VerodogSLAMF1) were generated by transduction and puromycin selection of a second generation lentiviral vector (kindly provided by Dr. Lukkana Suksanpaisan, Imanis Life Science, Rochester, MN, USA) encoding control of the focus-forming spleen virus codon-optimized canine (Canis lupus familiaris) SLAMF1 molecule (GenBank NP_001003084.1) with an N-terminal FLAG tag sequence (DYKDDDD). Cells were maintained in DMEM supplemented with 5% FBS. Chinese hamster ovary (CHO) cell line, CHO-CD46 cells, CHO-hSLAMF1 cells, CHO-dogSLAMF1 cells, CHO-nectin-4 cells, CHO-aHIS cells, CHO-CD38 cells, and human myeloma cell line U266.B1 (courtesy provided by Dr. David Dingli, Mayo Clinic, Rochester, MN) were grown in RPMI 1640 supplemented with 10% FBS. Cells were incubated at 37°C in an atmosphere of 5% CO2 at saturating humidity.

Plasmids and construction of whole-genome recombinant measles virus (MeV).

To create expression plasmids based on the canine distemper virus (CDV) SPA.Madrid/22458/16, total RNA was isolated from Vero/dog SLAMF1 cells (passage 1) infected with the CDV isolate SPA.Madrid/22458/16 (passage 1) using the RNeasy Mini Kit (Qiagen , Hilden, Germany). Both CDV hemagglutinin (H) and CDV fusion (F) genes were reverse transcribed using SuperScript III Reverse Transcriptase (cat. no. 11752050, Thermo Fisher Scientific) and amplified by PCR using the following primers: CDVH7050(+): 5'AGAAAACTTAGGGCTCAGGTAGTCC-3' CDVH8949(-): 5'-TCGTCTGTAAGGGATTTCTCACC-3'; CDVF4857(+): 5'-AGGACATAGCAAGCCAACAGG-3' and CDVH7050(-): 5'GGACTACCTGAGCCCTAAGTTTTCT-3'. PCR products were sequenced directly using Sanger sequencing (Genewiz, Plainfield NJ, USA) and cloned into the pJET1.2 vector (Thermo Fisher Scientific). The CDV-H open reading frame was then PCR amplified with a forward primer (5'-CCG GTA GTT AAT TAA AAC TTA GGG TGC AAG ATC ATC GAT AAT GCT CTC CTA CCA AGA TAA GGT G-3') and a reverse primer (5' -CTA TTT CAC ACT AGT GGG TAT GCC TGA TGT CTG GGT GAC ATC ATG TGA TTG GTT CAC TAG CAG CCT CAA GGT TTT GAA CGG TTA CAG GAG-3'), and cloned into PacI- and SpeI-digested (New England Biolabs, Ipswich , MA, USA) pCG vector (Cathomen et al., J. Virol., 72(2):1224-34 (1998)) using the InFusion HD kit (Takara, Shinagawa, Tokyo, Japan). The primers contained PacI and SpeI restriction sites (underlined), as well as the coding sequence of the MeV-H untranslated region (italics). Similarly, the open reading frame of CDV-F (amino acid residues 136-662) was cloned into the HpaI/SpeI digested plasmid pCG-CDV-F (von Messling et al., J. Virol., 75(14):6418-27 ( 2001)). The resulting plasmid pCG-CDV-F SPA.Madrid/22458/16 contained the coding sequences of the MeV-F untranslated region and the signal peptide. Expression plasmids for the CDVH/F Onderstepoort vaccine and isolate 5804P (von Messling et al., J. Virol., 75(14):6418-27 (2001)), as well as the MeV Nse strain, have been described elsewhere (Catomen et al. ., J. Virol., 72(2):1224-34 (1998)). The signal peptide for CDV-F 5804 was replaced by heterologous MeV-F as described above for CDV-F SPA.Madrid/22458/16. The open reading frames of the Nipah-G and Nipah-F glycoprotein genes were amplified from commercial RNA templates (cat. no. NR-37391, BEI Resources), and the Nipah-F gene (GenBank AF212302.2) was inserted into the pCG vector using NarI sites and PacI. Specificity-redirected H/G protein variants were generated by inserting a homologous PacI/SfiI digested PCR product into pCGHX a-CD38 (Peng et al., Blood, 101(7):2557-62 (2003)). The insertion of the coding sequence of the scFv recognizing CD46 was made by replacing the anti-CD38 scFv at the SfiI and NotI restriction sites. Site-directed mutagenesis (QuickChange Site-Directed Mutagenesis Kit, Agilent Technologies, Santa Clara CA, USA) was used to eliminate H tropism and remove the SpeI site in CDV-F.

The viruses used in this example were derived from a molecular cDNA clone of the Moraten/Schwart vaccine strain pB(+)MVvac2(ATU)P with an additional transcription unit after the phosphoprotein gene (Cathomen et al., J. Virol. 72(2):1224 -34 (1998) and Munoz-Alia et al., Viruses, 11(8) (2019)). To avoid instability of the plasmid during propagation in bacteria and to increase the efficiency of virus rescue, the plasmid backbone was replaced with the pSMART LCkan vector (cat. no. 40821-1; Lucigen, Middleton, WI, USA) with an optimized T7 promoter followed by

- 27 046760 hammerhead self-cleaving ribozyme (Hrbz) (Beaty et al., mSphere, 2(2) (2017); and Munoz-Alia et al., Viruses, 11(8) (2019)). eGFP or firefly luciferase was cloned into an infectious clone using unique MluI/AatII restriction sites. Rescue of rMeV particles was performed using the START system (Nakamura et al., Nat. Biotechnol., 23(2):209-14 (2005)).

Expression of recombinant proteins.

A plasmid encoding the CD46-Fc fusion protein was prepared by fusing the CD46 ectodomain (residues 35–328) with the Fc domain of IgG1 (pfc1-hg1e3; InvivoGen, San Diego, CA, USA). ScFv K1, K2 and A09 were constructed with VL and VH sequences separated by a flexible linker GSSGGSSSG, which were codon optimized and synthesized and cloned into pUC57-Kan (GenScript). A fourth scFv (N1E) was constructed with VH and VL sequences separated by the SSGGGGS linker, which was codon optimized, synthesized at Creative Biolabs (Shirley, NY), and cloned into pCDNA3.1+ (Invitrogen). For IgG constructs, scFv was cloned into the unique AgeI and KpnI sites in pHL-FcHIS (cat. no. 99846, Addgene, Cambridge, MA, USA), carrying the secretion signal coding sequence and the C-terminal fragment of the human Fc region followed by 6*NK-mark. Recombinant proteins were expressed by transfecting a suspension of Expi293F cells (Thermo Fisher) in serum-free Expi293 expression medium (Thermo Fisher) in conical flasks according to the manufacturer's instructions. Culture supernatants containing recombinant proteins were collected and passed through a Protein G chromatography cartridge (cat. no. 89926, ThermoFisher). Bound recombinant proteins were eluted with 0.1 M glycine (pH 2.0), followed by immediate neutralization with 1 M Tris (pH 8.0), and the isolated proteins were concentrated using an Amicon Ultra centrifugal concentrator (Millipore Sigma, Burlington, MA, USA). CD46 and nectin-4 were released from the Fc region by incubation with HRV 3C protease (Thermo Fisher) at a ratio of 1:200. The final purification step was performed using a Superdex 75 10/300 gel filtration column (GE Healthcare) equilibrated with phosphate-buffered saline (PBS). Protein concentrations were calculated from the protein extinction coefficient determined from the amino acid composition.

Fusion assays.

Cells were transfected using Fugene HD (PROMEGA, Fitchburg WI, USA) or TransIT-LT1 transfection reagent (Mirus Bio LLC, Madison WI, USA). For quantitative fusion analysis, a split reporter system was used (Kondo et al., J. Biol. Chem., 285(19):14681-8 (2010)); and Ishikawa et al., Protein Eng. Des. Sel., 25(12):813-20 (2012)), as described elsewhere (Munoz-Alia et al., Viruses, 11(8) (2019)), using BHK cells as effector cells. For semiquantitative assessment of fusion, Vero cells and derived cell lines were transfected with a total of 0.1 μg of DNA (1:1 ratio of H to F expression plasmids) including a GFP expression plasmid to further visualize syncytia formation. Images were obtained using a microscope (Eclipse Ti-S; Nikon) at 40* or 100* magnification.

Expression analysis of Morbillivirus attachment proteins.

To assess the level of polypeptide H, transfected cells were analyzed by flow cytometry or cell enzyme-linked immunosorbent assay (CELISA) using a monoclonal antibody against the 6xIi tag (cat. no. 130-120-787, Miltenyi Biotec or cat. no. MA1-135, Thermo Fisher Scientific) as described elsewhere (Munoz-Alia et al., Viruses, 11(8) (2019); and Saw et al., Methods, 90:68-75 (2015). To analyze total protein expression with Using flow cytometry (FACSCanbt, BD Biosciences, San Jose, CA, USA), cells were treated with eBioscience Intracellular Fixation & Permeabilization buffer (cat. no. 88-8823-88, Thermo Fisher Scientific).

Coimmunoprecipitation of envelope glycoprotein (co-IP).

Three micrograms (1 μg H and 2 μg F) of total DNA were transfected into HEK293T cells (4e5 cells). After 24 h, cells were washed twice with PBS and treated with 1 mM 3-3'diothiobis(sulfosuccinimidyl propionate) cross-linker (DTSSP; cat. no. 21578, Thermo Fisher Scientific), followed by quenching with 20 mM Tris/HCl (pH 7). ,4) and lysis using 0.4 ml of M-PER Mammalian Protein Isolation Reagent (Thermo Fisher Scientific) containing 1 ^x Halt Protease and Phosphatase Inhibitor Mixture (cat. no. 1861281, Thermo Fisher Scientific). Soluble fractions were collected after centrifugation at 10,000 g for 10 min at 4°C, and one-thirtieth of the volume was retained as the original cell lysate. The remaining material was incubated with 0.5 μg of anti-FLAG monoclonal antibody M2 (Sigma-Aldrich) and EZview red protein G affinity gel (Sigma-Aldrich, St. Louis, MO, USA). The precipitated material was washed (20 mM Tris-HCl, pH 7.4, 140 mM sodium chloride) and denatured by boiling in Laemmli buffer containing β-mercaptoethanol.

SDS-PAGE electrophoresis and immunoblotting.

Samples were fractionated by gel electrophoresis on a 4-12% NuPAGE Bis-Tris gel (Thermo Fisher) and transferred to polyvinylidene difluoride (PVDF) membranes using the iBLOT 2 dry blotting system (cat. no. IB21001, Thermo Fisher Scientific). Protein material

- 28 046760 was detected by incubation with antibodies against MeV-H606 (Hudacek et al., Cancer Gene Therapy, 20(2):109-16 (2013)), against MeV-F431 (von Messling et al., J. Virol. , 78(15):7894-903 (2004)), anti-Fcyt (von Messling et al., J. Virol., 75(14):6418-27 (2001)), anti-MeV-N (Toth et al. , J. Virol., 83(2):961-8 (2009)), anti-HIS (Cat. No. A01857-40, GenScript, Piscataway NJ, USA), anti-β-actin (Cat. No. A3854, Sigma -Aldrich) and against CD46 (cat. no. sc-7056, Santa Cruz, Dallas TX, USA). Immunoblots were visualized using a horseradish peroxidase (HRP)-conjugated secondary antibody and a KwikQuant Imager (Kindle Bioscience LLC, Greenwich CT, USA). Representative results from two independent replicates are shown. Bands were quantified using a KwikQuant Image Analyzer 1.4. (cat. no. D1016, Kindle Biosciences, LLC).

Antibody binding assay.

An enzyme-linked immunosorbent assay (ELISA) was used to measure scFv binding to CD46. Nunc-Immuno MicroWell 96-well plates were coated overnight at 4°C with 1 μg of purified CD46 or N4 in 0.05 M carbonate-bicarbonate buffer, pH 9.6 (cat. no. E107, Bethyl Laboratories, Montgomery TX, USA ). Purified scFv-Fc fusion proteins were then diluted in PBS and added at a concentration of 12.5 μg/ml. Bound antibodies were detected with a secondary (Fc-specific) HRP-conjugated anti-human IgG antibody (1:70,000; cat number A0170, Sigma-Aldrich). In parallel, 125 ng of scFv-Fc fusion protein was first bound in the wells and protein levels were monitored by measuring the optical density (OD490 nm) after incubation with the secondary antibody alone.

Surface plasmon resonance (SPR).

The interaction between scFv A09, 2B10, K1, K2 and CD46 was measured using CM5 S-series sensor chips on a Biacore T-100 system (GE Healthcare, Waukesha, WI, USA). For A09, N1E, K1 and K2, 50 μg/ml anti-Fc antibody (MAB1302, EMD Millipore, Burlington, MA, USA) diluted in 10 mM Na acetate, pH 4.5, was immobilized on the active and reference channel of the CM5 chip with using the amino group coupling kit reagents (EDC (1-ethyl-3-[3-dimethylaminopropyl]carbodiimide), NHS (N-hydroxysuccinimide) and ethanolamine). Antibody immobilization resulted in ~12,000 response units. Interactions between CD46 and scFv captured by anti-Fc antibody were measured at 25°C with a data rate of 10 Hz using HBS EP buffer (0.01 M HEPES, pH 7.4, 0.15 M NaCl, 3 mM EDTA, 0.005% v/v surfactant P20). Each binding cycle began with application of 15 μg/ml scFv to the active channel for 300 s at a flow rate of 10 μl/min. After a buffer wash for 100 s and a stabilization period of 120 s, CD46 (concentration range 50–1000 nM for A09, 37.5–100 nM for K2, and 50–1000 nM for N1E and K1) was passed over the active and reference channels for 100 s at a flow rate of 40 µl/min. The association phase was followed by a 200 s dissociation period followed by 10 mM glycine, pH 2.0, for 60 s at a flow rate of 30 μL/min to regenerate the anti-Fc antibody immobilized on the surface. All sensorgrams were fit to a 1:1 binding model using Biacore T100 data analysis software version 2.04.

Infection and analysis of virus growth kinetics.

For virus infection, cells were infected at the indicated MOIs for 90 min at 37°C in reduced-serum Opti-MEM I medium. After the absorption phase, the inoculum was removed, washed, and viral growth medium (DMEM+5% FBS) was added. When using eGFP-expressing viruses, fluorescence microscopy photographs were taken 48 h postinfection. During infection with Fluc-expressing viruses, luciferase expression was measured using an Infinite M200 Pro Multimode Microplate Analyzer (Tecan Trading AG) after adding 0.5 mM D-luciferin to the infected cells.

To analyze virus growth kinetics, Vero cells and derivative cell lines seeded in 6-well plates 16 to 18 h before infection were infected at a multiplicity of infection (MOI) of 0.03 for 90 min in Opti-MEM (cat. no. 31985070, Thermo Fisher Scientific). The inoculum was then removed, cell monolayers were washed three times with Dulbecco's phosphate-buffered saline (DPBS; cat. no. MT21-031-CVRF, Mediatech, Inc., Manassas, VA, USA) and the medium was replaced with 1 ml DMEM supplemented with 5% FBS. At the indicated time points, cell supernatants were collected and cells were scraped into 1 ml of Opti-MEM, followed by 3 freeze/thaw cycles. Cellular debris was removed by centrifugation (2000*g for 5 min) and virus titers in Vero-aHIS cells were determined.

Immunization research.

Male and female four- to six-week-old HuCD46Ge-IFNARKO mice (Mrkic et al., J. Virol., 74(3):1364-72 (2000)), deficient in the type I IFN receptor and transgenically expressing human CD46, inoculated intraperitoneally (i.p.) with 1*10 ⁵ TCID ₅₀ particles of MeV or Stealth-A09. On day 28, serum samples were collected and stored at −20°C until assessed for neutralizing antibodies.

Neutralization assays.

Fluorescent plaque microneutralization (PRMN) assay was performed as described elsewhere (Munoz-Alia et al., J. Virol., 91(11) (2017)). Briefly, Vero-aHIS cells were seeded at 96

-

Claims (2)

well plate, and serial dilutions of serum samples were premixed for 1 h at 37°C with the viral inoculum before being added to the cells. The data were plotted as logarithm (serum dilution) versus normalized response (variable slope) using GraphPad (Prism 8) and a 50% neutralizing dose (ND50) was calculated. The inclusion of the World Health Organization International Serum Standard 3 (3 IU/ml) allowed antibody titers to be converted to mIU/ml when calculating the unit conversion constant (Haralambieva et al., Vaccine, 29(27):4485-91 (2011)). Mixed human serum from 60–80 AB blood type donors was used (cat. no. HS1017; lot no. C80553, Valley Biomedical Inc., Winchester, VA, USA). The following reagents were obtained from the Biodefense and Emerging Infections Research Resource Repository of the US National Institutes of Health, NIAID, NIH: MeV polyclonal antibody, Edmonston (guinea pig antiserum), NR-4024 and CDV polyclonal antibody, Lederle avirulent (antiserum , ferret), HP-4025. The lack of cross-neutralization between measles virus and Stealth was assessed (Fig. 28A-B). Experimental oncolytic therapy. To engraft subcutaneous tumors, 6-week-old female mice with severe combined immunodeficiency (SCID) were injected with P107 U266.B1 tumor cells into the right flank. When the tumor reached a diameter of 0.5 cm, mice received a single intravenous dose of MeV (n=5) or Stealth (n=5) at a dose of 1χ10 ⁵ 50% of the tissue culture infectious dose (TCID ₅₀ ). Control mice (n=5) received an equal volume of PBS. Animals were sacrificed when the tumors ulcerated or when the load reached 20% of body weight. Tumor diameter was measured every two days, and tumor volume was calculated using the formula length ^x length ^x width ^x 0.5. To create an orthotopic model of ovarian cancer, 5 ^χ 10 ⁶ SKOV3ip.1 cells expressing firefly luciferase (SKOV3ip.1-Fluc) were injected into the abdominal cavity of athymic nude mice. Ten days later, animals were treated with 600 mIU measles serum (cat. no. HS1017; lot no. C80553, Valley Biomedical Inc.) or an equal volume of saline, and three hours later they were given a single intraperitoneal dose of (1χ10 ⁶ TCID ₅₀ ) MeV (n=5 ) or Stealth (n=5). Mice in the control group received a similar volume of Vero cell lysates (n=5). For the therapeutic experiment, 5χ10 ⁶ SKOV3ip.1-Fluc cells were implanted instead. The tumor was monitored weekly using in vivo bioluminescence imaging using an IVIS Spectrum instrument (Perking Elmer, Waltham, MA, USA). Mice were sacrificed at the end of the study (80 days) when they developed ascites or lost 20% of their body weight. Statistical comparisons between groups were performed using the log-rank (Mantel-Cox) test, and p < 0.05 was considered statistically significant. Statistical analysis. Statistical analysis was performed using GraphPad Prism version 8.3.1 for Mac OS X. Significant differences between groups were determined using one-way analysis of variance (ANOVA) with the Holm-Šidák multiple comparisons test. Survival data were analyzed using the Kaplan-Meier method, and the log-rank test was used to identify significant differences between groups. Other embodiments It should be understood that although the invention has been described in connection with the detailed description thereof, the foregoing description is intended to illustrate and not to limit the scope of the invention, which is defined by the scope of the appended claims. Other aspects, advantages and modifications are included within the scope of the following claims. CLAIM 1. A CDV F polypeptide comprising SEQ ID NO: 4, provided that said CDV F polypeptide does not contain at least amino acid residues 1-60 or does not contain at least amino acid residues 1-105 of SEQ ID NO: 4. 2. A replication-competent vesicular stomatitis virus comprising an RNA molecule, wherein said RNA molecule includes a nucleic acid sequence that is a template for a positive sense transcript encoding a VSV N polypeptide, a nucleic acid sequence that is a template for a positive sense transcript encoding VSV P polypeptide, a nucleic acid sequence that is the template for the positive sense transcript encoding the VSV M polypeptide, a nucleic acid sequence that is the template for the positive sense transcript encoding the CDV F polypeptide, a nucleic acid sequence that is the template for the positive a sense transcript encoding a CDV H polypeptide, and a nucleic acid sequence that is a template for a positive sense transcript encoding a VSV L polypeptide, wherein said RNA molecule does not contain a nucleic acid sequence that is a template for a positive sense transcript encoding a functional G -