AU2005209571A1 - Methods and compostions of matter concerning APRIL/G70, BCMA, BLYS/AGP-3, and TACI - Google Patents

Description

-1-

AUSTRALIA

PATENTS ACT 1990 COMPLETE SPECIFICATION FOR A STANDARD PATENT

ORIGINAL

Name of Applicant/s: Actual Inventor/s: Amgen Inc.

Lars Eyde Theill and Gang Yu Address for Service is: SHELSTON IP Margaret Street SYDNEY NSW 2000 CCN: 3710000352 Attorney Code: SW Telephone No: Facsimile No.

(02) 9777 1111 (02) 9241 4666 Invention Title: METHODS AND COMPOSITIONS OF MATTER CONCERNING BCMA, BLYS/AGP-3, AND TACI Details of Original Application No. 2001261557 dated 14 May 2001 The following statement is a full description of this invention, including the best method of performing it known to me/us:- File: 37147AUP01 500680675 1.DOC/5844 la- METHODS AND COMPOSITIONS OF MATTER CONCERNING APRIL/G70, BCMA, BLYS/AGP-3, AND TACI The present application is a divisional application of Australian Application No. 2001261557, which is incorporated in its entirety herein by reference.

This application claims the benefit of U.S. Provisional Application Serial No.

60/204,039, filed May 12, 2000 and U.S. Provisional Application Serial No. 60/214,591, filed June 27, 2000, which are incorporated herein by reference.

Field of the Invention The present invention relates to proteins that are involved in inflammation and immunomodulation, survival, or activation or in lymphoproliferative disorders or other cancers. The invention further relates to proteins related to the tumor necrosis factor (TNF)/nerve growth factor (NGF) superfamily and related nucleic acids, expression vectors, host cells, and binding assays. The specification also describes compositions and methods for the treatment of immune-related and inflammatory, autoimmune and other immune-related diseases or disorders, such as rheumatoid arthritis Crohn's disease lupus, and graft versus host disease (GvHD) as well as for treatment of lymphoproliferative diseases and other cancers.

Background of the invention After years of study in necrosis of tumors, tumor necrosis factors (TNFs) ax and 3 were finally cloned in 1984. The ensuing years witnessed the emergence of a superfamily ofTNF cytokines, including fas ligand (FasL), CD27 ligand (CD27L), ligand (CD30L), CD40 ligand (CD40L), TNF-related apoptosis-inducing ligand (TRAIL, also designated AGP-1), osteoprotegerin binding protein (OPG-BP or OPG ligand), 4-1BB ligand, LIGHT, APRIL, and TALL-1. Smith et al. (1994), Cell, 76: 959- 962; -2- Lacey et al: (1998), Cell 93:165-176; Chihepotiche et al. (1997), T. Biol.

Chem, 272:32401-32410; Mauri et at (1998), Immunity, 8: 21-30; Hahne et A (1998), T. Ex. Med., 188: 1185-90; Shu et al. (1999), T. Leukocyte Biology, 680-3. This family is unified by its structure, particularly at the Cterminus. In addition, most members known to date are expressed in immune compartments, although some members are also expressed in other tissues or organs, as well. Smith et al. (1994) Cell 76: 959-62. All ligand members, with the exception of LT-o, are type II transmembrane proteins, characterized by a conserved 150 amirio acid region within C- 1.*c terminal extracellular domain. Though restricted to only ?0-25% identity, the conserved 150 amino acid dmain folds into a characteristic n-pleated sheet sandwich and trimerizes. This conserved region can be proteolyticaly released, thus generting a oluble functional form Banner et al. (1993), Cell, 73; 431-445.

Many members within this ligapdn family are expressed in lymphoid enriched tissues and play nimportant roles in the immune system development and modulation. Smith et al. (1994). For example, TNFat is mainly synthesized by macrphages and is animportant mediator for inflanmatry responses and immune defenses. Tracey &Cerami (1994), Annu. Rev. Med., 45: 4917-503. Fas-L, predominantly expressed in activated T cell, modulates TCR-mediated apoptosis of thymocyts.

Naga, S Suda, T. (1995) Immunology Today 16:39-43; Castrim et al.

Immunity 5:617-27. CD40L, also expressed by activated T cells, provides an essential signal for B cell survival, proliferation and immunoglobulin isotype switching. Noelle (1996), ImmunitLy,4:415-9.

The cognate receptors for most of the TNF ligand family members have been identified. These receptors share characteristic multiple cysteine-rich repeats within their extracellular domains, and do not possess catalytic motifs within cytoplasmic regions. Smith et al. (1994).

3 The receptors-signal through direct interactions with death domain proteins TRADD, FADD, and RIP), or with the TRAF proteins (eg& TRAF2, TRAF3, TRAF5, and,TRAF6), triggering divergent and overlapping signaling pathways, apoptosis, NF-iB activation, or JNK activation. Wallach et al. (1999), Annual Review of Immunology 17:331- 67. These signaling events lead to cell death, proliferation, activation or differentiation.. The expression profile of each receptor member varies.

S For example, TNFR1 is expressed on a broad spectrum of tissues and cells, S- whereas the cell surface receptor of QPGL is mainly restricted to the osteoclasts. Hsu et al. (1999) Proc. Natl Acad. Sd. USA, 96:3540-5. Such Sproteins are believed to play a role in inflammatory and immune processes, suggesting their usefulness in treating autoimmune and inflammatory disorders.

A number of research groups have recently identified TNF family ligands with the same or substantially similar sequence; but they have not identified the associated receptor. The ligand has been variously named neutrokine-a (WO 98/18921, published May 7,1998), 63954 (WO 98/27114, published June 25,1998), TL5 (EP 869 180, published October 7, 1998), NTN-2 (WO 98/55620 and WO 98/55621, published December 1998), TNRLl-alpha (WO 9911791, published March 11, 1999), kay ligand (W099/12964, published March 18,1999), and AGP-3 Prov. App.

Nos. 60/119,906, filed February 12; 1999and 60/166,271, filed November 18; 1999, respectively). Each of these references is hereby incorporated by reference Hereinafter, this protein sequence is referred to as "AGP-3." A recent paper has identified two'previously known proteins as receptors for AGP-3. Gross etal. (2000), Nature 404: 995-9. The first receptor was previously identified as a lymphocyte surface receptor named Transmembrane Activator and CAML Interactor (TACI). See WO 98/39361, published September 11,1998, and von Bulow Brai (1997), -4- Sience 278:138-140, each of which is hereby incorporated by reference in its entirety., According to these references, TACI binds an intracellular cclophiiih ligand designated CAML, which modulates the calcium sighaling pathway in lymphocytes.

The second receptor identified forAGP-3 is the so-called B cell maturation protein (BCMA). The human BCMA gene was discovered by molecular arialysis of a t(4;16) trinslocation, which characteristic of a human T celllymphoma. Laabi et al. (1993), EMBO T. 11: 3897-3904. BCMA mRNA was reported to be found mainly iin lymphoid tissues. Human BCMA cDNAericodesa 184 amino acids protein (185 residues for the mouse) and the.literature reports no obvious'similarity with any known protein or motif; ard its ftmnctionrLimained unikniiwi. The protein was reported to reside in the Golgi apparatus (Gras eal (1995), Intl. Immunol.

7: 1093-1106)..'Recent speculation suggested that BCMA may be a distant member of the TNFR super family. Madryiit A (1998)y, Itl. Immunol. 1693-170Z A ligand. called APRIL or G70 is a TNF famiily ligald that remains without a receptorreported in the literattife. Accordiig.to the literature, APRIL is associated with prostate cancer, breast cancer, Alzheimer's 'disease, immune disorders, iriflaminatory disorders, and gestational abnormalities See WO 99/00518 (June 26, 1997); WO 99/11791 (Sept. 1997); WO 99/12965 (Sept..12; 1997); EP.911 633 (Oct. 8,1997); EP 919 620 S (Nov. 26,1997); WO 99/28462 (Dec. 3,1997); WO 99/33980 (Dec. 30,1997); WO 99/35170 (Jan 5, 1998); arid Hahne et al.. (1998), T. Exp. Med. 188: 1185-90. (Each of the foregoing references is hereby incorporated by reference in its entirety.) A recent paper described APRIL isoforms and suggested that APRIL causes cell death. Kelly et al. (2000), Cancer Res. 1021-7. Theart would benefit from identification of a receptor for APRIL and a clarification of its activity.

5 Summary-of the Invention It has now been found that (APRIL) binds to cell-surface receptors on T and B lymphoma cells resulting in stimulation of proliferation of primary human and mouse B and T cells both in vitro and in vivo. It has now been found that BCMA and TACI are cell-surface receptors for APRIL. It has also been found that APRIL competes with AGP3's binding to TACI and BCMA.

Furthermore it is shown here that sBCMA inhibits APRIL and AGP3 binding tp its receptors. sBCMA ameliorates T cell dependent and T cell independent humoral immune responses in vivo. In addition it has now been found that sTACI inhibits APRIL and AGP3 binding to its receptors and ameliorates T cell dependent and T cell independent humoral immune responses in vivo. In addition it has now been:found that sBCMA reduces lymphoma and colon carcinoma cell tumor growth in vivo. It has also now been found that sBCMA iinceases survival and reduces incidence of proteinurea, and development of anti-dsDNA antibodies in an animal model of lupus. It has also beeni found that BCMA exhibits similarity with TACI within a single cysteine rich domain located N-terminal to a potential transmembrane domain. It has also been found that APRIL stiniulatesiB cell growth and immuinglobulin production in vitro and in vivo. Furthermore, treatment with a blocking anti-APRIL antibody ameliorates generation of antigen specific immunoglobulin, suggesting that endogenous APRIL is required for Shumoral immunity in vivo. This invention concerns novel methods of use and compositions of matter that exploit these discoveries. The discoveries provides a strategy for development of therapeutics for treatment of autoimmune diseases, and cancer, for prevention of transplant rejection.

These discoveries show that activity, disease states, and disease parameters associated with APRIL and AGP-3 may be affected by modulation of BCMA. Likewise, disease states and disease parameters -6associated with TACI can be affected by modulation of APRIL. Further, such disease states and disease. parameters cain be affected by modulation of any of TACI, BCMA, APRIL and AGP-3 together. This discovery further suggests molecules and methods of treatment by which more than 5 one of TACI, BCMA, APRIL, and AGP-3 may be modilated by a single molecule.

Description of the Figures Figure 1 shows the sequence of human APRIL (SEQ ID NOS: 1 and 2) Start and stop codons are underlined.' Figures 2A aind 2B show the DNA and amino acid sequences of mousd APRIL/G70 (SEQ ID NOS: 3 and 47-esp§ectively). Start and stop codonis are underlined. The amino acid equence of FAG-tagged soluble mouse APRIL (SEQ ID:NO: 19) is also provided.

Figure,3 shows an alignment of huiitan (SEQ ID NO: 22) and mouse (SEQ ID NO:f 23) APRIL. The middle lineof each row shows the consensus sequence (SEQ ID NO: 24).

Figure 4 shows that G70/APRIL is a potent stimuilator for B and T cell lymphoma Figure 4A shows doc&dependet sitiiulation of 2 0. proliferation of Jurkat cells (human leukemic T cells), Raji (human Burkitt lymphoma) and K562 cells (human chioic myeldgenous leukemia cells)..

The proliferation of 'cells was determinedby incubating 3 x 10' cells/well in 100 pd medium with indicated concentration of recombinant and phosphate-buffered saline (PBS, no ligand) as a control.

After 48 hours; the number of viable cells iwere measureidby Celiltiter 96 AQ proliferation assay (Promnega, Madisonri; WI). If Figure 4B, U937 cell (monoocyte -like leukemia cells) NIH/3T3 (imouse embryo cell line) and 293 (transformed humrian primary embryonal kidney cell line) did not respond to sG70/APRIL stimiulation: -7- .Figures 5A and 5B show FACS analysis of G70/APRIL receptor binding. G70/APRIL receptor expression was assessed on indicated cell line using anti-Flag monoclonal antibody followed by FITC-conjugated goat antibody to mouse IgG .A anti-mouse CD16/CD32 monoclonal antibody (Fc Block) was used to block non-specific binding to cells.

.Figure 6, shows the effect of sG70/APRIL on htuian peripheral blood B cell, T cell and granulocyte proliferation. Human peripheral T cell (CD4+- and CD8+), B cells, and granulocyte were purified from three different donors by.using RosetteSep cocktail antibodies (Stem cell Tech.

Vancquver). Purified cells were cultured in tissue culture-treated plastic ,wells( Becton-Dickinson, Lincoln park, NJ) for 6 Days in RPMI-1640 Smedium supplemented with 10% fetal calf serum-,2 mM L-glutamine and 2-ME (50uM) in the present different concentration of sG70/APRL. For the. B cell proliferation assay, plastic wells were coated with purified mouse anti-human IgrM monodclonal antibody (3 gg/ml, Pharmingen, San Diego, CA). The positive control for T cell stimulation is IL-2.

SFigure 7 shows.the effect.of G70/APRIL on murine T-and B-cell proliferation in.vitro. T-and B-cells from the spleens of C57B1 mice were purified by selection through a murine T-cell and B-cell enrichment 2Q Q.columns. 1x10 cells per well were cultured in the absence or presence of various G70/APRIL for 48 hours; pulsed during the last 18 hours with pCi K thymidine and hayested to count the incorporated radioactivity.

Figure 8 shows the effect of G70/APRIL on murine T cell proliferation costimulated though anti-CD28 antibody. T-cells from the spleens of C57B1 mice were purified by selection through a murine T-cell enrichment column. 1x10 T-cells per well were treated with in the absence or presence of subliminal concentration of anti-CD28 antibody (0.9 pg/ml) for 48 hours, pulsed during the last 18 hours with gCi 3 H thymidine and harvested to count the incorporated radioactivity.

-8 Figure 9 shows the effect'of 070/APRIL on murine T cell proliferation costimulated though, anti-CD antibody. T-cells from the spleens of C57B1 mice were purified by sele4ction through a murine T-cell enrichment column. 1x10 5 T-cells per well werte tieated with G70/A1'RIL in 'the 'absence or presence of subliminal concentraition r of antf-CD3 antibody pg/mnl) for 48 hours, pulsed 'during the last 18 hours with jii jthymidineiad har7vestedto count h ihmcorporated rAdioacfi Vity. Table1 shows FACS analysis of spleen (Table 1A), and miesentteric lymph nodes (Table iB) aftLer in vivo sytmcadministration o6f NF fami mmbers. Sveral, m ereis oTNF'family have been tested in vitvo: ~eac group have65nmice (BDF-.I ,gw 8 we2 of agDose: 1 mg/kg/ day 0.2 ml. for, 5 days t Splen, tliyius and mfiesenteric lymph ndes frm, three kof each grouip hakre b een iolite~ for FACS analysis using a panel of T cell andI B cell sufc it r atbdes. eut4fFC ;analysis haveDbeen summarized as following tables.

Figures 10k and 105 show the sequeneo hr. BCMA (SEQ If) NO:, BCMVAfs extracellulai- domamn (SEQ ID) NO: 6) extends from aa 1 to aa 51 and, is, identified by arrows.% The cysteinae-rc con~surein(E lD O:7, esribe-d further herei~tr isshwn 6n1odface. The transmfemrane region (SEQ IID NO: 8) is underliined. huBCMA-Fc (SEQ ID NO: 9mCMA-Fc (SEQ. IID NO: Figurerl11 shows an alignment of human BCMA amidno acd sequence and m urine BCMA: amidnoacid seunc (SEQ ID NO: 11). The huizmn sequenic6 is shown on the top line, the murinte on"the bottom line in. each row; The humnan-murine cons ensus sequence (SEQ ID) NO: 12) appears as the mriddle line of each row. A nthe conisensus, sequence indicates a conservative substitution. The cysteine-rich. portion of the consensus sequence (SEQ ID NO: 13) appears in boldface.

-9- Figuresl12A and 12B show the sequence of hTACI (SEQ ID NO: 14).

TACI's extracellular domain (SEQ ID NO: 15) exteds frdm aa 1 to aa 166.

The cysteine-rich consensus region (SEQ ID NO: 16) is shown in boldface, and the:tifansmembrane region (SEQ ID NO: 17) is underlined. hTACI-Fc (SEQ ID NO: 18).

Figure 13 shows an align ment of cysteine rich extracellular regions of human TACI and human BCMA. The BCMA cysteine rich consensus region (SEQ ID NO: 20) appears as the top line, the TACI cysteine rich -consensus region (SEQ ID NO: 21)' appears as the bottoniline of each row.

Conserived amino acid residues are indicated by a vertical bar Related amino acid residues are indicated with a colon Figures 14A, 14Band i4C -show soluble m6use G70/APRIL binding to 293 cells expressing the BC1A gene. Human,293 cells transfected with the pmBCMA and pcDNA3 vectors were incubated iith Flag, followed by FITC-coftjugated anti-Flag antibody staining for FACS analysis. AJ 293 cells transfected with pcDNA3 vector only. B. 293 cells transfected with antisense pmBCMA vector. C. 293 cells transfected with sense pnmBCMA vector..

Table 2 shows BIACore analysis of the stoichiometric binding kinetics of APRIL and AGP-3 to BCMA and TACL Flag-APRIL specifically binds to murine and human BCMA with affinities of 0.25 nM and 0.29 nM, respectively, and to human TAC iwith an affinity of 1.48 nM.

Also a longer version of Flag-tagged APRIL (aa 50-240) binds to BCMA and TACI with high affinity similar to that of Fc-AGP-3 (Table In 25 separate experiments, we determined that neither APRIL nor AGP-3 bind to OPG and also that TNFa, OPGL, LIGHT, TWEAK, anrid TRAIL. do not bind to BCMA or TACI. Hence, APRIL and AGP-3 specifically bind to both BCMA and TACI with high affinity.

10 shows G70/APRIL binding to 293 cells expressing the hTACI gene. Human 293 cells transfected with the phTACI and pcDNA3 vectors were incubated with G70/APRIL-Flag; followed by FITCconjugated anti-Flag antibody staining for FACS analysis. In Figure 293 cells were transfected with phTACI vector. In Figure 15B, 293 cells were transfected with pcDNA3 vector only.

.Figure 16 shows G70/APRIL completely blocks AGP3 binding to its receptor. Mouse B lymphoma cells A20 were stained with AGP3-Fc or plus 10 fold excess G70/APRIL, CD40 ligand; TRAILligand and Tweak.

After washing 3 times, cells were incubated with FITC-conjugated goat anti-human IgG-Fc secondary antibody. In Figure46Ai, 10 fold completely blocked AGP3 binding to'A20 cells. In Figures 16BC and. PD,10 fold CD40 ligand, Tweak and TRAIL do not have that effect on AGP3 binding.

Figure 17 shows.that soluble TACI receptor (sTACI) binding competes with G70/APRIL binding to A20 cells. A20 cells were incubated with G70/APRIL or at saime time plus 10 fold soluble;TACI, TRAIL R2 and TRAIL R3 receptor, followed by FITC-conjugated anti-Flag antibody staining for FACS analysis In Figuire 17A, soluble TACI receptor partially competed in binding G70/APRIL binding to A20 cells.In Figures 17B and C, soluble TRAIL R2 and TRAIL R3 receptors did not interfere with binding Figure 18 shows that soluble human BCMA-Fc receptor fusion protein (shBCMA-Fc) and soluble humart TACI-Fc receptor fusion protein (shTACI-Fc) completely blocks soluble human AGP-3-Fc' receptor fusion protein. (shAGP3-Fc) binding to A20 cells. A20 cells were incubated shAGP3-Fe or at same time plus 10 fold shBCMA-Fc or shTACI-Fc followed by FITC-conjugated anti-Flag antibody staining for FACS analysis.

11 Figure 19A shows shBCMA-Fe completely blocks shAGP3-Fc binding to A20 cells& A20 cells were incubated with shAGP3-Fc with or without 10 fold soluble hBCMA-Fc followed by-FITC-conjugated anti-Flag antibody staining for FACS analysis,. Figure 19B shows that shBCMA-Fe blocks soluble. mirine APRIL (smAPRIL) binding to A20 cells. A20 cells were incubated with smAPRIL with or without 10 fold shBCMA-Fc followed by FITC-conjugated anti-Flag antibody staining for FAGS analysis.

Figure 20 shows serum levels of anti-KLH1 IgG and IgM and anti- Pneumovax IgM in mice treated with TACI-Fc or BCMA-Fe fusion proteins or non-fused Fc as a control. p vales ifer to the comparison with the Fc-treated gfoup. n 7. See Matrials arid Methods hereinafter.

Figure 21 shows the sequence of Fc-hiianAPRIL SFigure 22 shows the effect of Fc-humanAPRIL and soluble huanAGP3/BlyS/Tall-1 on proliferation of primary murine B cells.

Purified murine spleen B cells were cultured in preseie of various amounts ofhuman Fc-APRIL and untagged AGP3 plus 2. gg/ml of anti- IgM Data show incorporatidn of 3H thyinidine as cpmand represent mean of triplicate wells.

Figitre 23 shows thathBCMA-Fc aid hTACI-Fc inhibits FlagmAPRIL mediated mouse B cellproliferation. Purified murine spleen B Scells were cultured with the indicated amounts of soluble BCMA-Fc and TACI-Fc in presence of 10 ng/inl of Flag-miPRIL and 2 .g/ml of an anti- IgM for 72 hr. Incorporation of 3H thymidiihe is indicated as cpm. Data shown represent meanof triplicate wells.

Figure 24 shows that administration of hBCMA-Fc (15 mg/kg intraperitoreally (ip- on day 0, 3, and 6) reduces mouse peripheral blood B cell levels measured at day seven. Normal mice were treated with 12 human BCMA-Fc and notf used Fc as a control in day 0,3, and 6 mg/kg). Peripheral blood B cell levels were measuted-bn day 7.

Figure 25 shows that administration of hBC4A,-Fc (15 mg/kg ip on day 0, 3, and 6) reduces mouse spleen B cell. levels measured at day seven.

ue 26 shows thatFiag-nAPRIL and hAGP3' mediated IgA production is ihibited by hBCMA-Fc and hTACI-Fc in vitio. Purified murine spleen B clswere cul tured with, LPS (100 nig/ffil),AGP3 (10 ng/nil),Flag- APRIL ng/rnl) or plus BCM4A-Fc (100 ng/mld) and TACI-Fc (100 ng/nil) for 12 slays., Culture sup~ra t wee collected 'on day 7 and day 12 for detecting IgA level..

Figure.27 shows that Vlag-ifAPRfL and hAGP3 mediated IgG production -is inhi-b ite d -by hB CMA-Fc and hT AC I-Fc in vitro.

Fgure 28. shows reduced total IgE and IgA levels In normal mice treated with mBCMA7Fc and truncated hTACI-Fc (5 mg/kg ip day 0, 3, and 6).

Normal mie(n=7)- were treated with huiriaxt BCMA-Fc ,trutcated TACI- Fc and~inonfused Pc as a contiol in day 0,3,1 and Imimunoglqbulin levels in serum were measured on day; 7.

Figure. 29 shows' that- treatment with hBCMA-Fc -or truncatedhTACI-Fc reduces anti-Pneumovacs specific IgM levels in normal mice.

Normal-Pice were treated with-hum BCMA-Fc, truncated TACI- Fc and. nonfused Pc as a. control in daily: doses of 0.5mnglkg to 15 mg/kg for, 7 days. At preniunztion anti KLH- and apti-Pnteumovax were undetectable. Antibodies were measured on. day 7 and day 14.

Figure 30 shows that the pA#t-mAPRJL specific monoclonal antibody #c19 inhibits Flag-n-APRTL mediated mouse B cell proliferation.

Purified murirte spleen B cells were cultured in presence 10 ng/nil niFlag- APRIL plus 2 ug/n-l of. anti-IgM. anti-Flag-wAPRIL monoclonal antibody c-19 and rat IgG control were added into culture in same time. Data show 13 *incorporation f.3H thymiddine ascpmand represent mean of triplicate wells.

Figure '31 shows that treatment with the anti-mAPRIL specific monoclonal antibody #c19 inhibits generation of anti-Pneumovacs specific antibodies in. vivo.

Figure 32 shows that that treatmnent with hBCMAFc increases survival in the NZB/NZWF1 motusermodel. of SLE.

Figure 33'shows that; that-treatment with hBCMA-Fc reduces the incidence o~f poroteinurea in the NZB/NZWF1 mousemodel of SLE.

Figure,34 shows that treatment with hBCMA-Fc. decreases antids]DNA specific antibody levels in the-NZB/NZWF1 mouse model of SLE.

Figure 35. shows that that treatment with hBCMA-Fc decreases peripheral bloodO/o cells in the NZB/NZWFI mou~emodel of SLE.

Fig 36 shows that APRIL binds to a number 6f tumor cell lines.

APRIL binding to, human tumor cell lines were determined byincubating cell wit lug/nil Fc-APRIL or Flag-AP1jIL following F1TC labeled -secondr antibodystaining, and FACS analss Figr 3.7 shows that APRL stimulates U266-B1 cell growth and that this can be inhibited by BCMA-Fc or TAq U-266,cefls were cultured in presence lOng/nil humxan Fc-APRIL or plus, 5Ong/rnl soluble BCMA-FcTruncated TACI-Fc for 48 hr. Incorporatio4 of 3H thymidine is indicated as cpm.L Data shown 7ersn menoPrpiaewls Figr.e 38,shows thatAPRIL stimulates mouse B lymphoma. cell growth and that this can be inhibited by BCMA-Fc or TACI-Pc. mouse B3 cell lymphom4 were cultured in presence 'Ong/rl mouse Flag- APRI~huma Flag-G '3 r plus 100 ng/rrl soluble B 2MA-Fc for 48 hr.

Incorporation of 3H thymiddine is indicated as cprn. Data shown represent mean of triplicate wells.

14 Figure 39 shows that treatment with BCMA-Fc reduces A20 B lymphomna tumor cell growth in Balb/c nmice. A20 (0.2 million) cells were implanted, id, on day 0. Treatments with PBS, CHO-Fc, mBCMA-Fc, hIBCMA-Fc, mTACI-Fc, hTACI-Fc were given (10 mg/kg, ip) on days 0, 7, 10, 13, 16,19,22. Tumor measurements were made twice per Week. Mice were sacrificed on days. 27-31, tumors were snap frozen for RNA isolation, blood was collected and serum samples were frozen.

Figure 40 shows thatttreatment with hBCMA-Fc reduces human &coln- carcinomna'cell line HT 29 tumior volumhe 'ro~th in mice. 2x10 6 cells plus 50% ,matrigel injected subcutaneouisly into athymic nude mice. Rx: hu n .~AF t2; 5; anfd'15 mg/kg Q215; stirting'at'day 0 Control 1: CHO-Fc at,15 mg/kg Q2D. Control 2: 0.2 ml of PBS Q2DIP1W Tunibr 'Volm: 3/week, from'd'y 7.'Tii weigfit at end of study. Body weight2/wek.l Figure 41 shows that treatment with iiiCMAFcreduces human colo cacn- 'cell ~ine HT29 tumor volume growth ininice. 2x10 6 cells plus 50% mtatrigel iject&6 sitan~oustrint6 at'?~ymic n ude mice. Rx: mi~ouse BC2MA-Pc' at 2; ,5,-;and 15 gk Q2D, start~ing at day 0 (n i0/grop4y Control1: CHO-FcR at15mgkgQ Contro 12: 0.2 ml of !PBS Q2j II. Tu~mor volume 3/wek fto day 7. tuni weight at end of'study. Body TWeiht'ZweekMetaie ecito of the Invention DjefiAnito of termsU The tr~s ued thoughot this- spdation are defined as follows, unIes oth erwiiitedin spedffic instaces.

termn "comprising' t means that a compound may include additional- aminoacids on either or both of te N4- or C- termini of the given sequence. Of course, the se adiinlamino acidssol o significantly interfere with the activity of the co mpound.t 15 "AGP-3" activity" refers to modulation of cell growth, survival, or activation resulting from binding by natural human AGP-3 to TACI or BCMA, particularly in B cells. Conversely, "AGP-3 antagonist activity" refers to activity in opposition to AGP-3 activity, as would result, for example, by inhibition of binding of AGP-3 to TACI.orBCMA. Such activity can be determined, for example by such assays as described in "Biological activityof AGP-3" in the Materials Methods of PCT/US00/93653, which is hereby incorporated by reference. Additional assays by which AGP-3 activity may be identified appear in the references WO 98/18921 (May 7, 1998); WO 98/27114 (June 25,1998); EP 869 180 (October 7,1998); WO 98/55620 and WO 98/55621 (December 10,1998); WO 9911791 (March11, 1999); W099/12964 (March 18,1999); and Gross Set al. (2000), Nature 404: 995-9. Any of the assays described therein may be Smodified as needed by methods known to persons having ordinary skill in the art "APRIL activity" refers to modulation of cell growth, survival, or activation resulting from binding of natural human APRIL to TACI or BCMA, particularly in T cells. Conversely, "APRIL antagonist activity" refers to activity i opposition to APRIL activity, as would result, for example, by inhibition of binding of APRIL to TACI or BCMA. Such activity can be determined, for example, by such assays as described in the Materials Methods hereinafter. Additional assays by which APRIL activity may be identified appear in the references WO 99/00518 (June 26, 1997); WO 99/11791 (Sept. 5,1997); WO 99/12965 (Sept. 12, 1997);EP 911 633 (Oct. 8, 1997); EP 919 620 (Nov. 26,1997); WO 99/28462 (Dec. 3,1997); WO 99/33980 (Dec. 30, 1997); WO 99/35170 (Jan. 5,1998); and Hahne et al.

(1998), T. Exp. Med. 188 1185-90. Any of the assays described therein and herein may be modified as needed by methods known to persons having ordinary skill in the art.

16 -BCMAk activity" refers to modulation* of 'cell growth, survival, or activation resulting from binding by'natural hu -nAPRIL or natural hun AGP'-3 to BCMA' Conversely, $iBC1~{A'antagonisf activity" refers to activity in o*''po sition't BCMA activity, as would resilt, for example, by ihibition of bindinig of AGP-3 or: ARl to BCMA. Sucl activity can be deterind fo xmle, by such asasa eie in the. Materials Metlkbd heieinafti r. Additionala""says by whk~h B'CMA activity may be identifiect ppear in the 6WOain reference WO 99/005187 (June 26,1997); WO 99/11791 (S§ept. 5,1997); WO 99/12965 (Sept'. 1 2, 1997);: El? 911 633 (Oct. 8, .197);'EP 919(6 20'-(No V. 26-; 1997); WO 99/28462 (Dec.'3, 1997);WO *99/338 0,"O f0~7); WO 99/35170 (Jan. 5, 1998) an et al. (1998), L

B

4 Med 188 1180; WO 9818921(May 7,1998)' WO 98/27114(Juine ,1998); EP S69:180 (October 7,19 98); WO'98/55620, ana WO 98/55621 ~ierlO,, 1998); WO 49/11791 (Mrc11 1999); Wb99/12964 (arc 18,1999); and Gross et al. (2000), Nature 404: 995-9. Any o-f the assays desribedlhieeixand h~rein may bem m odif as needed by methods n~Wn t~ersons hiavig ordinaiy sldin th~e a.

TT acity" ref ersio modulationi of cell groWth, survival, or activation. resulting from -binfding b y naturA hina AGP-3 or natural h anAPRIL to TACT. Con i'&rey "TAC atgonist activity" refers to acity i oppositidn to TACI ativity, a. would ,r -esult, for example, by inhibition'do binin 6 AP- or APRIL to TACT. S'ch activity can be deterind fr example, ly such Assay as de s cie in the Materials& Methods of PCT/USOO/03653, WO 98/18921 (MaUy 7,1998), WO 98/27114 Jn 25, 1998), EP 869 180 (Octobe 7,1998) w6 98/ 5 5620 and WO 98/55621 (December 10, 1998), WO 99/11791 (March 11, 1999), *W099/12964 (March 18, 1999), WO 98/39361 (September 11, 1998), von 13low Brain (1997), Scee 278:138-140, and Gross et"al. (2000), Nature 17 404: 995-9. Any of the assays described therein may be modified as needed by methods known to persons having ordinary skill in the art.

The term "specific binding partner" refers to any molecule that preferentially binds to a protein of interest, regardless of the antagonistic or agonistic activity of the molecule toward the protein of interest.

Exemplary specific binding partners include antibodies, solubilized receptors, peptides, modified peptides as described hereinafter, and the like.

SThe term "vehicle" refers to a molecule that prevents degradation and/or increases half-life, reduces toxicity, redtcies imn ogenicity, or increases biological activity of a therapeutic pobtein. Exemplary vehicles include an Fc domain (which is preferred) as well as a linear polymer polyethylene glycol (PEG), polylysine, dextran, etc.)i branchedchain polymer, (see, for example; U.S. Patent No.: ,289,872 to Denkenwalter et al., issued September 15,1981; 5,229,490 to Tam; issued July 20,1993; WO 93/21259 by Frech et al., published 28 October 1993);'a lipid; a cholesterol group (such as a steroid); a carbohydrate or oligosaccharide; or any natural or synthetic protein, polypeptidd or peptide that binds to a Ssalvage receptor.Vehicles are further described hereinafter.

The term "'native Fc" refers to molecule or sequence comprising the sequence of a non-antigen-binding fragment resulting from digestion of whole antibody,whether in-monomeric or mlltimeric form. The original S immuhoglobulin source of the native Fc is preferably of human origin and may be any of the immunoglobulins, although IgG1 and IgG2 are preferred. Native Fc's are made up of monomeric polypeptides that may be linked into dimeric or multimeric forms by covalent disulfide bonds) and non-covalent association. The number of intermolecular disulfide bonds between monomeric subunits of native Fc molecules ranges from 1 to 4 depending on class IgG, IgA, IgE) or subclass 18 SIgG1, IgG2;IgG3, IgAl, IgGA2). One example of a native Fc is a disulfidebonded dimer resulting from papain digestion of an IgG (see Ellison etal.

S(1982), Nucleic Acds Res. 10: 4071-9). The term "native Fc" as used herein is generic to the monomeric, dimeric, and multimeric forms.

The term "Fc variant" refers to a molecule or sequence that is modified from a native Fc but still comprises a binding site for the salvage receptor, FcRn. International applicatioris WO 97/34631 (published September 1997) and WO 96/32478 describe exemplary Fc variants, as well as interaction with the salvage receptor; and are hereby incorporated by reference,.Thus, the term "Fc variant" comprises a molecule or Ssequence that ishumanized from a non-humai native Fc. Furthermore, a S native Fe comprises sites that may be removed because they provide S structural features or biological activity that arendt.required for'the fusion molecules of the present invention. Thus, the tentm"Fc variant" comprises .a molecule or sequence that lacks one orore native Fc sites or residues that affect or are involved in disulfide bond:fornatio, (2) incompatibility with a selected host cell N-terminat heterogeneity upon .expression in a selected host cell, glycosylation, interaction with complement, binding to an Fc receptor other than a salvage receptor, or S. 20 antibody-dependent cellular cytotoxicity (ADCC). Fc variarits are described in further detail in WO 00/24782, published:May 4,2000, which is hereby incorporated by reference in its entirety.

The term "Fc domain" encompasses nitive Fc and.Fc variant molecules and sequences as defined above. As with Fc variants and native Fc's, the term "Fc domain" includes molecules in monomeric or multimeric form, whether digested from whole antibody or produced by other means.

The term "multimer" as applied to Fc domains or molecules comprising Pc domains refers to molecules having two or more 19 polypeptide chains associated covalently, noncovalently, or by both covalent and non-covalent interactions. IgG molecules typically form dimers; IgM, pentamers; IgD, dimers; and IgA, monomers, dimers, trimers, or tetramers. Multimers may be formed by exploiting the sequence and resulting activity of the native Ig source of the Fc or by derivatizing (as defined below) such a native Fc.

The term "dimer"'as applied to Fc domains.or molecules comprising Fc domains refers to molecules having two polypeptide chains S.associated covalently or non-covalently.

The terns "'der vatizing" and "derivative" or "derivatized" comprise processes and resulting compounds respectively in which the compound has a cyclic portion; for example, cross-linking between cysteinyl residues within the compound; the compound is cross-linked or has a cross-linkmg site; for example, the compound has a cysteinyl residue and thus,forms cross-linked dimers in culture or in one or more peptidyl linkage is replaced by a non-peptidyl linkage; the Nterminus is replacedby -NRR, NRC(O)R,. -NRC(O)OR, -NRS(O)R, S NHC(O)NHR,,a succnimide group, or substituted or unsubstituted Sbenzyloxycarbonyl-NH- wherein R and R and the ring substituents are as defined hereinafter; the C-terminus is replaced by -C(O)R 2 or -NR'R' wherein R 2 R and R' are as defined hereinafter; and compounds in which individual amino acid moieties are modified through treatment with agents.capable of reacting with selected side chains or terminal residues. Derivatives are further described hereinafter..

The term "peptide" refers to molecules of 2 to 40 amino acds, with molecules of 3 to, 20 amino acids preferred and those of 6 to 15 amino acds most preferred. Exemplary peptides may be randomly generated by any of the methods cited above, carried in a peptide library a phage display library), or derived by digestion of proteins.

20 The term "randomized" as used to refer to peptide sequences refers to fully random sequences (6.g.,'selected by phage diisplay methods) and sequences in which orinedt more fesidues of a naturally occurring molecule is replaced by anaminto acid residue not appearing in that position in the naturally'occurrihg iolicle. Exemplary methods for identifying.peptide sequenes include phage ilay; .oli display, ribosome display, ye st-based screening, RNA-peptide screening, chemicalcreening, ritional design; pr6tein stiuctiral analysis, and the like. Randomized peptides aid inethods of geierating them appear in to; WO 00/24782, ptblished May 2000,' whih is hiereby incorporated by refereice in its entirety, h- e term "phans acologically active" niithat a substance so described isiterined to.have actifity thait affects amedical parameter T cell proliferation) or diseaie state cancer aitoimnmune disorders): Thus, pharmacolo. ally active comipoiids comprise agonistic or mietLtic and antagoititic coinpoundi as defied beldd.

The terms "-miltiic" and "agdiist" refer to '&iimlecule having biologicalactivity comparable toWapiotin APRILA GP-3) that interacts with a protein of interest. These terms further include molecules that indirectly mimicthe activity of a protein of interest; Such as by potentiatirigthe effects ofthe natuialigicd of the protn of interest The terms "antagonist" or "inhibitor" refer to a molecule that blocks or in some way interferes with thie biological activity of thlie associated protein of interest, or has biological activity comparable to a known 2 25 a ntagonist or' inhibitor of the associatd protein 6f interest.

Additionally, physiologically acceptable salts of the compounds of S this iiventioi are also encompassed herein. By"physiologically acceptable salts" is meant any salts that are knoivn or later discovered to be pharmaceutically acceptable. Sdme specific examples are: acetate; 21 trifluoroacetafe; hydrohalides, such as hydrochloride and hydrobromide; sulfate; citrate; tartrate; glycolate; and oxalate.

Methods of Treatment The present invention concerns a method of inhibiting T cell proliferation in a mammal, which comprises administering a therapeutic agent comprising: a. a specific binding partner for TACI, wherein the specific binding partner has TACI antagonist activity; b. a specific binding partner for BCMA, wherein the specific binding partner has BCMA antagonist activity," c. both'a and b;or d a specific biding partner for TACI andBCMA, wherein the specific binding partner has TACT antagonist activity, BCMA antagonist activity or both.

The present invention also concerns a method of inhibiting APRIL activity in a mammal, whichTcomprises administering a therapeutic agent S, comprising a through d above.

S The invention also concerns a method of inhibiting TACI activity, BCMAactivity, or both in a mammal, which comprises administering a specific binding partner for APRIL. This method may further comprise administering a specific binding partner for AGP-3.

Some indications benefit from an increase in the immune response.

Accordingly, the invention further relates to a method of increasing T cell S proliferation in a mamnial which comprises administeriig a therapeutic agent comprising: Sa. a specific binding partner for TACI, wherein the specific binding partner has TACI agonist activity; b. a specific binding partner for BCMA, wherein the specific binding partner has BCMA agonist activity; 22 c. both-a andb; or d. a specific binding partner for TACI and BCMA, wherein the specific binding partner has TACI agonist activity, BCMA agonist activity or both.

The invention also concerns a method of increasing APRIL activity in a mammal, which comprises administering a therapeutic agent Scomprising a through d above.

The inventors contemplate carrying out the foregoing methods of treatment with any of several different types of molecules, including small molecules, antibodies; and engineered peptides and fusion molecules described hereinafter. These molecules may alsobe used in assays to identify cells and tissues that express AGE3, TACI, APRIL, or BCMA. The .invention further concerns nucleic acids; vectors, and host cells useful in preparing such molecules.

The invention further concerns methods of identifying compounds .that are useful in the aforementioned methods of use. Such compounds include nucleic acds, peptides, proteins, carbohydrates, lipids or small molecular weight organic molecules and may act'either as agonists or antagonists of BCMA, TACI,AGP-3 or APRI-proteinractivity.

AGP-3, APRIL, BCMA, andTACI are believed to play a role in regulation of immune function. Accordingly, these molecules, their S soluble forms; and agonists and antagonists there6f may be useful for the S diagnosis and/or treatment of inflammation and immune function S diseases. Indications for antagonists include, but are not limited to the following: S infections such as bacterial, fungal, protozoan and viral infections, especially HIV-1 or HIV-2; diarrhorea; psoriasis; 23 iniflammnation; *allergies; *atopic dermatitis; *respiratory allergic diseases such as'asthma, allergic rhinitis, hypersensitivity lung disease, hypersensitivity pneumonitis, easinophilic pneumionia g. Loeffler's synidririe, chronic eosinophilic pneumonia,,interstitial lung disease (ILD), such as idiopathic pulmonary fibrosis. orTILD- associated with rheumatoid arthritis; systemic lupus erythemnatosus, ankylosing spondylitis, systemic sclerosis, Sjogren's, syndrome; polymyositis or dermatoinysitis); systemic anaphylaxis':or hypersensitivity;responses; *drug allergy;, i nsect sting allergy; inflammuatory bowel disease, such As Crohns disease and ulcerative colitis;.

spondyloarthropiathy;; sdlerodernia; poiasis nlxntrdrntsissc as demtis eczema, at6pic deraitsaleic col~tact &d h~tis, urticia vasculitis (e.g nerotizing, cutaneous~ and hypersefisitivity vasulitis), eosinphilic myosftis and'eoinophilic fasciitis; *aultohin'mu ne Idiseases9 suc as rli~imatoid irthriitis, psorai arthritis, multiple 8aierosis,, systemic lupus e~~e~tsus, m yasthenia gravis, juvenile onset dibetes, glomerulonephritis, autoimmuune thyroitis and Be~t'~disese *graft rejection, inclduding allograft rejection or graft-versus-host dlisease; 24 cancers with leukocyte infiltration of the skin or organs; reperfusion injury; atherosclerosis; certain haematologic malignancies; shock, including septic shock and endotoxic shock.

.Ag nistscan be used fortreating: immunosuppression e.g. in AIDS patients or individuals S undergoing radiation therapy; chemotherapy, therapy for autoimmune disease or other diug therapy; and immurnosuppression due congenital deficiency in rece ttr function or other causes; and Sinfectious diseases such as parasitic diseases, including helminth infections, such as nematodes (round worms).

Compositions of matter Any numbe r of molecules may serve as specific binding artners within the present invention. Of particular interest are antibodies, peptides, and Pc-peptide fusion molecules.

Antibodies.

The invention also provides for an antibody or antigen binding doa thereof, or a fragmet, variant, oreriative thereof, which binds to eie onan of the target molecules P AGP-3, TACI, or BCMA) and has partia or cmpet agonist or antagonist activity.

Preferably, the target molecul is naummalian, more preferably human, and ma be in soluble or cell surface assqciated forms, or fragments, derivatives and variants thereof: A number of methods for antibody generation are known in the art.

All such methods are useful in generating molecules useful in accordance with the present invention. Conventionally, an antibody may be prepared by immunizing an animal with the target molecule murine or human 25 BCMA or TACI) or with an immunogenic fragment, derivative or variant thereof. In addition, an animal may be immunized with cells transfected with a vector containing a nucleic acid molecule encoding the target molecule such that the target molecule is expressed and associated with the surface of the transfected cells, Alternatively; specific binding partners that are antibodies may be obtained by screening a library comprising antibody or antigen binding domain sequences for binding to the target molecule. -Such a library is conveniently prepared in bacteriophage as protein or peptide fusions to a bacteriophage coatprotein which are excpressed on the surface of assembled phage particles and the encoding ,DNA setqiences contained within.the phage particles (so-called "phage display library"). In one example, a phage display library contains DNA sequences encoding human antibpdies, such as variable light and heavy chains,Sequences bing to the target molecule may be further evolved by multiple rounds of mutagenesis and screenifng Specific binding partners that are antibodies or antigen binding domains may be tetrameric glycoproteins similar to native antibodies, or S they maybe sinigle chain antibodies; for example, Fv, Fab, Fab' or F(ab)' fragments,,bispecific antibodies, heteroantibodies; or other fragments, 2 v ariants, or derivatives thereof,,which are capable of binding.the target molecule and partially or completely neutralize the target molecule activity. Antibodies or antigen, binding domains may be produced in hybridoma cell lines (antibody-producing cells such as spleen cells fused to mouse myeloma cells, for example) or may be produced in heterologous cell lines transfected with nucleic acid molecules encoding said antibody pr antigen binding domain..

Antibodies of the invention include polyclonal monospecific polyclonal, monoclonal, recombinant, chimeric, humanized, fully human, single chain and/or bispecific antibodies. -Antibody fragments include 26 those-portions of an antibody that bind to an epitope ofia target molecule.

Examples of such fragments include Fab Fv, and sFv frgments. The antibodies may be generated by enzyinatic cleavage of full-length antibodies dr by recombinant DNA techniques, such as expression of recombinant plasmids containg nuceic acid sequences encoding antibody variable regions.

Polyclonal antibodies are heterogeneous populations of antibody molecules derived from the sera of adimals immiunhized'vith an antigen.

An antigen is a moletulk or a portion of ad moleculd capable of being S 10 bound by at antibody which is additionally capable of liducing an animal S to produce antib6dy capable ofbindifig to' aft epitope of that antigen. An S-antigen caniav eone'or iorelepitope. Th&2'secificdretion referred to above iS meant to ididte that the anitigenii ill react, i a highly'selective mainner, with its corresponlding anitibody aid'not with the multitude of other antibodies which cai be eoked by otheranttigens.

FPolycidal antibobdies directed toward a target molecule generally are raised in ainimals Ce, irabbitsor mice) b multipl8e subcutaneous or iintraperitoneal injections of thd taiget mol&ul'e and aiadjuvant. In a cordance with the invention, it riay be useful to conjuigate the target molecule; or a variant, fragimeit, or deiivative theiebf to a cahier protein that is immurntogenic ii the species to b4 iimunized, such as keyhole .limi.pet heocyanin serumn, albuinin, bovine thyroglobulin, or soybean typsin inhibitor. Also, aggregating agets such asalim are used to enhance the immunie respoise. After imiinimizatiomr the animals are bled and th seruimx is assayed forinti-target aftibody titer:' Monoclonal antibodies (mAbs) contiin a subs'tantially homogeneous populatibr of aintibodies specific to antigens, which population contains substantially similar epitope bindiig sites. Such antibodies may be of ahy immunoglobulin class includiig IgG, IgM, IgE, 27 IgA, IgD and any subclass thereof. A hybridoma producing a monoclonal antibody of the present invention may be cultivated in vitro, mi situ, or in vivo. Production of high titers in vivo or in situ is a preferred method of production.

Monodonal antibodies directed toward the target'molecule are produced using any.method which provides for the production of antibody moleculesby continuous cell lines in culture. Examples of suitable methods for. preparing monoclonal antibodies include hybridoma methods of Kohler eta, Nature 256,495-497 (1975), and the human B-cell S 10 hybridoma method, Kozbor, T.ImmunoL 133, 3001 (1984); Brodeur etal., Monoclonal Antibody Production Techniques'and Applications, pp. 51-63 (Marce Dekker, Inc., New York, 1987); and Harlow and Lane, Antibodies: ALaboratory Manual, Cold Spring HarborLaboratory (1988); the contents of which references are incorporated entirely herein:by reference.

Preferred specific binding partners iiclude monoclonal antibodies which will inhibit partially or completely the binding of the human target molecule, to its cognate ligand or receptor or an antibody having substantially the same specific binding characteristics, as well as fragments and regions thereof. Preferred methods for determining monodonal antibody specificity and affinity by competitive inhibition can be found in Harlow et al., Antibodies: A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, 1988), Colligan et al., Seds., Current Protocols in Immunology, Greene Publishing Assoc. and Wiley Interscience, (1992, 1993), and Muller, Meth Enzvmol., 92:589- 601 (1983). Each of these references is incorporated herein by reference in its entirety.

Also provided by the invention are hybridoma cell lines which produce monoclonal antibodies reactive with target polypeptides.

28 Chimeic, antibodies are molecules in which different portions are derived from different animal species, such as those having a variable region derjied from- a murine mbnoclonal antibody and a human inimunoglobulin constant region. Chimeric antibo'dies are primarily used to reduce immunogenicity in application anid to" in crease yields in ,.production,. for example,. where mnurinemoniioclonial antibodies have higher yields from hybridomas but higher in-imunogeiity in humans, such thathunian/mrurine chimeric monodo nal antibodies are used.

Chim~iic antibodies, and methods for their prc~duction are known la.: in the art.,; CabAily etal, Pr6c. Natl. Acad Set USA, 81:3273-3277 (1984); Morrison ~et FLIroc: NatI. Acad~ Set, USA, 81:6851-685'(1984); Boulianne eta. Natur 31:4-646 (1984); Neuberger'etgj Natie 314268-270 (185); Liugt j. Proc. NAtl ,Acad. Set. USA, 84:3439,-3443 (l987);'and -HarloW and. Lane Antibodies-A Laboratory Manual Cold Spring Harbor Laboratory (1988);. These references are incorporated herein by reference in their entirety.

Achimeric monoclonal antibody of the finvention may be used as a therapeutic, agent In such a chimeric jAitibod a portibon of the heavy an/rlight. chain is identi&al with. or. homt gdu trcorresponding 20f sequence in antibodies derived from a particular speciesbr bdonging to one particular antibody. class or subclass,, while the remainder of the chaift(s)' is identical with 6r homologous to correspondig sequence in antibodie-s derived from another sp&des or beloriging to another antibody 'cdass 6r subclass,: as well A& fragmenits of such antibodies, so long as they exhibit the desired biological activity (see U.S. PatenitNo. 4,816,567; Morrison et al., Proc. NatI. Acad. Sci. 81, 6851-6855 (1985).

As used herein, the term "chtimeric antibody" includes monovalent, divalent or polyvalent immunoglobuins?! A monovalent dhimeric antibody is a climer (BL) formed bk a cimeric H chain associated through 29 disulfide bridges with a chimeric L chain. A divalent chimeric antibody is tetramer (IL) formed by two HL dimers assbciated through at least one disulfide bridge.,! A polyvalent chimeric antibody can also be produced, for example, by employing a C- region that aggregates from an IgM H chaiii, or R chain).

Murine and chimeric antibodies, firagmentS and regions of the present inventiorinmay comprise individual hdavy and/or light (L) immunoglobulin' chains. A chimeric H chain comnprises an antigen binding region drived from the H chain ofa rin-human antibody specific for the target molecule, which is linked to at least a portion of a human H chairt& regiofi siuch is C1'or C.

Achinefit chain accordiig to the pihit inviention comprises an aitigen'binding regib derivd from the'Lchain of a 'ri-humari antibody .specific for the target iolecule, lik d to at lIast a &ortion of a human L chain Cr egiott (C Spdeific bindifig partners, such as antibodies, fragments, or derivativeshaving chimeric H chains and L ch~iri of the same or different vrariable"region binding speifficityj can also be prepared by appropriate association of the individual pil yeptide chains, according to knowfmethod steps,', according to Ausubl et al., eds. Current Protocols in Molecular Biolbgy WilefrintersieeIrye, NY. (1993), and HLarlow et al-, Antibodies: A Laboratory Manial, Cold Spring Harbor Laboratory Press, Cold Spriiig Harbor, N.Y. (1988). The contents of these S refrences are incorporated entirely herein byreference. With this approach, hosts expressing chimeric H chains (or their derivatives) are separately cultured from hosts expressing chimeric L chains (or their derivatives), and the immunoglobulin chains are separately recovered and then associated: Alternatively, the hosts can be co-cultured and the chains 30 allowed to asociate spontaneously in the culture medium, followed by recovery of the assembled immunoglobulin, fragment or derivative.

As an example, the antigen binding region of the specific binding partner (such as a chimeric antibody) of the present invention is preferably derived from a non-human antibody specific for the human analog of the target molecule. Preferred sources for the DNA encoding such a nonhuman antibody include cell lines which produce antibodies, such as hybrid cell lines commonly known as hybridomas The invention also provides for fragments,,variants and derivatives, and fusions of anti-target antibodies, wherein the terms "fragments", "variants", "derivatives" and 'fusions" are defined herein. The invention .encompasses fragments; variants, derivatives, and fusions of anti-target antibodies which are functionally similar to the unmodified antibody, that is, they retain at least one of the activities of the unmodified antibody. In addition to the modifications set forth above, also included is the addition of geneic sequences coding for cytotoxic proteins such as plant and bacterial toxins. The fragments, variants, derivatives and fusions of the antibodies can be produced from any of the hosts of this invention.

Suitable fragments include, for, example; FabFab', F(ab), Fv and scFv. These fragments lack the Fc fragment of an intact antibody, clear Smore rapidly from the circulation, and can have less non-specific tissue binding than an intact antibody. See Wahl et al., I. Nucl Med., 24:316-325 .1983). These fragments are,produced from intact antibodies using methods well known in the art, for example by proteolytic cleavage with enzymes such as papain (to produce Fab fragments) or pepsin (to produce fragments). The identification of these antigen binding regions and/or epitopes recognized by monoclonal antibodies of the present invention provides the information necessary to generate additional monoclonal antibodies with similar binding characteristics and 31 therapeutic or diagnostic utility, that parallel the embodiments of this invention.

Variants of specific binding, partners are also provided. In one embodiment, yariants of antibodies and antigen binding domains comprise changes in light and/or, heavy chain amino acid sequences that are naturally occurring or are.introduced by in vitro engineering of native sequences using recombinant DNA,techniques. Naturally occurring variants include "somatic" variants which are generated in vivo in the corresponding germ line nuceotide sequences during.the generation of an antibody response to a foreign antigen..

Variantpof antibodies and antigen binding domains are also S: prepared by mutagenesis techniques knownin the art In one example, amino acid changes may be introduced at random throughout ani antibody S coding region and the resultig, variants may be screened for a desired activity, such as binding affinity for the target molecule. Alternatively, amino acid changes may be introduced in selected regions of an antibody, such as in the light and/or heavy chain CDRs, and framework regions, and the resulting antibodies may be,screened forbinding to the target molecule,or some other activity. Amino acid changes encompass one or more amino acid substitutions in a CDR, ranging from a single amino acid Sdifference to the introduction of all possible permutations of amino acids within a given CDR, such as CDR3-. Inanother method, the contribution of each residue within a CDR to target binding may be assessed by substituting at least one residue within,the CDR with alanine (Lewis et al.

(1995), Mol. Immunol. 32: 1065-72). Residues which are not optimal for Sbinding to the target molecule may then be changed in order to determine a more optimum sequence. Also encompassed are variants generated by insertion of amino acids to increase the size of a CDR, such as CDR3. For example, most light chain CDR3 sequences are nine amino acids in length.

32 Light chain'CDR3 sequences in an antibody ihich are shorter than nine residues may be optimized for binding to the target molecule by insertion of appropriate amtino acids to increase the length of the CDR.

In one embodiment, antibody or anitigen binding domain variants comprise one or mre amiino acd changes in dnie' or more of the heavy or light chain CDRI1, CDR2 or CDR3 and optionally one or more of the heavy or light chain framework egions FR, FR2 or FR3: Amino acid changes comprise substitutions, deletions and/or insetions of anino acid residues.

Variants may also be prepared by "chain shuffling" of either light or heavy cha Marks et al.'(1992), Biotechnolog 10: 779-83. Typically, a single light (or heavy) chain is;'onibiied'with. alibrary hiving a repertoire of heAry (i light) chains and the resulting population is §creened for a Sdesired activity; sith a bindinig to the tairgetf mibcutile. This technique permitsscreening ofa greater sampl& of different heavy (or light) chains in combinatii withi a single light (or heavy) diain than is possible with libraries comprising repertoires of both heavy and ligltf chains.

The specifi binding partriers of the invention canbe bispecific.

Bisecifi pecific pecfl binding partners of thi invetii6n can be of several 'configurations. For example, bispecific antibdies resemble single antibodies (or antibidy'fragments) but have two different antigen binding sites (viaiable regions) Bispecific ahtibodies can be prodixced by chemical techitiqis (See Kranz etal., Proc Natl. Aid. Sci. USA, 78:5807 (1981)), by "polydona" techniiques (see U.S. Pat. N6. 4,474,893 to Reading) orby rec6mbinant DNA techniques. For eample, a bispecific antibody in accordaice with this invention may bind to APRIL arid AGP-3. As another example, a bispecific antibody may bind to TACI and BCMA.

The specific binding partne's of the inventidii may also be heteroantibodies. Heteroantibodies are two or moe antibodies, or 33 antibody binding fragments (Fab) linked together, each antibody or fragment having a different specificity.

The invention also relates to "humanized" antibodies. Methods for humanizing non-human antibodies are well known in the art. Generally, a humanized antibody has one or more amino acid residues introduced into a human antibody from a source which is non-human. In general, non-human residues will be present in CDRs. Humanization can be performed following methods known in the art (Jones etal, Nature 321 522-525 (1986); Riechmann et at:Nature 332, 323-327 (1988); Verhoeyen et a, Science 239, 1534-1536 (1988)), by substituting rodent c. cmplementarily-determiningregions (CDRs) for the corresponding regions of a human antibody..

Th specific binding partners of the invention, including chimeric, CDR-grafted, and humanized antibodies can be produced by recombinant methods known in the art- Nucleic acds encoding the antibodies are introduced into host cells and expressed using materials and procedures described herein and known in the art. In a preferred embodiment, the antibodies are produced in mammalian host cells, such as CHO cells.

SFully human antibodies may be produced by expression of recombinant DNA transfected into host cells or by expression in hybridoma cells as described above.

Techniques for creating recombinant DNA versions of the antigen- S binding regions of antibody molecules which bypass the generation of monoclonal antibodies are encompassed within the practice of this invention., To do so, antibody-specific messenger RNA molecules are extracted from immune system cells taken from an immunized animal, and transcribed into complementary DNA (cDNA). The cDNA is then cloned into a bacterial expression system. One example of such a technique suitable for the practice of this invention uses a bacteriophage 34 lambda vectorsystem having a leader sequence that causes the expressed Fab protein to migrate to the periplasinic space (between the bacterial cell membrane and the cell wall) or to be secreted. Oie can rapidly generate and screen great numbers of functional Fab fragments for'those which bind the antigen.: Such target molecule specific binding partners (Fab fragments with specificity for the target molecule) are specifically encompassed within the term "antibody"' as it is defined, discussed, and claimed herein.

Also within the scope of the invention are techniques developed for the production of chimeric antibodies by splicing the genes from a mouse 4 antibody molecule of appropriate antigen-speificity together with genes from a human antibody molecule of apprdpriate biologial activity, such Sas the ability to activate hui f an complement and mediate ADCC.

.(Morrison gt a Proc. Natl AcadeSci'k 81:6851 (1984); Neuberger etal.

Nature 312:604 (1984)). One example is the replacement of a Fc region Swith that of a different isotype. Specific binding partneis such as antibodies produced by this technique are within the scope of the invention. In a preferred embodiment of the invention, the antibodies are fully :human antibodies. Thus encompassed by the invention are antibodies that bind target molecules and are encoded by nucleic acid sequences Swhich are naturally occurring somatic variants of human germline immunoglobulin nucleic acid sequence, and fragments, synthetic variants, derivatives and fusions thereof. Such antibodies may be produced by any method known in the art. Exemplary methods include immunization with S a target antigen (any target pblypeptide capable of elicing an immune response, and optionally conjugated to a carrier) of transgenic animals Se mice) that are capable of producing a repertoire of human antibodies in the absence of endogenous imminoglobulin production. See, for r 35 example, Jakobovits et Proc. Natl. Acad. Sd. 902551-2555 (1993); Jakobovits et al. Nature 362, 255-258 (1993); Bruggermann et al, Year in Immunol., 7, 33 (1993).

Alternatively, human antibodies may be generated through the in vitro screening, of phage display antibody libraries. See Hoogenboom et Mol. Biol, 227, 381 (1991); Marks et T. Mol. Biol. 222, 581 (1991), incorporated herein by reference. Various antibody-containing phage display libraries have been described and may be readily prepared by one Sskilled in the art. Libraries may contain a diversity of human antibody sequences, such as human Fab, Fv, and scFv fragments, that may be screened against an appropriate target. As described further below, phage display libraries may comprise peptides or proteins other than antibodies which may be screened to identify specific binding partners of the target molecule.

An anti-idiotypic (anti-Id) antibody is an antibody which recognizes unique determinants generally associated with the antigenbinding, site of an antibody. An.Id antibody can be prepared by .immunizing an animal of the same species and genetic type mouse strain) as the source of the monoclonal antibody with the monodonal S0 antibody towhich an anti-Id is being prepared. The immunized animal will recognize and respond to. the idiotypic determinants of the immunizing antibody by producing an antibody to these idiotypic determinants (the anti-Id antibody). See, for example, U.S. Pat. No.

4,699,880, yhich is herein entirely incorporated by reference. The anti-Id antibody may also be used as an "immunogen" to induce an immune response in yet another animal, producing a so-called anti-anti-Id antibody. The anti-anti-Id may be epitopically identical to the original monoclonal antibody which induced the anti-Id. Thus, by using 36 antibodies to thie'idiotypic determinants of a mAb, it is possible to identify other dones expreing anibodies of identica speicty Peptides and Peptide fusion molecules.- The patenit application WO 00/24782, published May 4,2000, mentioned previously herein -describes in, etivaous Petd generation techniquek. That patent applicaticmt further dekcribes various derivatives and fusion molecuks.

~~~~inparticular, a peptide used as a specifi biingparnrmyb comfprised *ithih a mole6cule of the formhula:, 1 2~ wierem F; ig a vehicL;' XV and k 2 aike-each mhidiependntl saeed, from t-(L')P 1

-(L

1 )vP 1 (Ud 2

{L

1 ),jPL)d4V 2

{L

3

)P

3 I and -WV)~P )-P 2 3 -l' 3

-(L

4 )f-P[ 4

L

1

P

1 aiid F' arei each independehtly peptide sequences, wherein at least one is a specifi bindig, partn'er; L, and U'are eachindependently lners; n a, c,d, e, 'and f ir~ each lndependenilY 0~ or1,poiethta e6t onefaand bisl Preferably; suchi a molecl opie; titi of the formulae or A'more preferred molecule comprses astuur'othfrml or a struicture of the forffiula wherein 1" and/or 2 is a specific bihding partner for TACI or BCMvA.

Such molecules facilitate modulation of both TACI and BCMA; for 37 example, one of P' and P2 is a specific binding partner for TACI and the other is a specific binding partner for BCMA. Conversely, in a ligand inhibitor, one of P' and P 2 is a specific binding partner, for APRIL and the S other is a specific binding partner for AGP-3.

For all of these molecules, the preferred vehicle is an Fc domain.

SAmong Pc domains, IgG Fc, particularly IgG1, are preferred.

S The Fc domains, linkers, and processes of preparation of the foregoing molecules is described in WO 00/24782, published May 4,2000.

S. Soluble receptor fragments Another class of specific binding partners are soluble receptor fragments. Of p icular nterest are the fragments identified in the figures:.. S a. the extracellular region of TACI (SEQ ID NO: the extracellular region of BCMA (SEQ ID NO: 6).

15 c. the consensus region of TACI (SEQ ID NO: 16).

d. the consensus region of BCMA (SEQ ID NO: e. the TACI/BCMA extracellular consensus sequence (SEQ ID NO: 13>. These molecules have the heretofore unrecognized advantage of binding both APRIL and AGE-3. Like the aforementioned peptides, these specific binding partners may also be covalently linked to a vehicle, preferably an Fc domain. Muteins Additional useful peptide sequences may result from conservative and/or non-conservative modifications of the amino acid sequences of the aforementioned antibodies, peptides, Fc-fusion peptides, and receptor fragments.

Conservative modifications will produce molecules having functional and chemical characteristics similar to those of the molecule 38 from which'such modifications are made. Iri contrast, siubstantial modifications in the functional and/or chemical characteristics of the molecules may be accomplished by selecting substitutions in the amino acid sequence that differ significantly iin their effect ofit aintaining the striudcture of the molecular backbone in the area of the substitution, for example; as a sheet or helical coiifo-mation the charge or hydrophobicity of the molecule at the target sitetor the size of the molecule. For example, a "conservativeaminLo acid substitution" may involve 'a substitution-of a native Amino acid fesidue w ith a nonnative residue such that there is little or no' effect on tha polaiit r charge of the amino acid residue at that position. Furthermore, any native residue in the polypeptide maylso be substituted with alanin as has been previously desribed for ,"alanifte scaanning ziutagitnesiW' (see for example, MacLennazi e 1998, Acta Physibl. Saind. Si l. 643:55-67; Sasalki et al., 1998, Adv. Biophys. 35:1-24, which discuss alanit scanning mutagenesis).

Desired amimno acid substiitiOs (hliethier conservative or nonconservative) can be determined by those skilled in the art at the time such suibstitutionsare desired. For example, aimio idad substitutions canbe used to identify important resi'dies of the miolecile sequence, or to increase or decrease the affinity of the miolecules described herein.

Exemplary amino acid substitutions are set forth in Table 3.

39 Table 3-Amino Add Substitutions Original Exemplary referred Residues Sibsiutions Substitutions Al-a Val, Leu, lie Val.

Arg Lys Gin, Asn

LYS

Asn Gin Gin Asp Glu Glu Cys C)9er, Ala Ser Gin (Q *Asn As'n Giu (E),.Asp Asp ~ly G)'ProAlaAla His Asn, -Gin, Ly,4rg AFg Ilie Leu,:Vali, Met, Ala, Leu Phe, Norleucine Leu Norleucine, lHe, Val, Ilie: Lys Arg,'4 Diamino-Ar butyric Acid, Gin,-Asn, Me Le u, Phe, lie Leu Phe Leu-,.Val, lie, Ala, Tyr, Leu Pro Ala Gl Ser, Thr, Ala, Cys h Thr(T)~ Se Trp, Tyr, Phe' 'Tyr(Y Trp, Phe, Thr, Ser Phe Val' lid, Met Leu, Phe,'e Ala, Norleucine 40 In certain embodiments, conservative amino acid substitutions also encompass non-natuially occurring amino acid residues which are typically incorporated by chemical peptide synthesis rather than by synthesis in biological systems.

As noted in the foregoing section ')efinition of Terms," naturally occurring residues may be divided into classes based on common sidechain properties that miay be useful for modifications of sequence. For example, non-conservtive siibstitutions may ivolve the exchange of a iember of one of these cldasses for a member from another class. Such Aitbstituted residues may be introduced int irgibns of the molecule that areihomologous with i'io-1r ortholog or into the non-homologous regions of the mole ul. In addition, one ntr alb make modifications using P or G for the putipose of influeincing hain orientation.

In making stich modifications, the hydropathic index of amino adds may be conisidered. Eac amino icid has been assigned a hydropathic index op the basis of thei ihy ophobiity and charge characteristics, thse are; isoleicine valine leucine phenylalanine cysteine/c'ystine methionine alanine glycine threoniie serine tryptophan tyrosine proline histidine glutamate glutamine aspartate asparagine lysiTe and arginine The importance of the hydropathic amino acid index in conferring interactive biological function on a protein is understood in the art. Kyte et l. T. 1Mo1. Biol., 157: 105-131 (1982). It is lakown that certain amino acids iiay be substituted for other amino acids having a simnilar hydropathic index or score and still retain a similar biological activity. In making changes based upon the hydropathic index, the substitution of amino acids whose hydropathic indices are within ±2 is preferred, those which 41 are within :1 are particularly preferred, and those within ±0.5 are even more particularly preferred.

It is also understood in the art that the substitution of like amino acids can be made effectively ori the basis of hydrophilicity. The greatest local average hydrophilicity of a protein, as governed by the hydrophilicity of its adjacent amino acids, correlates with its immunogenicity and antigeiicity, i.e. with a biological property of the protein.

The followiig hydrophilicity value have ber assigned to amino acid residiesargiie (+3.0);-lysini aspartate glutamate seine aipargine (0.2glutinine glycine threohirne proline alanine histidine cysteine methionine Valine leucine isoleucie tyrosine phenylaanine trypitophan In making changes based upon imilar hydrophilicity Values, the substitution of ainino acids whose hydrbphiii cty values are withinh ±2 is preferred; those i hich are within ±1 are particularly preferred, and those within -±05 are even more particularly preferred. One may also identify epitopes from primary armino acid sequences on the basis of hydrophilicity. These regions are alsDreferred tb'o as "epitopic core regionsY' A skilled artisan will be able to determiine suitable variants of the polypeptide as set forth in the foregoing sequences using well known techniques. For identifying suitable areas of the molecule that may be changed withot destroying activity, one skilled in the art may target areas not believed to be important for activity. For example, when similar polp eptides with similar activities from the same species or from other species are known, one skilled in the art may compare the amino acid sequence of a molecule to similar molecules. With such a comparison, one can identify residues and portions of the molecules that are conserved 42 among similarpolypeptides. It will be appreciated that changes in areas of a molecule that are not conserved relative to such similar molecules would.beless likely to adversely affect the biological activity and/or S. structure of the molecule. One skilled in the art would also know that, even in relatively conserved regions; one may substitute chemically similar amino acids for the naturally occurring residues while retaining activity (conservative amino acid residue substitutions). Therefore, even areas that may be important for biological activity or for structure may be subjec to conservative amino add substitutions without destroying the 10 biological actiyity or without adversely affecting. the molecule structure.

Additionally, one skilled in the art can review structure-function studies identifying residues in similar molecules that are important for activity or structure. In view of such a comparispn, one can predict the importance of amino acid residues in a molecule that correspond to amino acid residues that are important for activity or structure in similar molecules. One skilled in the art may opt for chemically similar amino add substitutions for such predicted important amino acid residues of the molecules.

One skilled in the art can also analyze thethree-dimensional structure and amino acid sequence in relation to, that structure in similar polymolecules. In view of that information, one skilled in the art may S predict the alignment of amino acid residues ofa molecule with respect to S its three dimensional structure. One skilled in the art may choose not to make radical changes to amino acid.residues predicted- to be on the surface of the protein, since such residues may be involved in important S interactions with other molecules. Moreover, one skilled in the art may generate test variants containing a single amino acid substitution at each desired amino acid residue. The variants can then be screened using activity assays know to those skilled in the art. Such data could be used to 43 gather information about suitable variants. For example, if one discovered that a change to a particular amino acd residue resulted in destroyed, undesirably reduced, or unsuitable activity, variants with such a change would be avoided. In other words,;based on information gathered from such routine experiments, one skilled in the art can readily determine the amino acids where further substitutions should be avoided either alone or in combination with other mutations.

A number of scientific publications have been devoted to the prediction of secondary structure. See Moult Curr. Op. in Biotech. 7(4): S,0 422-427 (1996), Chou et al. Biochemistry, 13(2): 222-245 (1974); Chou et al.

Biochemistry, 113(2): 211-222 (1974); Cho eta..l Adv. Enzymol. Relat.

Areas-Mol. Biol. 47:45-148 (1978); Chou et al, Ann. Rev. Biochem., 47: .251-276 and Chou et al Biophys 26:367-384 (1979). Moreover, computer programs are currently available tb assist withpredicting ,scondrystructure. One method of predicting secondary structure is based upon homology modeling, For example, two polypeptides or proteins which have a sequence identity of greater than 30%, or similarity Sgreater than 40% often have similar structural topologies; The recent growth of the protein structural data base (PDB) has provided enhanced predictability of secondary structure, including the potential number of folds within a polypeptide's or protein's structure; See Holm et al., Nud.

Acid.Res..27(1): 244-247 (1999). It has been suggested (Brenner etal., Curr. Op. Struct. Biol. 369-376 (1997)) that there are a limited number of folds in a given polypeptide orprotein and that once a critical number of structures have been resolved, structural prediction will gain dramatically in accuracy.

Additional methods of predicting secondary structure include "threading" (Jones, Curr. Opin. Struct. BioL, 377-87 (1997); Sippl et., Structure, 15-9 (1996)), "profile analysis" (Bowie etal., Science, 44 253: 164-170 (1991); Gribskov et al, Meth. Enzym., 183: 146-159 (1990); Gribskov et al., Proc. Nat. Acad. Sci, 84(13): 4355-8 (1987)), and "evolutionary linkage" (See Home, supra, ard-Brenneri upra).

Production of spdfi binding partteis When the specific binding partner to be piepared is a proteinaceous specific-binding partner, such as ani antibody or an antigen binding domain or an Fc-peptide fusion molecule, various biological or chemical methods for producing said partner are available.

Biological minethods are preferable for prodiucing sufficient o quantities of a specific bindiig partner for therapeutic use. Sta dard ,recombixant DNA techiriques are particularlyusefi r ft he production of antibodies and antigen biding'doirais of thh invdntiont Exemplary expression idctois, host cell aid nmethods for recovryo df the expressed product are described below* 'A nucleic acidci6lecue enbcoding an antibody or, intigei binding domain is inserted iito an appropriate expredsionhvectdr iising standard ligation techniues. The vector is typically telected to be functional in the particular host cell employed the vector is compatible with the host cell machiiiery such that animplification of the gene and/o expression of the gene' can occur). A nucleic acid molecule'eicoding an antibody may be amplified/exprbssed in prokaryotic, yeast, iisect (baculovirus systems) and/or eukcaryotic host'cells. Selection of the hiost cell will depend in part on whether ani antibody is to be post-trainsitionally modified glycosylated and/or phosphorylated). If so, yeast, insect, or mammalian host cells are preferable. For a review of expression vectors, see Meth.

Enz. v. 185, Goeddel, Academit Press Inc., SanDiego, CA (1990) Typically, expressioni vectors used in any host cells will contain one or more of the following comporietits: a promoter, one or more enhancer 45 sequences, an-origin of replication, a transcriptional termination sequence, a complete intron sequence containing a donor and acceptor splice site, a leader sequence for secretion, a ribosome binding site, a polyadenylation sequence, a polylinker region for inserting the nucleic acid encoding the polypeptide to be expressed, and a selectable marker element. Each of these sequences is discussed in more detail below.

The vector components may be homologous from the same species and/or strain as the host cell), heterologous from a species other than the host cell species or strain), hybrid a combination of different sequences fronmore than onA source), synthetid, or native sequences which normally function to regulate immunoglobulin expression, As such, a source of vector components may be any prokaryotic or eukaryotic organism, any vertebrate or invertebrate organism, or any plant, provided that the components are functional in, and'can be activated by, the host cell machinery.

.An origin of replication is selected based upon the type of host cell being used for expression. For example, the origin of replication from the plasmid pBR322 (Product No. 303-3s, New Englard Biolabs, Beverly, MA) is suitable for most,Gram-negative bacteria while various origins from SV40, polyoma, adenovirus, vesicular stomatitus virus (VSV) or papillomaviruses (such asHPV or BPV) are useful for cloning vectors in mammalian cells. Generally, the origin of replication component is not needed for mammalian expression vectors (for example, the SV40 origin is often used only because it contains the early promoter).

25 A transcription termination sequence is typically located 3' of the end of a polypeptide coding regions and serves to terminate transcription.

Usually, a transcription termination sequence in prokaryotic cells is a G-C rich fragment followed by a poly T sequence. While the sequence is easily cloned from a library or even purchased commercially as part of a vector, 46 it can also be readily synthesized using methods for nucleic acd synthesis such as those described above.

SA selectable marker gene element encodes a protein necessary for the survival and growth of a host cell grown in a selective culture medium. Typical selection marker genes encode proteins that confer resistance to antibiotics or. other toxins, eg., ampicillin, tetracycline, or kanamycin for prokaryotic host cells, complement auxotrophic deficiencies of the cell; or supply critical nutrients not available from complex media. Preferred selectable markers'are the kaiamycin resistance gene, the ampicillin resistance gene, and the tetracycline Sresistance gene. A neomycin resistance gene may also be used for S selection in prokaryotic and eukaryotic host cells..

Other selection geries maybe used to amplify the gene which will Sbe expressed. Amplification is the process-wherein genes which are in greater demand for the production of a protein critical for growth are reiterated in tandem within the chromosomes of successive generations of recombinant cells. Examples of suitable selectable markers for maammalian cells include dihydrofolate reduiicta (DHFR) and thymidine kinase The mammalian cell transforinants are placed under selection pressure which only the transformants are uniquely adapted to survive by virtue of the marker present in the vector. Selection pressure is imposed by culturingthe transformed cells under c6nditions in which the concentration of selection partner in the medium is successively changed, thereby leading to amplification of both the selection gene and the DNA that encodes an antibody. As a result, increased quantities of an antibody are synthesized from the amplified DNA.

A ribosome binding site is usually necessary for translation initiation of mRNA and is characterized by a Shine-Dalgarno sequence (prokaryotes) or a Kozak sequence (eukaryotes). The element is typically 47 located 3' to the promoter and 5' to the coding sequence of the polypeptide to be expressed. The Shine-Dalgarno sequence is varied but is typically a polypurine having a high A-G content). Many Shine-Dalgarno sequences have been identified, each of which can be readily synthesized .,using methods set forth above and used in a prokaryotic vector.

A leader, or signal, sequence is used to direct secretion of a polypeptide., A signal sequence may be positioned within or directly at the 5' end of a polypeptide coding region. Many signal, sequences have been identified and may be selected based upon the host, cell used for expression. In the present invention, a signal sequence may be S homologous (naturally occurring) or heterologous to a nucleic acid sequence encoding an antibody or antigen binding domain. A S.heterologous signalsequence.selected should be one that is recognized and processed, cleaved, by a signal peptidase, by the host cell. For prokaryotichost cells that do not recognize and process a native immunoglobulin signal sequence, the signal sequence is substituted by a prokaryotisignal sequence selected, for example, from the group of the alkaline phsphatase, penicillinase, or heat-stable enterotoxin II leaders.

S. For yeast secretion, a native immunoglobulin signal sequence may be substituted by the yeast invertase, alpha factor oracid.phosphatase leaders.. In mammalian cell expression the native signal sequence is satisfactory, although other mammalian signal sequences may be suitable.

In most cases, secretion of an antibody.or antigen binding domain Sfrom a host cell will result in the removal of the signal peptide from the antibody. Thus the mature antibody will lack any leader or signal sequence.

In some cases, such as where glycosylation is desired in a eukaryotic host cell expression system, one may manipulate the various presequences to improve glycosylation or yield. For example, one may 48 alter the peptialase cleavage site of a particular signal peptide, or add prosequences, which also may affect glycosylation. The final protein produict may have, in the -1 positioin (relative to the first amino acid of the mature protein) one or more additional amino acids incident to expression, which miay not have be6n iotally reirioved. For example, the final protein product may have one or two anino acid found in the pepitidase cldavge site, attached to the N-teriini. Alternatively, use of somte-enzymne cleavage sites-may result in a slightly truncated formof the desiid Polypeptide, if the eizye cut. at such aiea within the mature polypeptide.

The expiession vecto-rs of thie present ihvention will typically contaiia prmioter thit is re6gizedby the hist organism and operably linld to A nuleic acid iLolule encobdiig nt itibody dor antigen binding ddfiaiin. Eithe a native or hetrolbgous promitner may be used depending the host ell uised for expression arind the yielt of protein desired.

Proioters suitable for use with prdkaiyotic hosts include the beta- Sactamas aind lactose proinoter sysfem allike phosphatase, a tryptophari (frp) promoter system; aird hybrid promioters such as the tac proioier. Other known bacterial promoters are also suitable. Their selubnces have been published, thereby enablig one skilled in the art to ligat6 thh to the desired DNA sequence(s), using linkers or adapters as needd t6 sdu pply any required restriction sites.

Siuitable promi-oters for ise with yeast hosts are also well known in the art. Yedst enhancers are advaitageously us&d with yeast prorimoters.

Suitable promoters for use with mammalian host cells are well known and include those obtained from the genomes of viruses such as polyoma virus, fowlpox virus, adeniovirus (such as Adenovirus bovine papilloia irirus, avian sarcoma virus, cytomegalovirus a retrovirus, hepatitis-B virus and most preferably Simian Virus 40 (SV40). Other 49 suitable-marnialian promoters include heterologous mammalian promoters, heat-shock promoters and the actin promoter.

Additional promoters which may be used for expressing the specific binding partners of the invention include, but are not limited to: the SV40 early promoter region (Benoist and Chambon (1981), Nature 290:304-310); the CMV promoter; the promoter contaired in the 3' long terminal repeat of Rous sarcoma virus (Yamamoto et al (1980), Cell, 22: 787-97); the herpes thymidine kinase promoter (Wagner et al. (1981), Proc.

Natl. Acad Sci. U.S.A, 78: 1444-5); the regulatory sequences of the metallothionine gene (Brinster et al (1982), Nature, 296: 39-42): prokaryotic expressiont vectors such as the beta-lactamase promoter (Villa-Kamaroff et al. (1978), Proc. Natl. Acad Sci U.S.A.,75: 3727-31); or the tac promoter (DeBoer, et al. (1983), Proc..Natl. Acad. Sci. 80: 21- Also of interest are;the following animal transcriptional control 5 regions, which exhibit tissue specificity and have been utilized in transgenic animals; the elastase I gene controlregionwhich is active in pancreatic acinar cells (Swift et al. (1984), Cel 38: 639-46; Ornitz et al.

(1986), Cold Spring Harbor Svmp. Ouant. Biol. 50:399-409; MacDonald (1987), Hepatology 7: :425-515); the insulin gene control region which is active in pancreatic beta cells (Hanahan (1985), Nature, 315: 115-122); the immunoglobulin gene control region.which is active in lymphoid cells S (Grosschedl et al. (1984), Cel; 38: 647-58; Adames et al. (1985), Nature 318: S 533-8; Alexander et al. (1987), MoL Cell. Biol., Z: 1436-44); the mouse mammary tumor virus control region which is active in testicular, breast, S 25, lymphoid and mast cells (Leder et al. (1986), Cell 45:485-95), albumin gene control region which is active in liver (Pinkert et al. (1987), Genes and Devl., 1: 268-76); the alphafetoprotein gene control region which is active in liver (Krumauf et al. (1987), Mol. Cell. Biol., 5:1639-48; Hammer et al.

(1987), Science 235: 53-58); the alpha 1-antitrypsin' gene control region 50 which is active in the liver (Kelsey etal. (1987), Genes and Devel., 1: 161- 171); the beta-globin gene control region which is active in myeloid cells (Mogram etal. (1985), Natur 315: 338-340; Kollias et al. (1986), Cell, 46: '89-94); the myelin basic protein gene control region which is active in oligodendrocyte cells in the brain (Readhead etal: (1987), Cell 48: 703- 712); the myosin light chain-2 gene control regior which is active in skeletal muscle (Sani (1985), Nature 314: 283-286); and the gonadotropic releasing hormone gene control region which is active in the hypothalamus (Mason et al. (1986), Science,234:1372-8).

An enhancer sequence may be inserted into the vector to increase transcription in eucaryotic host cells. Several enhancer sequences available from mammalian genes are known globin, elastase, albumin, alpha-feto-protein and insulin). Typically, however, art enhancer from a virus willbe used. The SV40 enhancer, the cytomegalovirus early promoter enhancer, the polyoma enhancer, and adenovirus enhancers are exemplary enhancing elements for the activation of eukaryotic promoters.

SWhile'an enhancer may be spliced into the vector at a position 5' or 3' to the polypeptide coding region, it is typically located at a site 5' from the promoter.

2 0 Preferred vectors for practicing this invention are those which are compatible with bacterial, insect, and mammalian host cells. Such vectors include, inter alia pCRII, pCR3, and pcDNA3.1 (Invitrogen Company, San Diego, CA), pBSII (Stratagene Company, La Jolla, CA), pET15 (Novagen, Madison, WI), pGEX (Pharmacia Biotech, Piscataway, NJ), pEGFP-N2 (Clontech, Palo Alto, CA), pETL (BlueBacIl; Invitrogen), pDSR-alpha (PCT Publication No. W090/14363) and pFastBacDual (Gibco/BRL, Grand Island, NY).

Additional possible vectors include, but are not limited to, cosmids, plasmids or modified viruses, but the vector system must be compatible 51 with the selected host cell. Such vectors include, but are not limited to plasmids such as Bluescript plasmid derivatives (a high copy number ColEl-based phagemid, Stratagene Cloning Systems Inc., La Jolla CA), PCR cloning plasmids designed for cloning Taq-amplified PCR products TOPOTM TA Cloning® Kit, PCR2.1® plasmid derivatives, Invitrogen, Carlsbad, CA), and mammalian, yeast or virus vectors such as a baiclovirus expression system (pBacPAK plasmid derivatives, Clontech, Palo Alto, CA). The'recombinant molecules can be introduced into host S cells via transformation, transfection, infection, electroporation, or other known techniques.

-Host cells of the invention may be prokaryotic host cells (such as E.

cli or eukaryotic host cells (such as a yeast cell, an insect cell, or a vertebrate cell). Prokaryotic host cells such as E. coli produce unglycosylated protein; for example, unglyclosylated shBCMA and uhglycosylated shTACI, which may possess advantages over the glycosylated eukaiyotic molecules. The host cell, when cultured under appropriate conditions, expresses an antibody or antigen binding domain of the invention which can subsequently be collected from the culture S medium (if the host cell secretes it into the medium) or directly from the host cell producing it (if it is not secreted). Selection of an appropriate host cell will depend upon various factors, such as desired expression levels, polypeptide modifications that are desirable or necessary for activity, such as glycosylation or phosphorylation, and ease of folding into a biologically active molecule.

A number of suitable host cells are known in the art and many are available from the American Type Culture Collection (ATCC), Manassas, VA. Examples include mammalian cells, such as Chinese hamster ovary cells (CHO) (ATCC No. CCL61) CHO DHFR- cells (Urlaub etal. (1980), Proc. Natl. Acad. Sci. USA 97, 4216-20), human embryonic kidney (HEK) 52 293 or 293T cells (ATCC No. CRL1573), or 3T3 cells (ATCC No. CCL92).

The selection of suitable mammalian host cells and methods for transformation,.culture, amplification, screening and product production and purification are known in the art. Other suitable mammalian cell lines, are the monkey COS-1 (ATCC No. CRL1650) and COS-7 cell lines (ATCC No. CRL1651), and the CV-1 cell line (ATCC No. CCL70). Further exemplary mammalian host cells include primate cell lines and rodent cell lines, including transformed cell lines. Normal diploid cells, cell strains derived from in vitro culture of primary tissue, as well as primary explants, are also suitable. Candidate cells maybe genotypically deficient in the selection gene, or may contain a dominantly acting selection gene.

Other suitable mammalian cell lines include but are not limited to, mouse neuroblastoma N2A cells, HeLa, mouse L-929 cells, 3T3 lines derived from Swiss, Balb-c or NIH mice, BHK or HaK hamster cell lines, which are available from the American Type Culture Collection, Manassas, VA).

Each of these cell lines is known by and available to those skilled in the art of protein expression.

Similarly useful as host cells suitable for the present invention are bacterial cells. For example, the various strains of E. coli HB101, 2 0 (ATCC No. 33694) DH5a, DHi0, and MC1061 (ATCC No, 53338)) are well-known as host cells in the field of biotechnology. Various strains of Pseudomonas spp., B. subtilis, other Bacillus spp., Streptomyces spp., and the like may also be employed in this method.

Many strains of yeast cells known to those skilled in the art are also available as host cells for expression of the polypeptides of the present invention. Preferred yeast cells include, for example, Saccharomyces cerivisae.

Additionally, where desired, insect cell systems may be utilized in the methods of the present invention. Such systems are described for 53 example in Kitis et al. (1993), Biotechniques, 14: 810-7, Lucklow (1993), Curr. Opin. Biotechnol., 4:564-72, and Lucklow etal. (1993), Virol. 67: 4566-79. Preferred insect cells are Sf-9 and Hi5 (Invitrogen, Carlsbad, CA).

Transformation or transfection of a nucleic acid molecule encoding a specific binding partner into a selected host cell may be accomplished by well known methods including methods such as calcium chloride, electroporation, microinjecti6n, lipofection or the DEAE-dextran method.

The method selected will in part be a function of the type of host cell to be S used. These methods and other suitable methods are well known to the i1 0 skilled artisan, and are set forth, for example, in Sambrook et al, supra.

One may also use transgenic animals to express glycosylated specific binding partners; such as antibodies and antigen binding domain.

For example, one may use a transgenic milk-producing animal (a cow or goat; for example) and obtain glycosylated binding partners in the animal milk. Alternatively, one may use plants to produce glycosylated specific binding partners.

Host cells comprising (as by transformation or transfection) an expression vector encoding a spedfic binding partner of the target molecule may be cultured using standard media: well known to the skilled artisan. The media will usually contain all nutrients necessary for the growth and survival of the cells. Suitable media for culturing E. coli cells are for example, Luria Broth (LB) and/or Terrific Broth Suitable media for culturing eukaryotic cells are RPMI 1640, MEM, EM,D M, all of which may be supplemented with serum and/or growth factors as required by the particular cell line being cultured. A suitable medium for insect cultures is Grace's medium supplemented with yeastolate, lactalbumin hydrolysate, and/or fetal calf serum as necessary.

Typically, an antibiotic or other compound useful for selective growth of transfected or transformed cells is added as a supplement to the 54 media. The-compound to be used will be dictated by the selectable marker element present on the plasmid with which the host cell was transformed. For example, where the selectable marker element is kanamycin resistance, the compound added to the culture medium will be kanamycin. Other compounds for selective growth include ampicillin, tetracycline and neomycin.

The amount of an antibody or antigen binding domain produced by a host cell can be evaluated using standard methods known in the art.

Such methods include, without limitation, Western blot analysis, SDSpolyacrylamide gel electrophoresis, non-denaturing gel electrophoresis, HPLC separation, immunoprecipitation, and/or activity assays.

.Purification of a specific binding partner that has been secreted into the cell media can be accomplished using a variety of techniques including affinity, immunoaffinity or ion exchange chromatography, molecular sieve chromatography, preparative gel electrophoresis or isoelectric focusing, chromatofocusing, and high pressure liquid chromatography. For example, antibodies comprising a Fc region may be conveniently purified by affinity chromatography with Protein A, which selectively binds the Fc region Modified forms of an antibody or antigen 2 0 binding domain may be prepared with affinity tags, such as hexahistidine or other small peptide such as FLAG (Eastman Kodak Co., New Haven, CT) or yc (Invitrogen) at either its carboxyl or amino terminus and purified by a one-step affinity column. For example, polyhistidine binds with great affinity and specificity to nickel, thus an affinity column of 2 5 nickel (such as the Qiagen® nickel columns) can be used for purification of polyhistidine-tagged specific binding partners. See for example, Ausubel et al., eds. (1993), Current Protocols in Molecular Biology, Section 10.11.8, John Wiley Sons, New York. In some instances, more than one purification step may be required." 55 Specificbinding partners of the invention which are expressed in procaryotic host cells may be present in soluble form either in the periplasmic space or in the cytoplasm or in an insoluble form as part of intracellular inclusion bodies. Specific binding partners can be extracted from the host cell using any standard technique known to the skilled artisan. For example; the host cells can be lysed to release the contents of the periplasm/cytoplasm by French press, homogeriization, and/or sonication followed by centrifugation.

SSoluble forms of an antibody or antigen binding domain present either in the cytoplasm or released from the periplasmic space may be S further purified using methods known in the art, for example Fab S fragments are released from the bacterial periplasmic space by osmotic Sshock techniques.

If an antibody or antigen binding domain has formed inclusion bodies, they can often bind to the inner and/or outer cellular membranes and thus will be found primarily in the pellet material after centrifugation.

The pellet material can then be treated at pH extremes or with chaotropic partner such as a detergent, guanidine, guanidine derivatives, urea, or urea derivatives in the presence of a reducing partner such as 2 0 dithiothreitol at alkaline pH or tris carboxyethyl phosphine at acd pH to release, break apart, and solubilize the inclusion bodies. The soluble -specific binding partner can then be analyzed using gel electrophoresis, immunoprecipitation or the like. If it is desired to isolate a solublized antibody or antigen binding domain, isolation may be accomplished using 2 5 standard methods such as those set forth below and in Marston et al.

(1990), Meth. Enz., 182: 264-75.

In some cases, an antibody or antigen binding domain may not be biologically active upon isolation. Various methods for "refolding" or converting the polypeptide to its tertiary structure and generating 56 disulfide linkages, can be used to restore biological activity. Such methods include exposing the solubilized polypeptide to a pH usually above 7 and in the presence of a particular concentration of a chaotrope.

The selection of chaotrope is very similar to the choices used for inclusion body solubilization, but usually the chaotrope is used at a lower concentration and is not necessarily the same as chaotropes used for the solubilization. In most cases the refolding/oxidation solution will also contain a reducing partner or the reducing partner plus its oxidized form in a specific ratio to generate a particular redox potential allowing for disulfide shuffling to occur in the formation of the protein's cysteine Sbridge(s). Some.of the commonly used redox couples include cysteine/cystamine, glutathione (GSH)/dithiobis GSH, cipric chloride, dithiothreitol(DTI)/dithiane DTT, and 2-mercaptoethanol(bME)/dithiob(ME). In many instances, a cosolvent may be used or may be needed to increase the efficiency of the refolding and the more common repartners used for this purpose include glycerol, polyethylene glycol of various molecular weights; arginine and the like Specific binding partners of the invention may also be prepared by chemical synthesis methods (such as solid phase peptide synthesis) using techniques known in the art such as those set forth by Merrifield et al.

S(1963), F. Am. Chem. Soc., 851 2149; Houghten etal. (1985), Proc Natl Acad.

Sci. USA, 82: 5132; and Stewart and Young (1984), Solid Phase Peptide Synthesis, Pierce Chemical Co., Rockford, IL. Such polypeptides may be synthesized with or without a methionine on the amino terminus.

Chemically synthesized antibodies and antigen binding domains may be oxidized using methods set forth in these references to form disulfide bridges. Antibodies so prepared will retain at least one biological activity associated with a native or recombinantly produced antibody or antigen binding domain.

f 57 The invention will now be further described by specific experimental examples. These examples are meant to be illustrative rather than limiting.

Working Examples Materials and Methods Isolation of BCMA and TACI cDNA' Mouse and human BCMA cDNA wiere isolated by PCR using the mouse BCMA sense primer 5'-CACAATACCTGTGGCCCTCTTAAGAG-3' (SEQ ID and antisense primer S5'-TGGTAAACGGTCATCCTAACGACATC-3' (SEQ ID NO:26), the human BCMA sense primer.

5'-TTACTGTCCCC AGGCTGTTCT-3' (SEQ ID NO; 27), and antisense primer 5'-CATAGAAACCAAGGAAGTTTCTACC-3' (SEQ ID NO: 28).

For isolation of human TACI cDNA, the sense primer 5'-AGCATCCTGAGTAATGAGTGGCCTGG-3' (SEQ ID NO:z29) and antisense primer 5'-GTGATGACGACCTACAGCTGCACTGG-3' (SEQ ID NO: were used. Poly RNA from the mouse B lymphoma cell line and human lymph Node were reverse -transcribed and cDNA were synthesized by using the Smart RACE cDNA amplification Kit (Clontech, palo Alto, California). The full-length cDNA of mouse and human BCMA 58 genes as well as human TACI gene were cloned into pcDNA3 vector for mammalian cell expression (Invitrogen, Carlsbad, California).

Recombinant proteins Soluble munne APRIL-Flag protein was generated by fusing Flag sequence in frame to the N-terminis of APRIL ainio acid 101-239.

Soluble mAPRIL-Flag protein was expressed in E.:coli and the refolded protein was affinity- purified by anti-Flag M2 antibody column. Fc-tagged AGP3 protein was generated by fusing OPG signal peptide followed by human IgG-yl Fc in frame to the N-terminus of AGP3 amino acid 128-285.

The protein was expressed in baculovirus and purified with protein A sepharose column. Fc-tagged human APRIL was ecoded by a similar construct (see Figure 21) and expressed in CHO cells and purified with protein A sepharose coloumn.

Soluble TACI protein (amino acids 1-165) and BCMA protein (amino acid 4-55) followed by human IgG-yl Fc in frame was expressed in E. coli. The inclusion bodies formed were solubilized. The refolded protein was purified by cation exchange chromatography.

Purification of human BCMA-Fc Purification'of human BCMA-Fc produced in recombinant E.coli was initiated by solubilizing 53.3 g of a washed inclusion body preparation in a solution having a final composition of 6M guanidinium chloride, 50mM Tiis(HCl), 8.0 mM dithiothreitol. The final volume of solution was approximately 200mL. The solution'was adjusted to pH at 23 degrees Centigrade and allowed to stir at room temperature for one hour.

The solubilized polypeptide solution was then added to 3.8 L of cold 4 M urea, 160 mM L-arginine, 20% (volume) glycerol, 4 mM cysteine, 1 mM cystamine, 50 mM Tris. The pH was adjusted to 8.9 at 12 degrees 59 Centigrade and the mixture was allowed to stir at 4-8 for approximately hours.

The refolding mixture was then clarified by filtration through a 0.9 square foot Cuno 10SP cartridge. The filtrate was then concentrated approximately five fold and diafiltered with five retehtate volumes of mM sodium phosphate/40 mM sodium chloride, pH 9 using a 1.0 square foot Pall Filtron regenerated cellulose TFF cassette having a nominal a molecular weight cutoff of 50 kDa. Processing was performed at 4-8 degrees Centigrade.

The retentate was removed from the unit, warmed to 20 degrees S Centigrade, and adjusted to pH 5.0 using 1 M acetic acid. The precipitated solids were then removed by centrifugation in a Beckman J6-B centrifuge operating at 20 degrees Centigrade at approximately 2500 x g for approximately 15 minutes. The supernatant was aliquotted and stored frozen at-30 degrees Centigrade. Aliquots of supernatant were thawed and processed over a 5.0 cm diameter x 27.5 cm height column of SP-Sepharose Fast Flow. All phases of the chromatographywere performed at 8 degrees Centigrade. The column was equilibrated using 25mM sodium phosphate, pH 6.5. The product was conditioned for loading by adjusting the pH to Following the load, the column was washed with equilibration buffer, and the product was then eluted with a linear gradient from 17 mM to 60 mM sodium chloride in a background of 25 mM sodium phosphate pH over 20 column volumes. Fractions were selected for further processing based on their appearance on a Coomassie blue-stained SDS polyacrylamide gel.

The pooled SP-Sepharose Fast Flow fractions were further processed over a 5.0 cm diameter x 27.5 cm height column of Butyl Toyopearl 650M, operated at 21 degrees Centigrade. The product was 60 conditioned prior to loading by adding solid potassium phosphate to give a final concentration of approximately 0.6 M. The column was equilibrated using 0.6M potassium phosphate, pH 7.0, After loading, the column was washed with 0.5 bed volumes of equilibration buffer, and the product was then eluted using a linear gradient from 0.5 M to 0.3 M potassium phosphate over 15 column volumes. Fractions from the predominant UV elution peak were pooled for further processing.

The pooled fractions from the Butyl Toyopearl chromatography were concentrated twenty fold and diafiltered against six retentate volumes of phosphate buffered saline using a Millipore XL TFF cassette, operated at 4-8 degrees Centigrade.: The final retentate pool was aliquotted and stored at -30 degrees Centigrade.

In vivo study ,;B6 mice (6-8 weeks old)were purchased from Charles River Laboratories and murine APRIL-Flag and other TNF proteins were Sinjected i.p. of 1 mg/kg/day for 5 days. On day 7, cells from mouse spleens and mesenteric lymph nodes were collected and B and T cell activation and differentiation was analyzed by FACS using specific monoclonal antibodies staining.

Cell lines and proliferation assays 293 human kidney epithelial cells, Raji Burkitt lymphoma, human T lymphoblastoma Jurkat cells and A20, mouse B lymphoma cell line were purchased from the American Type Culture Collection (Rockville, Maryland).Raji, Jurkat and A20 cells were maintained in a complete medium of RPM-1640 (life Technologies) supplemented with bovine serum (HyClone, Logan, Utah)and 25 mM HEPES. 293 cells were cultured in Dulbecco's modified Eagle's medium (Life Technologies) with fetal bovine serum.The proliferation of cells were determined by 61 incubating 5x10 4 cells /well in 100 pL medium with the indicated concentration of APRIL-flag protein using the celltiter 96 AQ proliferation assay Promega Corp.,Madison, WI) following the manufacturer's instructions. Alternatively, cells were pulsed for 18 h with 'H thymidine (0.5 gCi/well), after harvesting cells, H thymidine incorporation was monitored by liquid scintillation counting.

Transfectioti and Flow cytometric analysis For 293 cell expressing BCMA and TACI receptor, 2X10 6 293 cells S were plated into 6 well plate, cells were transfected with lipofectAMINE 2000 following the manufacturer' s procedure (Life Techiologies), 48 h after transfection, cells were collected and incubated at 4 C with 1 pg/ml APRIL-Flage ligand or Blys (AGP3)-Fc ligand for 60 nin, after washing 3 times with PBS (containing 2% FBS); cells were stained with FITCconjugated secondary antibody for 30 min:, then washed 3 times with PBS and fluorescence was analyzed by FACS scanner (Becton Dickinson, Mountain View, California).

Determination of the binding affinities of APRIL and TALL-1 for BCMA and TACI Biomolecular interaction analysis (BIA) was performed using a BIACORE 2000 (Biacore AB, Uppsala, Sweden). The receptors, BCMA-Fc and TACI-Fc (2 Lg/ml in 10 mM sodium acetate, pH were immobilized on Sensor Chip CM5 using the BIACORE standard amine coupling procedure.. An immobilization level of approximately 120 RU's was achieved. The analytes, Flag-APRIL and Fc-AGP-3 were diluted between 100 nM-0.01 nM in running buffer (10 mM HEPES, 0.5 M NaC1, 3 SmM EDTA, 0.005% Tween 20, 2mg/ml CM dextran, pH The analytes were injected over an immobilized receptor surface for 2 minutes at pl/min and allowed to dissociate for 10 minutes. Bound protein was removed by a Iminute injection of 50 mM HC1. Binding affinities were 62 determined using a 1:1 Langmuir model (BIA Evaluation software Version 3.1.2, BIACORE).

T cell co-stimulation assay T cells from the spleens of C57 BI/6 mice were purified by negative selection through a murine T cell.enrichment column (R&D Systems). T cells (1xl10 per well). were cultured in the absence or presence of various APRIL-Flag protein for 48 hr. Alternatively, 96 well plates were precoated Swith subliminal quantities of anti-CD3 antibody T cells were treated with APRIL-Flag protein for 72 hr, pulsed during the last 18 hr with 1 pCi of 3

H

thymidine and harvested to count the incorporation radioactivity.

B cell proliferation and Ig secretion: Mouse B cell were negatively selected from spleens by mouse B cell recovery column (Cedarlane, Homby, Ontario Canada). Ixl0'/ml were seeded in 96 -well flat bottom tissue culture plates in medium (RPMI- 1640, 5% FBS, 5x10 5 M 2 ME, affinity-purified goat anti-mouse:IgM ug/ml Pharmingen, San Diego). B cells were then treated with APRIL Flag protein plus different concentration of soluble BCMA-Fc protein for 72 hr and culture received 1 pCi of 3 H thymidine during the last 18 hr.

proliferation of B cell was quantitated by measuring the incorporation of radioactivity.

For analysis of Ig secretion from B cells, purified B cells 5x105/ml were cultured in 96-well flat bottom tissue culture plates in the presence of APRIL-Flag for six days. The culture supernatant were harvested and S IgG, IgM and IgA levels were determined by an isotype specific sandwich ELISA technique. Ig concentration in test samples were determined by comparing triplicate test values with isotype control standard.

Induction and detection of anti-keyhole limpet hemocyanin (KLH) and anti-Pneumovax antibodies.

63 Mice (Bab/c females of 9-11 wk and 19-21 g, Charles River Laboratories, Wilmington, MA) were immunized on day 0 with 100 Ag of KLH (Pierce, Rockford, IL) in CFA s.c. or with 115 gg of Pneumovax (Merck, West Point, PA) i.p. Starting on day 0, mice received 7 daily i.p.

injections of 5 mg/Kg of either TACI-Fc or BCMA-Fc fusion proteins or non-fused Fc and were then bled on day 7. Anti-KLH and anti- Pneumovax IgG and IgM were measured in serum by ELISA. Briefly, for the measurement of anti-KLH antibodies, plates were coated with KLH in PBS, blocked, and added with dilutions of standard and test samples.

Captured anti-KLH IgG or IgM were revealed using anti-IgG or anti-IgM biotinylated antibodies and neutravidin-conjugated HRP. For the measurement of anti-Pneumovax IgM, plates were coated with Pneumovax using poly-L-lysine, blocked, and added with dilutions of standard and test samples. Captured anti-Pneumovax Ig were revealed using an anti-IgMbiotinylated antibody and neutravidin-conjugated HR-P.

Results were compared with the Student t test.

In a second experiment, normal mice were treated with human BCMA-Fc,truncated TACI-Fc and nonfused Fc as a control in daily doses of 0.5mg/kg to 15 mg/kg for 7 days, At preimmunization anti-KLH and anti-Pneumovax were undetectable. Antibodies were measured on day 7 and Day 14. hBCMA-Fc effect on mouse peripheral blood B cell in vivo. Normal mice (15 mg/kg ip on day 0, 3, and 6; n=7) were treated with human BCMA-Fc and nonfused Pc as a control in day 0,3, and 6 Peripheral blood and spleen B cell level were measured on day 7.Flag-mAPRIL, hAGP3, hBCMA-Fc and hTACI-Fe effects on immunoglobulin production in vitro. Purified murine spleen B cells were cultured with LPS (100ng/ml),AGP3 (1Ong/ml),Flag-APRIL (10ng/m) or plus BCMA-Fc (100ng/ml) and TACI-Fc (100ng/ml) for 12 days. Culture supernatants were collected on day 7 and day 12 for detecting IgA and 64 IgG levels. mBCMA-Fc and trun hTACI-Fc effect on immunoglobulin levels in vitro.

Normal mice were treated with mBCMA-Fc; trun hTACI-Fc; or Fccontrol (5 mg/kg ip day 0, 3, and 6; Iinminoglobulin levels in serum were measured on day 7.

Generation and analysis of anti-mAPRIL specific monoclonal antibody.

Flag-mAPRIL was used as antigen for generation of rat anti-mAPRIL spefic monolonal antibodie following standard pr6cedures. Fourty i clones that recognized Flag-APRPL inri an ELISA were identified. The clones effect ofn iag-inAPRIL mediated muise B cell proliferation was deferinied. Purified mririee spleen B cel. weire cultried in presence 1Ong/ni iFlag-APRIL plhis 2 ug/ml of aiti-IgMi anti-Flag-APRIL moiodonal aibody and rat Ii control were added into culture in same 'time. Data sho incorporation of 3H thymidine as cpinand represent mean of triplicate wells. Anti-mAPRIL specific monoclonal antibody #c19 mg/ig ip on dy 3, and 6 as ue for determining the effect of blocking endogenous APRIL on anti-Pheumovacs IgM In Vivo.

NZBxNZ61 uIipuits model.

Five months old NZBxNZWF1 mice were treated 3 times/week (moi we, ie fdr a period of 5 months with the indicated amount of protein. 100 ug dose translates to 4mg/Kg. Mice were bled at day of treatment to later be analyzed for DNA-specific antibodies, histone proteins, etc. Also at day 0, urine:were taken and analyzed for proteins levels. Animals are exclided that already have high levels of protein in the urine, a sign of lupus. Treatment was injected i.p. Urine and blood are taken every 30 days and analyzed until day 1501 APRIL binding and stimulation of tumor cells.

65 APRIL binding to human tumor cell lines were determined by incubating cell with lug/ml Fc-APRIL or Flag-APRIL following FITC labeled secondary antibody staining and FACS analysis. APRIL and BCMA-Fc or TACI-Fc effect on U266-B1 cell growth: U266 cells were cultured in presence 10ng/ml human Fc-APRIL or plus 50ng/ml soluble BCMA-Fc, Truncated TACI-Fc for 48 hr. Incorporation of 3H thymidine is indicated as cpm. Data shown represent mean of triplicate mouse B cell lymphoma were cultured in presence 50ng/ml mouse Flag- APRIL,human Flag-AGP3 or plus 100 ng/ml soluble BCMA-Fc for 48 hr.

Incorporation of 3H thymidine is indicated as cpm. Data shown represent mean of triplicate wells.

B lymphoma tumor cell gro-kth in Balb/c mice.

(0.2 million) cells were' implanted, id, on day 0. Treatments with PBS, CHO-Fc, mBCMA-Fc, hBCMA-Fc, mTACI-Fc, hTACI-Fc were given (10 mg/kg, ip) on days 0, 7,10,13,16j 1922. Tumor measurements were made twice per week. Mice were sacrificed on days 27-31, tumors were snap frozen for RNA isolation, blood was collected and serum samples were frozen.

Human colon carcinoma cell line HT29 tumor growth.

2 x 10 6 HT29 cells plus 50% matrigel injected subcutaneously into athymic nude mice. Rx: human or mouse BCMA-Fc at 2,5, and 15 mg/kg Q2D, starting at day 0 (n=10/group). Control 1: CHO-Fc at 15 mg/kg Q2D. Control 2: 0.2 ml of PBS Q2D IF. Tumor volume: 3/week, from day 7. Tumor weight at end of study. Body weight 2/week.

Results in vitro function Human and mouse G70, also called APRIL, was isolated and characterized (Figures 1, 2, and FLAG-tagged soluble mouse was produced in E. coli purified and refolded (Figure Soluble 66 (smG70) specifically stimulates B and T cell lymphoma cell proliferation in a dose-dependent manner (Figure Furthermore, soluble 1) specifically binds to cell-surface receptors expressed on human B and T lymphoma cells (Figure 2) specifically stimulates proliferation of purified human peripheral blood B and T cells (Figure 6); 3) stimulates proliferation of purified murine spleen B and T cells in a dose-dependent manuier (Figure 4) acts synergistically with anti-CD28 antibody to stimulate proliferation of purified murine T cells (Figure 8); .,has a strong costimulatory activity on purified murine T cells (Figure 9) in the presence of sub-optimal concentration of the T cell receptor activator: anti-CD3 antibody.

G70/APRIL in vivo function A series of experiments were performed to elucidate soluble biological activity in normal mice in vivo. Each group consisted of 5 mice (BDF-1, 8 weeks of age, dosed at 1 mg/kg/day, 0.2 ml for 5 days). Spleen, thymus and mesenteric lymph nodes from three mice of each group was.used for FACS analysis using a panel of T cell and B cell surface marker antibodies and all the mice were analyzed by standard necropsy and pathological analysis.

Spleen (Table 1A): murine soluble G70 caused an average about decrease in the percentage of CD3* T cells. In addition, there was an average 5-fold increase in T-helper cells activation and an average 22-fold increase in cytotoxic T cell activation as measured by IL-2 receptor expression. In addition the percentage of immature B cells increased about 2-fold while the percentage of mature B-cells increased 3- to 4-fold.

67 The total percentage of lymphocytes was unchanged compared to control.

Mesenteric Lymph Nodes (Table 1B): soluble G70 treated mice had an average of 25% decrease in the percentage of T cells. There was an average 3-fold increase in activated T-helper cells and 36-fold increase in activated cytotoxic T-cells as measured by CD25/IL-2 receptor expression. In addition the percentage of immature B cells was increased S, on average 2-fold whereas mature B cells were up on average 4-fold.

In summary our preliminary observations indicate that stimulates both T and B cells in the spleen and mesenteric lymph nodes.

Pathological analysis revealed that soluble G70 treated mice have slightly enlarged spleens of normal morphology..

is a ligand for BCMA and TACI is related to the TNF ligand family member AGP3/BlyS. The TNFR receptor family member TACI (Fig.12) was recently shown to be a receptor for AGP3 ([A-570A patent application ser.

Furthermore, TACI has a match to the orphan TNFR receptor family member BCMA (Figure 10) in a conserved extracellular cysteine rich domain (Figure 13). These observations together prompted us to investigate whether G70/APRIL is a ligand for BCMA and TACI and to test whether in addition to TACI AGP3/BlyS is also a ligand for BCMA.

,Soluble mouse G70 specifically binds to 293 cells expressing exogenous BCMA (Figure 14). G70 also binds to 293 cells expressing TACI (Figure 15). Furthermore soluble G70 specifically blocks AGP3/BLyS binding to cell-surface receptors located on mouse B lymphoma cells (Figure 16). This, suggest that G70 and AGP3 both binds to BCMA and

TACL.

smBCMA-Fc and shTACI-Fc prevent G70 and AGP3 ligand binding to cell-surface receptors 68 SolubleBCMA (smBCMA-Fe; Fig. 10) and soluble TACI (shTACI- Fc) were produced in E. coli purified to homogeneity and refolded.

Soluble TACI receptor specifically prevents G70 from binding to mouse B cells. (Figure 17). Furthermore, shBCMA-Fc and shTACI-Fe both prevent binding of AGP3 to B cells (Figure 18; arid Figure 19 Soluble hBCMA-Fc also ameliorates G70 binding to A20 cells (Figure 19 B).

In summary 1) both G70 and AGP3 binds the orphan TNFR receptor family members TACI and BCMA; 2) soluible BCMA and TACE both effectively inhibits G70 and AGP3 from binding to B cells; 3) G70 and AGP3 competes for binding to cell-surface receptors.

Effects of TACI-Fc and BCMA-Fc treatment on the production of anti-KLH and anti-Pieumovax antibddies.

Treatment- with either TAC I-Fc or BiMA-Fc significantly inhibited the prodtion of anti-KLH and aniti-Pneiimovax antibodies. Serum levels of both aiti-KLH IgG and IgM were approximately 25% and 19% lower, respectively, in the TACI-Fc-treated mice than controls (Figure 20). Serum anti-KLH IgG aid IgM were approximately 526/% and 66% lower, respectively, in the BCMA-Fc-treated mice than controls (Figure Seirum levels of anti-Pneumovax IgMl were also li6er in the TACI-Fc- and BCMA-Fc-treated mice than controls (24% and 42%, respectively, Figure S 20). Second experiment: Treatment with hBCMA-Fc or Trun-hTACI-Fc reduces anti-Pneumovacs specific IgM levels in normal mice (Figure 29).

Normal mice were treated with human BCMA-Fctruncated TACI-Fc and nonfiused Fc as a control in daily doses of 0.5mg/kg to 15 mg/kg for 7 daiys. At preimmunization anti-KLH and anti-Pneumnovax were undetectable. Antibodies were measured oh day 7 and Day 14.

Effects of hBCMA-Fc on progression of lupus in NZBXNZWF1 mice.

69 Treatment of NZB/NXWF1 mice with human BCMA-Fc increased survival, decreased proteinurea, decreased the level of anti-dsDNA specific antibodies, and decreased B cells in peripheral blood. Protocol for SLE: Five months old NZBxNZWF1 mice were treated 3 times/week (mon, wed, fri) for a period of 5 months with the indicated amount of protein. 100 ug dose translates to 4mg/Kg. Mice were bled at day of treatment to later be analyzed for DNA-specific antibodies, histone proteins, etc. Also at day 0, urine were taken and analyzed for proteins t levels. Animals are excluded that already have high levels of protein in the urine, a sign of lupus. Treatment was injected i.p. Urine and blood are taken every 30 days and analyzed until day 150.

Effects of hBCMA-Fc on A20 lymphoma cell growth in vivo.

Treatment with BCMA-Fc reduces A20 B lymphoma tumor cell growth in Balb/c mice. A20 (0.2 million) cells were implanted, id, on day 0. Treatments with PBS, CHO-Fc, mBCMA-Fc, hBCMA-Fc, mTACI-Fc, hTACI-Fc were given (10 mg/kg, ip) on days 0,7,10,13, 16,19,22. Tumor measurements were made twice per week. Mice were sacrificed on days 27-31, tumors were snap frozen for RNA isolation, blood was collected and serum samples were frozen.

2 0 Effects of hBCMA-Fc on human colon carcinoma HT29 cell growth in vivo.

Treatment with hBCMA-Fc reduces human colon carcinoma cell line HT29 tumor volume growth in mice. 2x106.cells plus 50% matrigel injected subcutaneously into athymic nude mice. Rx: human BCMA-Fc at 2,5, and 15 mg/kg Q2D, starting at day 0 (n=10/group). Control 1: CIO- Fc at 15 mg/kg Q2D. Control 2: 0.2 ml of PBS Q2D IP. Tumor volume: 3/week, from day 7. Tumor weight at end of study. Body weight 2/week.

70 Fc-A1'RIL effect on B cells Fc-humanAPRTL and soluble humanAGP3/BlyS/Tall-1 stimulates incorporation of 3H thymidine in primary murine B cells. Purified murine spleen B cells were cultured in pre sence of various amounts of human Fc- APRIL and untaged AGM3 plus 2 u g/rnl of anti-IgM. Data show incorporation of 3H thymiddirte as cpmand repriesent mean of triplicate hBCMA-Fc effect' on'APRI mnedited B cell proliferation h BCMA-Fc and hTACI-Fc inhibits Flag-nxAPR L mediated mouse B cell proliferation (Figu re 23). Purifed murine spleenBcelwreutrd with the indicat Ida'mot ts of soluble BCMA-Fc and TACI-Fc in presence of 10 fig /1m of Flag-iA lUL ahid 2 ug/mi of: an anti-IgM for 72 hr.

Incorp oratiion. of 3H thymiddine is indicated as cpm. Data shown represent mean of'triplicate wells.

hBCM1A-Fc effect on peripheral blood B cell IevelshBCMA-Fit (15 mg/kg ip, on''day 0, 3, and 6) reduces m'ouse peripheral blood and B cell levels measured at day 'seven (Figure 24 and 25). Normal mice were treated With human BCMA-Fc and nonf used Pc: as a control in day 0,3, and 6 (15mg/kg). Peripheral blood and spleen B cell (B220) level were -measured on day 7.

Flag-n-APRIL and hAGP3 mediated IgA production is inhibited by hBCMA-Fc "and ;hTACI-Fc in ivito(Figure 26. Purified murine spleen B cells were cultured: with LI'S (lOOng/ml),AGP3 (l0ng/in-),Flag--APRIL (lOng/min) or plus BCMA-Fc (lO0ng/mI) an~d TACI-Fc (100ng/mi) for 12 days. Culture supernatants were collected on day 7 and day 12 for detecting IgA level. Flag-inAPRIL and hAGP3 mediated IgG production is inhibited by hBCMA-Fc and hTACI-Fc in vitro (Figure 27).

BCMA-Fc effect on total immunoglobulin in vivo.

71 Total IgE and IgA levels are reduced in normal mice treated with mBCMA-Fc and trun hTACI-Fc (5 mg/kg ip day 0, 3, and Normal mice were treated with human BCMA-Fc ,Truncated TACI-Fc and nonfused Fc as a control in day 0,3, and 6 (5mg/kg). Immunoglobulin levels in serum were measured on day 7.

anti-mAPRIL specific monoclonal antibody effect.

The anti-mAPRIL specific monoclonal antibody #c19 inhibits FlagmAPRIL mediated mouse B cell proliferation (Figure 30). Purified murine spleen B cells were cultured in presence lOng/ml mFlag-APRIL plus 2 ug/ml of anti-IgM anti-Flag-APRIL monoconal antibody c-19 and rat IgG control were added into culture in same time. Data show iicorporation of 3H thymidine as cpm,and represent mean of triplicate wel2. Treatment with the anti-mAPR specific monoclonal antibody #c19 inhibits generation of anti-Pneumovacs specific antibodies in vivo (Figure 31).

APRIL binding and effect on tumor cells.

APRIL binds to a number of tumor cell lines (Fig 36). APRIL binding to human tumor cell lines were determined by incubating cell with lug/ml Fc-APRIL or Flag-APRIL following FITC labeled secondary antibody staining and FACS analysis. APRIL stimulates U266-B1 cell growth and that this can be inhibited by BCMA-Fc or TACI-Fc (Figure 37).

U266 cells were cultured in presence lOng/ml human Fc-APRIL or plus soluble BCMA-Fc, Truncated TACI-Fc for 48 hr. Incorporation of 3H thymidine is indicated as cpm. Data shown represent mean of triplicate wells.

Claims (29)

1. A method of inhibiting B or T cell proliferation or activation in a mammal, which comprises administering a therapeutic agent comprising: a. a specific binding partner for TACI, wherein the specific binding partner has TACI antagonist activity; b. a specific binding partner for BCMA, wherein the specific binding partner has BCMA antagonist activity; c. both a and b; or d. a specific binding partner for TACI and BCMA, wherein the specific binding partner has TACI antagonist activity, BCMA antagonist activity or both.

2. A method of inhibiting APRIL activity in a mammal, which comprises administering a therapeutic agent comprising: a. a specific binding partner for TACI, wherein the specific binding partner has TACI antagonist activity; b. a specific binding partner for BCMA, wherein the specific binding partner has BCMA antagonist activity; c. both a and b; or d. a specific binding partner for TACI and BCMA, wherein the specific binding partner has TACI antagonist activity, BCMA antagonist activity or both.

3. A method of inhibiting TACI activity, BCMA activity, or both in a mammal, which comprises administering a specific binding partner for APRIL.

4. The method of Claim 3, further comprising administering a specific binding partner for AGP-3. 73 A method of increasing T cell proliferation in a mammal, which comprises administering a therapeutic agent comprising: a. a specific binding partner for TACI, wherein the specific binding partner has TACI agonist activity; b. a specific binding partner for BCMA, wherein the specific binding partner has BCMA agonist activity; c. both a and b; or d. a specific binding partner for TACI and BCMA, wherein the specific binding partner has TACI agonist activity, BCMA agonist activity or both.

6. A method of increasing APRIL activity in a mammal, which comprises administering a therapeutic agent comprising: a. a specific binding partner for TACI, wherein the specific binding partner has TACI agonist activity; a specific binding partner for BCMA, wherein the specific binding partner has BCMA agonist activity; c. both a and b; or d. a specific binding partner for TACI and BCMA, wherein the specific binding partner has TACI agonist activity, BCMA agonist activity or both.

7. A method of treating B-cell lymphoproliferative disorders, which comprises administering a therapeutic agent comprising an amino acid sequence selected from: Sa. the extracellular region of TACI (SEQ ID NO: b. the extracellular region of BCMA (SEQ ID NO: 6; c. the consensus region of TACI (SEQ ID NO: 16); d. the consensus region of BCMA (SEQ ID NO: 7); e. the TACI/BCMA extracellular consensus sequence (SEQ ID NO: 13). O 74 S8. A method bf treating T-cell lymphoproliferative disorders, which comprises administering a therapeutic agent comprising an amino acid sequence selected from selected from: a. the extracellular region of TACI (SEQ ID NO: C" 5 b. the extracellular region of BCMA (SEQ ID NO: 6; Sc. the consensus region of TACI (SEQ ID NO: 16); C, d. the consensus region of BCMA (SEQ ID NO: 7); e. the TACI/BCMA extracellular consensus sequence (SEQ ID NO: 13)..

9. A method of treating one or more solid tumors, which comprises administering a therapeutic agent comprising an amirib acid sequence selected from selected from a. the extracellular region of TACI (SEQ ID NO: b. the extracellular region of BCMA (SEQ ID NO: 6; c. the consensus region of TACI (SEQ ID NO: 16); d. the consensus region of BCMA (SEQ ID NO: 7); e. the TACI/BCMA extracellular consensus sequence (SEQ ID NO: 13).. The method of Claim 9, wherein the tumor is selected from lung, gastrointestinal, pancreatic and prostate

11. The method of any of Claims 1 to 6, wherein the specific binding partier is an antibody.

12. The method of Claim 11, wherein the antibody is a monoclonal antibody.

13. The method of Claim 11, wherein the antibody is a fully human antibody, a humanized antibody, or an antibody derived from a phage display library.

14. The methods of any of Claims 1 to 6, wherein the specific binding partner is a peptide. 75 The method of Claim 14, wherein the specific binding partner is comprised within a molecule of the formula wherein: F 1 is. a.vehicle; X 1 and X 2 are each independently selected from 1 and (L)P2-(L -P 3 i, P, .and P 4 are each independently peptide sequences, wherein at least one is a specific binding partner; L 2 L 3 and L' are each independently linkers; and a, b, c, and f are each independently 0 or 1, provided that at least one of a and b is 1.

16. The method ofClaim 15, wherein the molecule comprises a structure of the formulae X 1 -F 1 or '-X 2

17. The method of Claim 15, wherein the molecule comprises a structure of the formula F- -P.

18. The method of Claim 15, wherein the molecule comprises a structure of the formula F 1 -14L wherein one of P' and P 2 is a specific binding partner for TACI and the other is a specific binding partner for BCMA.

19. The method of Claim 15, wherein the molecule comprises a structure of the formula F 1 -(L 1 ^),Ps 3 76 wherein'orne of P' and P 2 is a specific binding'partner for APRIL and the other is a specific binding partner for AGP-3. The method of Claim 15 of the formula 2 wherein one of P' and P 2 is a specific binding partner for APRIL and the other is a specific bindiig partner for AGP-3.

21. The method of Claim 15, wherein the vehicle is an Fc domain.

22. The method of Claim 3, wherein the specific binding partner comprises a sequence selected from: a. the extracellular region of TACT (SEQ ID NO: b: the extracellular region of BCMA (SEQ ID NO: 6. c. the consensus region of TACI (SEQ ID NO: 16). d. the consensus region of BCMA (SEQ ID NO: 7). e. the TACI/BCMA extracellular consensus sequence (SEQ ID NO: 13).

23. The method of any of Claims 7,8,9 and 22, wherein the specific binding partner is covalently linked to a vehicle.

24. The inethod of Claim 23, wherein the vehicle is an Fc domain. The method of Claim 3, wherein the specific binding partner is comprised within a molecule having an antibody sequence in which one or more antibody CDR regions' are replaced by one or more sequences selected from: a. the extracellular region of TACI (SEQ ID NO: b. the extracellular region of BCMA(SEQ ID NO: 6); c. the consensus region of TACI (SEQ ID NO: 16); d. the consensus region of BCMA (SEQ ID NO: 7); e. the TACI/BCMA extracellular consensus sequence (SEQ ID NO: 13); 77 f. the sequence of a peptide capable of specifically binding APRIL; and g. the sequence of a peptide capable of specifically binding AGP-3.

26. The method of any of Claims 7,8, and 9, wherein said amino add sequence replaces a CDR region within an antibody molecule.

27. A composition of matter of the formula (X 1 )7F'*(X)b wherein: F'is a vehicle; X' and X 2 are each independently selected from -(L 1 )jP- P and -P 3 P' 2 P 3 and P are each independently peptide sequences, wherein at least one is a specific binding partner for TACI, BCMA, or APRIL; L, L0, L 3 and V' are each independently linkers; and a, b, c, d, e, and f are each independently 0 or 1, provided that at least one of a and b is 1.

28. The composition of matter of Claim 27 of the formulae or F'-X.

29. The composition of matter of Claim 27 of the formula Fl-(Ll)j-V. The composition of matter of Claim 27 of the formula F1(L)jp1(LZ)dP 2 wherein one of P and P is a specific binding partner for TACI and the other is a specific binding partner for BCMA.

31. The composition of matter of Claim 27, wherein the vehicle is an Fc domain. 78

32. An isolated nucleic acid encoding the composition of matter of Claim 31.

33. The nucleic acid of Claim 32 including one or more codons preferred for Escherichia coli expression.

34. An expression vector comprising the nucleic acid of Claim 32. A host cell transformed or transfected with the expression vector of Claim 34.

36. The host cell of Claim 35, wherein the cell is a prokaryotic cell.

37. The host cell of Claim 36, wherein the cell is Escherichia coil. DATED this 6 th day of September 2005 Shelston IP Attorneys for: Amgen Inc.