THEREOF

CROSS-REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit of U.S. Provisional Patent Application number 62/157,104, filed on May 5, 2015, the contents of which are herein incorporated by reference.

STATEMENT OF GOVERNMENT SUPPORT

[0002] This invention was made with government support under grant numbers AI101436, AI109762 and AI090249 awarded by the National Institutes of Health. The government has certain rights in the invention.

BACKGROUND OF THE INVENTION

[0003] Throughout this application various publications are referred to in parentheses. Full citations for these references may be found at the end of the specification. The disclosures of these publications are hereby incorporated by reference in their entirety into the subject application to more fully describe the art to which the subject invention pertains.

[0004] Multiple members of the family Filoviridae of enveloped negative-strand RNA viruses (filoviruses) cause a highly lethal hemorrhagic fever for which no approved treatments are available. While Ebola virus (EBOV) is responsible for the ongoing unprecedented epidemic in West Africa, the antigenically-distinct Sudan virus (SUDV), Bundibugyo virus (BDBV), Marburg virus (MARV), and Ravn virus (RAW) have also caused outbreaks with high case fatality rates (30-90%). Moreover, novel filoviruses with zoonotic potential likely await discovery in the filovirus-endemic zone of equatorial Africa and elsewhere. As a case in point, the founding member of a distinct filovirus clade, Lloviu virus (LLOV), was recently isolated in southern Spain and Portugal, where it is suspected to have caused mass die-offs of insectivorous cave bats (Negredo, PLoS Pathog, 2011).

[0005] The current outbreak in West Africa has seen ZMapp (Mapp Biopharmaceuticals), a mixture of three humanized mouse monoclonal antibodies (mAbs) against the EBOV spike glycoprotein (GP), emerge as a promising treatment for Ebola virus disease. More generally, it has provided a powerful proof-of-concept for mAb-based prophylactics and therapeutics against filoviruses. However, a major limitation of ZMapp and other planned mAb cocktails is their narrow spectrum of action (against EBOV only), dictated by the high antigenic diversity of filovirus GP proteins. Because the development, stockpiling, and deployment of separate mAb cocktails against each virulent filovirus is impractical, broad-spectrum mAb-based treatments that target multiple filoviruses are highly desirable.

[0006] The endo/lysosomal cholesterol transporter Niemann-Pick CI (NPCl) is a universal intracellular receptor for entry and infection by filoviruses, and is required for in vivo pathogenesis by both EBOV and MARV (and almost certainly by BDBV, SUDV, and RAW as well) (Carette, Nature, 2011 ; Ng, Virology, 2014). NPCl protein has a cytoplasmic C-terminus, 13 transmembrane domains, and 3 large loops in the lumen of the endosome (Davies et al, 2000). One complication in targeting NPCl with monoclonal antibodies (mAbs) is its exclusive localization to late endosomal compartments, where it is protected from extracellular antibodies.

[0007] The present invention uses bispecific antibodies (bsAbs) with potent anti- filovirus activity to address the need for methods for treating subjects infected with filoviruses or who are at risk for infection with filoviruses.

SUMMARY OF THE INVENTION

[0008] The invention provides methods of treating or preventing or reducing or inhibiting or reducing the risk or incidence of a filovirus infection in a subject comprising administering to the subject a bispecific antibody or a bispecific fusion construct that binds to Niemann-Pick CI (NPCl) receptor in an amount effective to treat or prevent or reduce or inhibit or reduce the risk or incidence of a filovirus infection in a subject.

[0009] The invention further provides bispecific antibodies and bispecific fusion constructs that bind to Niemann-Pick CI (NPCl) receptor for treating or preventing or reducing or inhibiting or reducing the risk or incidence of a filovirus infection in a subject.

BRIEF DESCRIPTION OF THE DRAWINGS

[0010] Fig. 1A. NPCl is required for Ebola virus infection. Niemann-Pick CI (NPCl), a ubiquitous multi-pass membrane protein localized to late endosomes is required for cytoplasmic entry and infection by all filoviruses. The filovirus spike glycoprotein, GP, must engage NPCl 's second luminal domain (domain C; arrow) to drive viral membrane fusion and cytoplasmic escape. [0011] Fig. IB. NPCl is required for Ebola in vivo pathogenesis. NPCl -knockout mice (NPCl ^"7" ), but not their WT littermates (NPC1 ^+/+ ), are completely resistant to infection and killing by EBOV and MARV.

[0012] Fig. 2A-2D. A illustrates the X-ray crystal structure of a cleaved EBOV GP (GPCL) trimer bound to three copies of a soluble human NPCl domain C protein. In B, side bottom views of NPCl domain C are shown, with residues contacting GPCL (left panel) and mAb 548 (right panel) highlighted. C-D demonstrate the kinetics of binding of 548 to human NPCl domain C by biolayer interferometry (BLI). D shows BLI-derived kinetic binding constants for the interaction of 548 with human and non-human primate (NHP; Colobus guereza) NPCl domain C. 548 recognizes these proteins with picomolar affinity, at both neutral and acid pH.

[0013] Fig 3. mAbs specific to human NPCl domain C block GP-NPC1 binding in vitro. Capacity of three anti-NPCl mAbs to block GP-NPC1 binding in an ELISA. A soluble, flag-tagged form of NPCl domain C was pre-incubated with the indicated concentrations of each mAb, and the protein-mAb mixtures were then added to plates coated with vesicular stomatitis virus (VSV) particles bearing EBOV GP. Bound domain C was detected with an anti-flag antibody. Irrelevant isotype-matched antibodies had no effect on GP-NPC1 domain C binding (not shown).

[0014] Fig. 4. Anti-NPCl domain C mAbs neutralize VSV-EBOV GP infection weakly or not at all. VSV-GP particles were exposed to U20S human osteosarcoma cells in the presence of increasing concentrations of NPCl -specific mAbs or a control GP-specific neutralizing mAb (KZ52). The number of infected cells was determined at 16 h postinfection, and normalized to that obtained in the absence of antibody (set to 100%).

[0015] Fig. 5A. Engineering antibodies with dual binding specificities for GP and NPCl. Constructs encoding bsAbs were generated by fusing NPCl -specific or control IgG sequences to a single-chain variable fragment (scFv) sequence derived from the EBOV GP- specific mAb KZ52. Fusions were done in four configurations: to the N- or C-terminus of the IgG heavy chain [HC] or light chain [LC].

[0016] Fig. 5B. Engineering antibodies with dual binding specificities for GP and NPCl . bsAbs and IgGs were expressed in 293-Freestyle cells by co-transfection of HC and LC expression vectors, and purified by protein A affinity chromatography.

[0017] Fig. 5C. bsAbs, but not NPCl-specific IgGs, can simultaneously bind to GP and NPCl domain C. bsAbs and IgGs were captured onto GP-coated ELISA plates and then incubated with NPC1 domain C-flag. Bound NPC1 domain C was detected with an anti-flag antibody.

[0018] Fig. 6. GP-NPC1 bsAbs neutralize VSV-EBOV GP infection with enhanced potency. Left panel, GP-NPC1 bsAbs show enhanced potency at viral neutralization relative to control GP-only bsAbs. VSV-EBOV GP particles were pre-incubated with the indicated concentrations of each bsAb for 1 h at room temp and then allowed to infect U20S cells. Infected cells were quantified at 16 h post-infection and normalized to a no-antibody control (100%). Right panel, the neutralization curves were fit to a logistic equation to extract the concentration of half-maximal neutralization (IC5 ₀ ± 95% confidence intervals).

[0019] Fig. 7. The enhanced neutralization potency of GP-NPC1 bsAbs can be attributed to their receptor blocking activity. Left panel, VSV-EBOV GP particles were pre-incubated with the indicated concentrations of each bsAb for 1 h at room temp and then allowed to infect matched U20S cell lines expressing endogenous levels of NPC1 or over- expressing NPC1 (NPCl ^hi ). Infected cells were quantified at 16 h post-infection and normalized to a no-antibody control (100%). Right panel, the neutralization curves were fit to a logistic equation to extract the concentration of half-maximal neutralization (IC50 ± 95% confidence intervals). The ratio of IC ₅₀ s for each bsAb in NPCl ^hi vs. WT U20S cells is shown.

[0020] Fig. 8. Fusion of a receptor-blocking IgG to a late endosome-localizing protein facilitates viral neutralization. bsAbs comprising an NPC1 -specific IgG (401 or 548) fused to NPC2, a soluble late endosomal/lysosomal protein, were expressed and purified, and examined for their capacity to neutralize VSV-EBOV GP infection, as described in Figs. 6- 7.

[0021] Fig. 9A-9D. A illustrates a bispecific antibody strategy for delivery of an NPC1- binding antibody to NPCl-positive endosomal/lysosomal compartments. In this strategy, the heavy and light chains of an NPC1 -binding antibody like 548 are genetically fused to the variable VH and VL domains of a delivery antibody like FVM09, to generate a dual- variable domain Ig (DVD-IgTM). FVM09 broadly recognizes ebolavirus GP in extracellular viral particles or on the cell surface. However, the delivery antibody could also target a cell-surface marker, instead of viral GP. In B, the binding of the FVM09-548 DVD-Ig to each of its antigens (ebolavirus GP, human NPC1-C) is measured by BLI. In C, two-phase BLI binding curves show that FVM09-548 can simultaneously bind to both of its antigens. D shows BLI-derived kinetic binding constants for the interaction of the FVM09-548 DVD-Ig with each of its antigens, and compares these constants to those obtained with the repective parent IgGs.

[0022] Fig. 10A-10B. A shows an SDS-polyacrylamide gel of purified FVM09-548 resolved under non-reducing (NR) and reducing (R) conditions and stained with Commassie Brilliant Blue. In B, FVM09-548 was subjected to size-exclusion chromatography coupled to multi-angle light scattering (SEC-MALS). The SEC-MALS trace indicates that the majority species of FVM09-548 (main peak; retention time = 5.10 min) is an IgG-like monomer with an apparent molecular weight of -175 kDa. A small (typical) amount of soluble aggregates (retention team = 4.46 min) is also present.

[0023] Fig. 11A-11C measures the capacity of the FVM09-548 bispecific antibody to neutralize ebolavirus GP-dependent entry and infection in human cells. In A, vesicular stomatitis viruses bearing EBOV GP (VSV-EBOV GP) were incubated with increasing concentrations of FVM09-548, its parent IgGs FVM09, 548, or an equimolar mixture of the parent IgGs (FVM09+548), and then exposed to cells. Only FVM09-548 showed potent, dose-dependent neutralization of VSV-EBOV GP infection. B shows that FVM09-548 can broadly neutralize infection mediated by multiple ebolavirus GPs. C shows that authentic EBOV, BDBV, and SUDV can be potently neutralized by FVM09-548.

[0024] Fig. 12A-12C examines the mechanism of action of the FVM09-548 bispecific antibody. In A, wild-type (WT) VSV-EBOV GP particles, or mutant particles containing two point mutations (E288DAV292R) in the FVM09 epitope ((FVM09mut) were incubated with increasing concentrations of the FVM09-549 antibody, and then exposed to cells. Loss of FVM09-GP binding is associated with a 99% reduction in the potency of antiviral neutralization, providing evidence that engagement of GP in extracellular virus particles by FVM09-548 is necessary for its neutralizing activity. In B, VSV-EBOV GP particles were incubated with FVM09-548 and then exposed to isogenic cell lines expressing physiological levels of human NPCl (WT), or expressing high levels of human NPCl (NPCl-overexp). The neutralization potency of FVM09-548 was abolished by NPCl over- expression, likely because of the titration of available antibody by excess NPCl . This experiment provides evidence that engagement of NPCl domain C by 548 in endosomes is required for the neutralizing activity of FVM09-548. In C, the delivery of FVM09-548 to NPCl -containing endosomes was directly examined. VSV-EBOV GP particles labeled with the fluorescent protein mNeongreen (mNG) were incubated with antibodies labeled with the fluorophore pHrodo-RedTM, which only fluoresces at acid pH (left panel). The virus- antibody mixture was exposed to cells at 37 ^° C for 30 min, after which time, cells were examined for both virus- and antibody-associated fluorescence by two-color flow cytometry. The flow cytometric analysis is shown in the right panel. Cells were gated into virus-negative and virus-positive populations based on mNeongreen fluorescence, and each subpopulation was futher examined for pHrodo-RedTM fluorescence. Strong pHrodo- RedTM fluorescence was seen with FVM09 in the virus-positive sample only, indicating that binding of this antibody to GP on extracellular virus particles affords its delivery to acidic endosomes. Little or no pHrodo-RedTM fluorescence was observed with 548 in the absence or presence of virus, indicating that this antibody cannot access endosomal compartments on its. By contrast FVM09-548 gave a strong pHrodo-RedTM signal in virus-positive cells. Therefore, combining FVM09 and 548 into a single bispecific antibody promotes efficient delivery of 548 to endosomes through the action of FVM09.

[0025] Fig. 13A-13C. The capacity of the human NPCl domain C binding antibody 548 to bind to a non-human primate (Colobus guereza) NPCl domain C protein and block EBOV GP-NPC1 binding is shown. In A, BLI is employed to detect and measure 548 binding to C. guereza NPC domain C at the acidic pH of late endosomes. B shows the capacity of 548 to block GP-C. guereza NPCl domain C binding in an ELISA. A soluble, flag-tagged form of C. guereza NPCl domain C was pre-incubated with the indicated concentrations of 548, and the protein-mAb mixtures were then added to plates coated with vesicular stomatitis virus (VSV) particles bearing EBOV GP. Bound domain C was detected with an anti-flag antibody. Because the sequences of C. guereza NPCl that contact GP are identical to those of NPCl proteins from rhesus macaques (Macaca mulatto) and cynomolgus macaques {Macaca fascicularis) (Fig. 13C), 548 can be evaluated for antiviral protection in these two established non-human primate (NHP) models of filovirus challenge. In C, sequences corresponding to NPCl residues 485-514 (human NPCl numbering) are shown. The 548 epitope is indicated— it is identical in humans and the three indicated NHP species. Homo sapiens - SEQ ID NO:24, Macaca mulatto - SEQ ID NO:25, Colobus guereza - SEQ ID NO:26, Macaca fascicularis - SEQ ID NO:27.

DETAILED DESCRIPTION OF THE INVENTION

[0026] The present invention provides a method for treating or preventing or reducing or inhibiting or reducing the risk or incidence of a filovirus infection in a subject comprising administering to the subject a bispecific antibody or a bispecific fusion construct that binds to Niemann-Pick CI (NPCl) receptor in an amount effective to treat or prevent or reduce or inhibit or reduce the risk or incidence of a filovirus infection in a subject.

[0027] The invention also provides a bispecific antibody or a bispecific fusion construct that binds to Niemann-Pick CI (NPCl) receptor for treating or preventing or reducing or inhibiting or reducing the risk or incidence of a filovirus infection in a subject.

[0028] The bispecific antibody or bispecific fusion construct can combine both antiviral and anti-NPCl specificities in the same molecule. For example, the bispecific antibody or bispecific fusion construct can bind both to NPCl and to filovirus glycoprotein (GP).

[0029] The bispecific antibody can be generated by fusing a NPCl -specific sequence to a single-chain variable fragment (scFv) sequence derived from EBOV GP-specific monoclonal antibody KZ52, where the fusion can be to a N- or C-terminus of an IgG heavy chain or light chain.

[0030] The bispecific antibody or bispecific fusion construct can combine an anti-NPCl specificity with an endosomal targeting functionality. This can include an antibody specific for a cellular molecule that affords endosomal delivery (e.g., NPC2) or a peptide or protein that would do the same (not necessarily just an antibody). Thus, the bispecific antibody or bispecific fusion construct can bind both to NPCl and to Niemann-Pick C2 (NPC2).

[0031] The bispecific antibody or bispecific fusion construct can be generated by fusing heavy and light chains of a NPCl -specific sequence to variable VH and VL domains of a delivery antibody to generate a dual-variable domain Ig. The dual variable domain-Ig (DVD-Ig) bispecific antibody format was developed by AbbVie (Wu et al., 2007). The delivery antibody can target, for example, a filovirus glycoprotein (GP) or a cell-surface marker. The delivery antibody can be, for example, FVM09 (Keck et al. 2015). FVM09 broadly recognizes ebolavirus GP in extracellular viral particles or on the cell surface.

[0032] The anti-NPCl amino acid sequence of a variable region of a light chain of the bispecific antibody or bispecific fusion construct can comprise the amino acid sequence set forth in SEQ ID NO:3, SEQ ID NO:5 or SEQ ID NO:7. The anti-NPCl amino acid sequence of a variable region of a heavy chain of the bispecific antibody or bispecific fusion construct can comprise the amino acid sequence set forth in SEQ ID NO:4, SEQ ID NO:6 or SEQ ID NO: 8. The anti-NPCl amino acid sequence of a variable region of a light chain can comprise the amino acid sequence set forth in SEQ ID NO:3 and a variable region of a heavy chain can comprise the amino acid sequence set forth in SEQ ID NO:4. The anti- NPCl amino acid sequence of a variable region of a light chain can comprise the amino acid sequence set forth in SEQ ID NO:5 and a variable region of a heavy chain can comprise the amino acid sequence set forth in SEQ ID NO:6. The anti-NPCl amino acid sequence of a variable region of a light chain can comprise the amino acid sequence set forth in SEQ ID NO: 7 and a variable region of a heavy chain can comprise the amino acid sequence set forth in SEQ ID NO: 8.

[0033] The bispecific antibody can comprise the amino acid sequence set forth in any one of SEQ ID NOs:9-20. The bispecific fusion construct can comprise the amino acid sequence set forth in SEQ ID NO:21 or SEQ ID NO:22.

[0034] In different uses, the subject can be infected with a filovirus. Alternatively, the subject can be at risk for infection with a filovirus. Subjects who are at risk for infection with filoviruses include subjects who have been exposed to filovirus or are at risk of exposure to filovirus. In addition to the natural occurrence of filoviruses, there is the potential for exposure to these pathogens if they are used as agents of bioterrorism or biological warfare. For example, the subject can be a family member or healthcare worker in an area of an outbreak of a filovirus infection. The subject can be a medical personnel, first responder or military personnel potentially exposed or exposed to a filovirus as the result of bioterrorism or biological warfare. The subject can be a biosafety level 3/4 laboratory personnel or animal worker potentially exposed or exposed to a filovirus.

[0035] The family Filoviridae is a family of viruses including genera Ebolavirus and Marburgvirus. The Ebola virus species can be Zaire ebolavirus or Sudan ebolavirus. Additional examples of filovirus include a Bundibugyo virus, a Sudan virus, a Ravn virus and a Lloviu virus.

[0036] The subject can be a mammal, such as, for example, livestock, a bat, a primate or a human.

[0037] To treat a subject with a filovirus infection means to reduce or stop the spread of filovirus in the subject, or to eliminate the filovirus from the subject, or to reduce or eliminate a sign or symptom of filovirus infection in the subject. Filovirus infection is characterized by hemorrhagic fever, including abnormalities in blood coagulation. As used herein, "preventing" a filovirus infection means reducing the development of, or reducing the extent of, one or more symptoms of the condition, as compared to the development or extent the condition takes in the absence of preventative treatment. In an embodiment, "preventing" as used herein does not mean an absolute prevention, but a lessened extent of the condition brought about prophylactically, or to reduce the risk or incidence of a filovirus infection.

[0038] Human NPC1 receptor protein has the amino acid sequence (SEQ ID NO: l)

(NCBI Reference Sequence: NM_000271.4):

MTARGLALGL LLLLLCPAQV FSQSCVWYGE CGIAYGDKRY NCEYSGPPKP LPKDGYDLVQ 60 ELCPGFFFGN VSLCCDVRQL QTLKDNLQLP LQFLSRCPSC FYNLLNLFCE LTCSPRQSQF 120 LNVTATEDYV DPVTNQTKTN VKELQYYVGQ SFANAMYNAC RDVEAPSSND KALGLLCGKD 180 ADACNATNWI EYMFNKDNGQ APFTITPVFS DFPVHGMEPM NNATKGCDES VDEVTAPCSC 240 QDCSIVCGPK PQPPPPPAPW TILGLDAMYV IMWITYMAFL LVFFGAFFAV WCYRKRYFVS 300 EYTPIDSNIA FSV ASDKGE ASCCDPVSAA FEGCLRRLFT RWGSFCVRNP GCVIFFSLVF 360 ITACSSGLVF VRVTTNPVDL WSAPSSQARL EKEYFDQHFG PFFRTEQLI I RAPLTDKHIY 420 QPYPSGADVP FGPPLDIQIL HQVLDLQIAI ENITASYDNE TVTLQDICLA PLSPYNTNCT 480 ILSVLNYFQN SHSVLDHKKG DDFFVYADYH THFLYCVRAP ASLNDTSLLH DPCLGTFGGP 540 VFPWLVLGGY DDQNYNNATA LVITFPV NY YNDTEKLQRA QAWEKEFINF VKNYKNPNLT 600 ISFTAERSIE DELNRESDSD VFTWISYAI MFLYI SLALG HMKSCRRLLV DSKVSLGIAG 660 ILIVLSSVAC SLGVFSYIGL PLTLIVIEVI PFLVLAVGVD NIFILVQAYQ RDERLQGETL 720 DQQLGRVLGE VAPSMFLSSF SETVAFFLGA LSVMPAVHTF SLFAGIAVFI DFLLQITCFV 780 SLLGLDIKRQ EKNRLDIFCC VRGAEDGTSV QASESCLFRF FKNSYSPLLL KDWMRPIVIA 840 IFVGVLSFSI AVLNKVDIGL DQSLSMPDDS YMVDYFKSIS QYLHAGPPVY FVLEEGHDYT 900 SSKGQNMVCG GMGCNNDSLV QQIFNAAQLD NYTRIGFAPS SWIDDYFDWV KPQSSCCRVD 960 NITDQFCNAS WDPACVRCR PLTPEGKQRP QGGDFMRFLP MFLSDNPNPK CGKGGHAAYS 1020 SAVNILLGHG TRVGATYFMT YHTVLQTSAD FIDALKKARL IASNVTETMG INGSAYRVFP 1080 YSVFYVFYEQ YLTIIDDTIF NLGVSLGAIF LVTMVLLGCE LWSAVIMCAT IAMVLVNMFG 1140 VMWLWGISLN AVSLV LVMS CGISVEFCSH ITRAFTVSMK GSRVERAEEA IAHMGSSVFS 1200 GITLTKFGGI WLAFAKSQI FQIFYFRMYL AMVLLGATHG LIFLPVLLSY IGPSV KAKS 1260 CATEERYKGT ERERLLNF 1278

[0039] Nucleic acid (mRNA) encoding human NPC 1 receptor protein has the nucleotide sequence (SEQ ID NO:2) (NCBI Reference Sequence: NM_000271.4):

1 gaagggcaac acggggacct tgaagcgggg tcgcggcggc gccccagccc gggccaggga 61 gtcccggcag cggcacctcc cagaaagggc ggagccgacg acgccttctt ccttcctgac 121 cggcgcgcgc agcctgctgc cgcggtcagc gcctgctcct gctcctccgc tcctcctgcg 181 cggggtgctg aaacagcccg gggaagtaga gccgcctccg gggagcccaa ccagccgaac 241 gccgccggcg tcagcagcct tgcgcggcca cagcatgacc gctcgcggcc tggcccttgg 301 cctcctcctg ctgctactgt gtccagcgca ggtgttttca cagtcctgtg tttggtatgg 361 agagtgtgga attgcatatg gggacaagag gtacaattgc gaatattctg gcccaccaaa 421 accattgcca aaggatggat atgacttagt gcaggaactc tgtccaggat tcttctttgg 481 caatgtcagt ctctgttgtg atgttcggca gcttcagaca ctaaaagaca acctgcagct 541 gcctctacag tttctgtcca gatgtccatc ctgtttttat aacctactga acctgttttg 601 tgagctgaca tgtagccctc gacagagtca gtttttgaat gttacagcta ctgaagatta 661 tgttgatcct gttacaaacc agacgaaaac aaatgtgaaa gagttacaat actacgtcgg 721 acagagtttt gccaatgcaa tgtacaatgc ctgccgggat gtggaggccc cctcaagtaa 781 tgacaaggcc ctgggactcc tgtgtgggaa ggacgctgac gcctgtaatg ccaccaactg 841 gattgaatac atgttcaata aggacaatgg acaggcacct tttaccatca ctcctgtgtt 901 ttcagatttt ccagtccatg ggatggagcc catgaacaat gccaccaaag gctgtgacga 961 gtctgtggat gaggtcacag caccatgtag ctgccaagac tgctctattg tctgtggccc 1021 caagccccag cccccacctc ctcctgctcc ctggacgatc cttggcttgg acgccatgta 1081 tgtcatcatg tggatcacct acatggcgtt tttgcttgtg ttttttggag cattttttgc 1141 agtgtggtgc tacagaaaac ggtattttgt ctccgagtac actcccatcg atagcaatat 1201 agctttttct gttaatgcaa gtgacaaagg agaggcgtcc tgctgtgacc ctgtcagcgc 1261 agcatttgag ggctgcttga ggcggctgtt cacacgctgg gggtctttct gcgtccgaaa 1321 ccctggctgt gtcattttct tctcgctggt cttcattact gcgtgttcgt caggcctggt 1381 gtttgtccgg gtcacaacca atccagttga cctctggtca gcccccagca gccaggctcg 1441 cctggaaaaa gagtactttg accagcactt tgggcctttc ttccggacgg agcagctcat 1501 catccgggcc cctctcactg acaaacacat ttaccagcca tacccttcgg gagctgatgt 1561 accctttgga cctccgcttg acatacagat actgcaccag gttcttgact tacaaatagc 1621 catcgaaaac attactgcct cttatgacaa tgagactgtg acacttcaag acatctgctt 1681 ggcccctctt tcaccgtata acacgaactg caccattttg agtgtgttaa attacttcca 1741 gaacagccat tccgtgctgg accacaagaa aggggacgac ttctttgtgt atgccgatta 1801 ccacacgcac tttctgtact gcgtacgggc tcctgcctct ctgaatgata caagtttgct 1861 ccatgaccct tgtctgggta cgtttggtgg accagtgttc ccgtggcttg tgttgggagg 1921 ctatgatgat caaaactaca ataacgccac tgcccttgtg attaccttcc ctgtcaataa 1981 ttactataat gatacagaga agctccagag ggcccaggcc tgggaaaaag agtttattaa 2041 ttttgtgaaa aactacaaga atcccaatct gaccatttcc ttcactgctg aacgaagtat 2101 tgaagatgaa ctaaatcgtg aaagtgacag tgatgtcttc accgttgtaa ttagctatgc 2161 catcatgttt ctatatattt ccctagcctt ggggcacatg aaaagctgtc gcaggcttct 2221 ggtggattcg aaggtctcac taggcatcgc gggcatcttg atcgtgctga gctcggtggc 2281 ttgctccttg ggtgtcttca gctacattgg gttgcccttg accctcattg tgattgaagt 2341 catcccgttc ctggtgctgg ctgttggagt ggacaacatc ttcattctgg tgcaggccta 2401 ccagagagat gaacgtcttc aaggggaaac cctggatcag cagctgggca gggtcctagg 2461 agaagtggct cccagtatgt tcctgtcatc cttttctgag actgtagcat ttttcttagg 2521 agcattgtcc gtgatgccag ccgtgcacac cttctctctc tttgcgggat tggcagtctt 2581 cattgacttt cttctgcaga ttacctgttt cgtgagtctc ttggggttag acattaaacg 2641 tcaagagaaa aatcggctag acatcttttg ctgtgtcaga ggtgctgaag atggaacaag 2701 cgtccaggcc tcagagagct gtttgtttcg cttcttcaaa aactcctatt ctccacttct 2761 gctaaaggac tggatgagac caattgtgat agcaatattt gtgggtgttc tgtcattcag 2821 catcgcagtc ctgaacaaag tagatattgg attggatcag tctctttcga tgccagatga 2881 ctcctacatg gtggattatt tcaaatccat cagtcagtac ctgcatgcgg gtccgcctgt 2941 gtactttgtc ctggaggaag ggcacgacta cacttcttcc aaggggcaga acatggtgtg 3001 cggcggcatg ggctgcaaca atgattccct ggtgcagcag atatttaacg cggcgcagct 3061 ggacaactat acccgaatag gcttcgcccc ctcgtcctgg atcgacgatt atttcgactg 3121 ggtgaagcca cagtcgtctt gctgtcgagt ggacaatatc actgaccagt tctgcaatgc 3181 ttcagtggtt gaccctgcct gcgttcgctg caggcctctg actccggaag gcaaacagag 3241 gcctcagggg ggagacttca tgagattcct gcccatgttc ctttcggata accctaaccc 3301 caagtgtggc aaagggggac atgctgccta tagttctgca gttaacatcc tccttggcca 3361 tggcaccagg gtcggagcca cgtacttcat gacctaccac accgtgctgc agacctctgc 3421 tgactttatt gacgctctga agaaagcccg acttatagcc agtaatgtca ccgaaaccat 3481 gggcattaac ggcagtgcct accgagtatt tccttacagt gtgttttatg tcttctacga 3541 acagtacctg accatcattg acgacactat cttcaacctc ggtgtgtccc tgggcgcgat 3601 atttctggtg accatggtcc tcctgggctg tgagctctgg tctgcagtca tcatgtgtgc 3661 caccatcgcc atggtcttgg tcaacatgtt tggagttatg tggctctggg gcatcagtct 3721 gaacgctgta tccttggtca acctggtgat gagctgtggc atctccgtgg agttctgcag 3781 ccacataacc agagcgttca cggtgagcat gaaaggcagc cgcgtggagc gcgcggaaga 3841 ggcacttgcc cacatgggca gctccgtgtt cagtggaatc acacttacaa aatttggagg 3901 gattgtggtg ttggcttttg ccaaatctca aattttccag atattctact tcaggatgta 3961 tttggccatg gtcttactgg gagccactca cggattaata tttctccctg tcttactcag 4021 ttacataggg ccatcagtaa ataaagccaa aagttgtgcc actgaagagc gatacaaagg 4081 aacagagcgc gaacggcttc taaatttcta gccctctcgc agggcatcct gactgaactg 4141 tgtctaaggg tcggtcggtt taccactgga cgggtgctgc atcggcaagg ccaagttgaa 4201 caccggatgg tgccaaccat cggttgtttg gcagcagctt tgaacgtagc gcctgtgaac 4261 tcaggaatgc acagttgact tgggaagcag tattactaga tctggaggca accacaggac 4321 actaaacttc tcccagcctc ttcaggaaag aaacctcatt ctttggcaag caggaggtga 4381 cactagatgg ctgtgaatgt gatccgctca ctgacactct gtaaaggcca atcaatgcac 4441 tgtctgtctc tccttttagg agtaagccat cccacaagtt ctataccata tttttagtga 4501 cagttgaggt tgtagataca ctttataaca ttttatagtt taaagagctt tattaatgca 4561 ataaattaac tttgtacaca tttttatata aaaaaacagc aagtgatttc agaatgttgt 4621 aggcctcatt agagcttggt ctccaaaaat ctgtttgaaa aaagcaacat gttcttcaca 4681 gtgttcccct agaaaggaag agatttaatt gccagttaga tgtggcatga aatgagggac 4741 aaagaaagca tctcgtaggt gtgtctactg ggttttaact tatttttctt taataaaata 4801 cattgttttc ctaaaaaaaa aaaaaaa

[0040] The invention also provides monoclonal antibodies to NPC1. [0041] Examples of anti-NPCl antibody amino acid sequences of variable regions for light and heavy chains include:

401 -variable light chain (SEQ ID NO:3) (mouse)

DIQMTQS PASLSASVGETV ITCRASENIYSYLAWYQQKQGKS PQLLVYNAK LVEAVPSR FSGSGSGTQFSLKINSLQPEDFGTYYCQHHYGSPWTFGGGTKLEIK,

401 -variable heavy chain (SEQ ID NO:4) (mouse)

EVQLQQSGAELVKPGASVKLSCTASGFNIKDTYMHWVKQRPEQGLEWIGRIDPANGNTEY D TKFQGKATITADTSSNTAYLQLSSLTSEDTAVYYCSRGYYWGRGTTLTVSS,

548-variable light chain (SEQ ID NO:5) (mouse)

DIQMTQS PASLSASVGETVTITCRASENIYSYLAWYQQKQGKS PQLLVYNAK LAEGVPSR FSGSGSGTQFSLKINSLQPEDFGIYYCQHHYGSPWAFGGGTKLEIK,

548-variable heavy chain (SEQ ID NO:6) (mouse)

EVQLQQSGAELVKPGASVKLSCTASGFNIKDTYMHWVKQRPEEGLEWIGRIDPADGNTEY V PKFQGKATITADTFSNTVYLQLSGLTSEDTAVYYCSRGYYWGQGTTLTVSS,

952-variable light chain (SEQ ID NO: 7) (mouse)

DIVMTQSHKFMSTSVGDRVS ITCKASQDVNTAWWYQQKPGQS PKLLIYWASTRHTGVPDR FTGSGSGTDFTLTISSVQAEDLALYYCQQHYTSPWTFGGGTKLEIK, and

952-variable heavy chain (SEQ ID NO: 8) (mouse)

DVQLQESGPDLVKPXQSLSLTCTVTGYSITSGYSWHWIRQFPGNRLEWMDYIHYSGSINY N PSLKSRISITRDTSKNQFFLQLNSVTTEDTATYYCARWGATGFDYWGQGTTLTVSS .

[0042] Examples of bispecific Ab design include the following. The underlined region designates glycine-rich linker polypeptide. The bold region designates fusion linker polypeptide. The italicized region corresponds to the constant sequences of the pMAZ- encoded heavy or light constant region sequences for human IgGl (heavy) and kappa constant domain (light).

401 -scFv-KZ52-HCN (heavy chain N-terminal fusion) (in all of the following scFv-KZ52 bispecifics: scFv is human, IgG HC and LC constant domains are human IgGl, IgG VH and VL domains are mouse) (SEQ ID NO: 9)

EVQLLESGGGLVKPGGSLRLSCAASGFTLINYRMNWVRQAPGKGLEWVSSISSSSSYIHY A DSVKGRFTISRDNAENSLYLQMNSLRAEDTAVYYCVREGPRATGYSMADVFDIWGQGTMV T VSSGGGGSGGGGSGGGGSELVMTQSPDSLAVSLGERATINCKSSQSVLYSSNNKSYLAWY Q QKPGQPPKLLIYWASTRESGVPDRFSGSGSGTDF L ISSLQAEDVAVYYCQQYYSAPL F GGGTKVEIKGGSAGSAGSAGSGGSEVQLQQSGAELVKPGASVKLSCTASGFNIKDTYMHW V

KQRPEQGLEWIGRIDPANGNTEYDTKFQGKATITADTSSNTAYLQLSSLTSEDTAVY YCSR GYYWGRGTTLTVSSASTKGPSVFPLAPSSKSTSGGTAALGCLVKDYFPEPVTVSWNSGAL T SGVHTFPAVLQSSGLYSLSSWTVPSSSLGTQTYICNVNHKPSNTKVDKKVEPKSCDKTHT CPPCPAPELLGRPSVFLFPPKPKDTLMISRTPEVTCVWDVSHEDPEVKFNWYVDGVEVHN AKTKPREEQYNSTYRVVSVLTVLHQDWLNGKEYKCKVSNKALPAPIEKTISKAKGQPREP Q VYTLPPSREEMTKNQVSLTCLVKGFYPSDIAVEWESNGQPENNYKTTPPVLDSDGSFFLY S KLTVDKSRWQQGNVFSCSVMHEALHNHYTQKSLSLSPGK,

401-scFv-KZ52-LCN (light chain N-terminal fusion) (SEQ ID NO: 10)

EVQLLESGGGLVKPGGSLRLSCAASGFTLINYRMNWVRQAPGKGLEWVSSISSSSSYIHY A DSVKGRFTISRDNAENSLYLQMNSLRAEDTAVYYCVREGPRATGYSMADVFDIWGQGTMV T VSSGGGGSGGGGSGGGGSELVMTQSPDSLAVSLGERATINCKSSQSVLYSSNNKSYLAWY Q QKPGQPPKLLIYWASTRESGVPDRFSGSGSGTDF L ISSLQAEDVAVYYCQQYYSAPL F GGGTKVEIKGGSAGSAGSAGSGGSDIQMTQSPASLSASVGETVTITCRASENIYSYLAWY Q QKQGKSPQLLVYNAKTLVEAVPSRFSGSGSGTQFSLKINSLQPEDFGTYYCQHHYGSPWT F GGGTKLEIKRTVAAPSVFIFPPSDEQLKSGTASVVCLLNNFYPREAKVQWKVDNALQSGN S QESVTEQDSKDSTYSLSSTLTLSKADYEKHKVYACEVTHQGLSSPVTKSFNRGEC _r

401-scFv-KZ52-HCC (heavy chain C-terminal fusion) (SEQ ID NO: 11)

EVQLQQSGAELVKPGASVKLSCTASGFNIKDTYMHWVKQRPEQGLEWIGRIDPANGNTEY D TKFQGKATITADTSSNTAYLQLSSLTSEDTAVYYCSRGYYWGRGTTLTVSSASTKGPSVF P LAPSSKSTSGGTAALGCLVKDYFPEPVTVSWNSGALTSGVHTFPAVLQSSGLYSLSSVVT V PSSSLGTQTYICNVNHKPSNTKVDKKVEPKSCDKTHTCPPCPAPELLGRPSVFLFPPKPK D TLMISRTPEVTCVWDVSHEDPEVKFNWYVDGVEVHNAKTKPREEQYNSTYRVVSVLTVLH QDWLNGKEYKCKVSNKALPAPIEKTISKAKGQPREPQVYTLPPSREEMTKNQVSLTCLVK G FYPSDIAVEWESNGQPENNYKTTPPVLDSDGSFFLYSKLTVDKSRWQQGNVFSCSVMHEA L

iiiViiYTQ SiSiSPG GGSAGSAGSAGSGGSEVQLLESGGGLVKPGGSLRLSCAASGFTLIN YRMNWVRQAPGKGLEWVSSISSSSSYIHYADSVKGRFTISRDNAENSLYLQMNSLRAEDT A VYYCVREGPRATGYSMADVFDIWGQGTMVTVSSGGGGSGGGGSGGGGSELVMTQSPDSLA V SLGERATINCKSSQSVLYSSNNKSYLAWYQQKPGQPPKLLIYWASTRESGVPDRFSGSGS G TDFTLTISSLQAEDVAVYYCQQYYSAPLTFGGGTKVEIK,

401-scFv-KZ52-LCC (light chain C-terminal fusion) (SEQ ID NO: 12) DIQMTQS PASLSASVGETV ITCRASENIYSYLAWYQQKQGKS PQLLVYNAK LVEAVPSR FSGSGSGTQFSLKINSLQPEDFGTYYCQHHYGS PWTFGGGTKLEIK. TVAAPSVFIFPPSD EQLKSGTASWCLLNNFYPREAKVQWKVDNALQSGNSQESVTEQDSKDSTYSLSSTLTLSK AD YEKHKVYA CE VTHQ GL SSP VTKS FNRGE CGG SAG SAG SAG SGGSEVQLLESGGGLVKPG GSLRLSCAASGFTLINYRMNWVRQAPGKGLEWVSSISSSSSYIHYADSVKGRFTISRDNA E NSLYLQMNSLRAEDTAVYYCVREGPRATGYSMADVFDIWGQGTMVTVSSGGGGSGGGGSG G GGSELVMTQSPDSLAVSLGERATINCKSSQSVLYSSNNKSYLAWYQQKPGQPPKLLIYWA S TRESGVPDRFSGSGSGTDFTLTISSLQAEDVAVYYCQQYYSAPLTFGGGTKVEIK,

548-scFv-KZ52-HCN (heavy chain N-terminal fusion) (SEQ ID NO:13)

EVQLLESGGGLVKPGGSLRLSCAASGFTLINYRMNWVRQAPGKGLEWVSSISSSSSYIHY A DSVKGRFTISRDNAENSLYLQMNSLRAEDTAVYYCVREGPRATGYSMADVFDIWGQGTMV T VSSGGGGSGGGGSGGGGSELVMTQSPDSLAVSLGERATINCKSSQSVLYSSNNKSYLAWY Q QKPGQPPKLLI YWASTRESGVPDRFSGSGSGTDF L ISSLQAEDVAVYYCQQYYSAPL F GGGTKVEIKGGSAGSAGSAGSGGSEVQLQQSGAELVKPGASVKLSCTASGFNIKDTYMHW V KQRPEEGLEWIGRIDPADGNTEYVPKFQGKATITADTFSNTVYLQLSGLTSEDTAVYYCS R GYYWGQGTTLTVSS AS TKGPSVFPLAPSSKSTSGGTAALGCLVKDYFPEPVTVSWNSGALT SGVHTFPAVLQSSGLYSLSSWTVPSSSLGTQTYICNVNHKPSNTKVDKKVEPKSCDKTHT CPPCPAPELLGRPSVFLFPPKPKDTLMISR TPEVTCVWDVSHEDPEVKFNWYVDGVEVHN AKTKPREEQYNS TYRVVS VL TVLHQD WLNGKE YKCKVSNKAL PAPIEKTISKAKGQPREPQ VYTLPPSREEMTKNQVSLTCLVKGFYPSDIAVEWESNGQPENNYKTTPPVLDSDGSFFLY S KLTVDKSRWQQGNVFSCSVMHEALHNHYTQKSLSLSPGK,

548-scFv-KZ52-LCN (light chain N-terminal fusion) (SEQ ID NO: 14)

EVQLLESGGGLVKPGGSLRLSCAASGFTLINYRMNWVRQAPGKGLEWVSSISSSSSYIHY A DSVKGRFTISRDNAENSLYLQMNSLRAEDTAVYYCVREGPRATGYSMADVFDIWGQGTMV T VSSGGGGSGGGGSGGGGSELVMTQSPDSLAVSLGERATINCKSSQSVLYSSNNKSYLAWY Q QKPGQPPKLLI YWASTRESGVPDRFS GSGSGTD F L IS SLQAEDVAVYYCQQ YY SAP L F GGGTKVEIKGGSAGSAGSAGSGGSDIQMTQSPASLSASVGETVTITCRASENIYSYLAWY Q QKQGKS PQLLVYNAKT LAEGVPSRFSGSGSGTQFSLKINSLQPEDFGIYYCQHHYGSPWAF GGGTKLEIKRTVAAPSVFIFPPSDEQLKSGTASVVCLLNNFYPREAKVQWKVDNALQSGN S QE S VTEQDSKD STYSLSSTLTL SKAD YEKHKVYA CE VTHQGL S SPVTKS FNRGE C _r

548-scFv-KZ52-HCC (heavy chain C-terminal fusion) (SEQ ID NO: 15)

EVQLQQSGAELVKPGASVKLSCTASGFNIKDTYMHWVKQRPEEGLEWIGRIDPADGNTEY V PKFQGKATITADTFSNTVYLQLSGLTSEDTAVYYCSRGYYWGQGTTLTVSSASTKGPSVF P LAPSSKSTSGGTAALGCLVKDYFPEPVTVSWNSGALTSGVHTFPAVLQSSGLYSLSSVVT V PSSSLGTQTYICNVNHKPSNTKVDKKVEPKSCDKTHTCPPCPAPELLGRPSVFLFPPKPK D TLMISRTPEVTCVWDVSHEDPEVKFNWYVDGVEVHNAKTKPREEQYNSTYRVVSVLTVLH QDWLNGKEYKCKVSNKALPAPIEKTISKAKGQPREPQVYTLPPSREEMTKNQVSLTCLVK G FYPSDIAVEWESNGQPENNYKTTPPVLDSDGSFFLYSKLTVDKSRWQQGNVFSCSVMHEA L

iiiViiYTQ SiSiSPG GGSAGSAGSAGSGGSEVQLLESGGGLVKPGGSLRLSCAASGFTLIN YRMNWVRQAPGKGLEWVSSISSSSSYIHYADSVKGRFTISRDNAENSLYLQMNSLRAEDT A VYYCVREGPRATGYSMADVFDIWGQGTMVTVSSGGGGSGGGGSGGGGSELVMTQSPDSLA V SLGERATINCKSSQSVLYSSNNKSYLAWYQQKPGQPPKLLIYWASTRESGVPDRFSGSGS G TDFTLTISSLQAEDVAVYYCQQYYSAPLTFGGGTKVEIK,

548-scFv-KZ52-LCC (light chain C-terminal fusion) (SEQ ID NO: 16)

DIQMTQS PASLSASVGETV ITCRASENIYSYLAWYQQKQGKS PQLLVYNAK LAEGVPSR FSGSGSGTQFSLKINSLQPEDFGIYYCQHHYGS PWAFGGGTKLEIK. TVAAPSVFIFPPSD EQLKSGTASWCLLNNFYPREAKVQWKVDNALQSGNSQESVTEQDSKDSTYSLSSTLTLSK ADYEKHKVYACEVTHQGLSSPVTKSFNRGECGGSAGSAGSAGSGGSEVQLLESGGGLVKP G GSLRLSCAASGFTLINYRMNWVRQAPGKGLEWVSSISSSSSYIHYADSVKGRFTISRDNA E NSLYLQMNSLRAEDTAVYYCVREGPRATGYSMADVFDIWGQGTMVTVSSGGGGSGGGGSG G GGSELVMTQSPDSLAVSLGERATINCKSSQSVLYSSNNKSYLAWYQQKPGQPPKLLIYWA S TRESGVPDRFSGSGSGTDFTLTISSLQAEDVAVYYCQQYYSAPLTFGGGTKVEIK,

952-scFv-KZ52-HCN (heavy chain N-terminal fusion) (SEQ ID NO:17)

EVQLLESGGGLVKPGGSLRLSCAASGFTLINYRMNWVRQAPGKGLEWVSSISSSSSYIHY A DSVKGRFTISRDNAENSLYLQMNSLRAEDTAVYYCVREGPRATGYSMADVFDIWGQGTMV T VSSGGGGSGGGGSGGGGSELVMTQSPDSLAVSLGERATINCKSSQSVLYSSNNKSYLAWY Q QKPGQPPKLLIYWASTRESGVPDRFSGSGSGTDF L ISSLQAEDVAVYYCQQYYSAPL F GGGTKVEIKGGSAGSAGSAGSGGSDVQLQESGPDLVKPXQSLSLTCTVTGYS I SGYSWHW IRQFPGNRLEWMDYIHYSGS INYNPSLKSRISI RD SKNQFFLQLNSVTTEDTATYYCAR WGATGFDYWGQGTTLTVSSASTKGPSVFPLAPSSKSTSGGTAALGCLVKDYFPEPVTVSW N SGALTSGVHTFPAVLQSSGLYSLSSVVTVPSSSLGTQTYICNVNHKPSNTKVDKKVEPKS C DKTHTCPPCPAPELLGRPSVFLFPPKPKDTLMISRTPEVTCWVDVSHEDPEVKFNWYVDG VEVHNAKTKPREEQYNSTYRWSVLTVLHQDWLNGKEYKCKVSNKALPAPIEKTISKAKGQ PREPQVYTLPPSREEMTKNQVSLTCLVKGFYPSDIAVEWESNGQPENNYKTTPPVLDSDG S FFLYSKLTVDKSRWQQGNVFSCSVMHEALHNHYTQKSLSLSPGK, 952-scFv-KZ52-LCN (light chain N-terminal fusion) (SEQ ID NO: 18)

EVQLLESGGGLVKPGGSLRLSCAASGFTLINYRMNWVRQAPGKGLEWVSSISSSSSY IHYA DSVKGRFTISRDNAENSLYLQMNSLRAEDTAVYYCVREGPRATGYSMADVFDIWGQGTMV T VSSGGGGSGGGGSGGGGSELVMTQSPDSLAVSLGERATINCKSSQSVLYSSNNKSYLAWY Q QKPGQPPKLLIYWASTRESGVPDRFSGSGSGTDF L ISSLQAEDVAVYYCQQYYSAPL F GGGTKVEIKGGSAGSAGSAGSGGSDIVMTQSHKFMSTSVGDRVSITCKASQDVNTAWWYQ QKPGQSPKLLIYWAS RHTGVPDRF GSGSG DFTL ISSVQAEDLALYYCQQHY SPW F GGGTKLEIKRTVAAPSVFIFPPSDEQLKSGTASVVCLLNNFYPREAKVQWKVDNALQSGN S QESVTEQDSKDSTYSLSSTLTLSKADYEKHKVYACEVTHQGLSSPVTKSFNRGEC _r

952-scFv-KZ52-HCC (heavy chain C-terminal fusion) (SEQ ID NO: 19)

DVQLQESGPDLVKPXQSLSLTCTVTGYSITSGYSWHWIRQFPGNRLEWMDYIHYSGSINY N PSLKSRISITRDTSKNQFFLQLNSVTTEDTATYYCARWGATGFDYWGQGTTLTVSSASTK G PSVFPLAPSSKSTSGGTAALGCLVKDYFPEPVTVSWNSGALTSGVHTFPAVLQSSGLYSL S SVVTVPSSSLGTQTYICNVNHKPSNTKVDKKVEPKSCDKTHTCPPCPAPELLGRPSVFLF P PKPKDTLMISRTPEVTCVWDVSHEDPEVKFNWYVDGVEVHNAKTKPREEQYNSTYRVVSV LTVLHQDWLNGKEYKCKVSNKALPAPIEKTISKAKGQPREPQVYTLPPSREEMTKNQVSL T CLVKGFYPSDIAVEWESNGQPENNYKTTPPVLDSDGSFFLYSKLTVDKSRWQQGNVFSCS V MHEALHNHYTQKSLSLSPGKGGSAGSAGSAGSGGSEVQLLESGGGLVKPGGSLRLSCAAS G FTLINYRMNWVRQAPGKGLEWVSSISSSSSYIHYADSVKGRF ISRDNAENSLYLQMNSLR AEDTAVYYCVREGPRATGYSMADVFDIWGQGTMVTVSSGGGGSGGGGSGGGGSELVMTQS P DSLAVSLGERATINCKSSQSVLYSSNNKSYLAWYQQKPGQPPKLLIYWASTRESGVPDRF S GSGSGTDFTLTISSLQAEDVAVYYCQQYYSAPLTFGGGTKVEIK, and

952-scFv-KZ52-LCC (light chain C-terminal fusion) (SEQ ID NO:20)

DIVMTQSHKFMSTSVGDRVS ITCKASQDVNTAWWYQQKPGQS PKLLIYWASTRHTGVPDR

FTGSGSGTDFTLTISSVQAEDLALYYCQQHYTSPWTFGGGTKLEIK. TVAAPSVFIFPPSD

EQLKSGTASWCLLNNFYPREAKVQWKVDNALQSGNSQESVTEQDSKDSTYSLSSTLT LSK

ADYEKHKVYACEVTHQGLSSPVTKSFNRGECGGSAGSAGSAGSGGSEVQLLESGGGL VKPG

GSLRLSCAASGFTLINYRMNWVRQAPGKGLEWVSSISSSSSYIHYADSVKGRFTISR DNAE

NSLYLQMNSLRAEDTAVYYCVREGPRATGYSMADVFDIWGQGTMVTVSSGGGGSGGG GSGG

GGSELVMTQSPDSLAVSLGERATINCKSSQSVLYSSNNKSYLAWYQQKPGQPPKLLI YWAS

TRESGVPDRFSGSGSGTDFTLTISSLQAEDVAVYYCQQYYSAPLTFGGGTKVEIK.

[0043] Additional examples incude the following. The italicized region corresponds to the constant sequences of the pMAZ-encoded heavy or light constant region sequences for human IgGl. The underlined region designates fusion linker polypeptide. The bold region designates the sequence of human NPC2.

401-NPC2-HCC (NPC2 C-terminal fusion to heavy chain of 401 IgGl) (in all of the following bispecifics: NPC2 is human, IgG HC and LC constant domains are human IgGl, IgG VH and VL domains are mouse) (SEQ ID NO:21)

EVQLQQSGAELVKPGASVKLSCTASGFNIKDTYMHWVKQRPEQGLEWIGRIDPANGNTEY D TKFQGKATITADTSSNTAYLQLSSLTSEDTAVYYCSRGYYWGRGTTLTVSS ASTKGPSVFP LAPSSKS TSGGTAALGCLVKDYFPEPVTVSWNSGAL TSGVHTFPAVLQSSGL YSLSSVVTV PSSSLGTQTYICNVNHKPSNTKVDKKVEPKSCDKTHTCPPCPAPELLGRPSVFLFPPKPK D TLMISR TPEVTCVWD VSHEDPEVKFNWYVDGVE VHNAKTKPREEQYNS TYRVVSVL TVLH QDWLNGKEYKCKVSNKALPAPIEKTISKAKGQPREPQVYTLPPSREEMTKNQVSL TCL VKG FYPSDIAVEWESNGQPENNYKTTPPVLDSDGSFFL YSKL TVDKSRWQQGNVFSCSVMHEAL

JiVgyrQ S-LS-LSPG GGSAGSAGSAGSGGSEPVQFKDCGSVDGVIKEVNVSPCPTQPCQLS KGQSYSVNVTFTSNIQSKSSKAWHGILMGVPVPFPIPEPDGCKSGINCPIQKDKTYSYLN KLPVKSEYPSIKLWEWQLQDDKNQSLFC EIPVQIVSHL , and

548-NPC2-HCC (NPC2 C-terminal fusion to heavy chain of 548 IgGl) (SEQ ID NO:22)

EVQLQQSGAELVKPGASVKLSCTASGFNIKDTYMHWVKQRPEEGLEWIGRIDPADGN TEYV PKFQGKATITADTFSNTVYLQLSGLTSEDTAVYYCSRGYYWGQGTTLTVSS ASTKGPSVFP LAPSSKS TSGGTAALGCLVKDYFPEPVTVSWNSGAL TSGVHTFPAVLQSSGL YSLSSVVTV PSSSLGTQTYICNVNHKPSNTKVDKKVEPKSCDKTHTCPPCPAPELLGRPSVFLFPPKPK D TLMLSR TPEVTCVWD VSHEDPEVKFNWYVDGVE VHNAKTKPREEQYNS TYRVVSVL TVLH QDWLNGKEYKCKVSNKALPAPLEKTISKAKGQPREPQVYTLPPSREEMTKNQVSL TCL VKG FYPSDLAVEWESNGQPENNYKTTPPVLDSDGSFFL YSKL TVDKSRWQQGNVFSCSVMHEAL

JiVgyrQ S-LS-LSPG GGSAGSAGSAGSGGSEPVQFKDCGSVDGVIKEVNVSPCPTQPCQLS KGQSYSVNVTFTSNIQSKSSKAWHGILMGVPVPFPIPEPDGCKSGINCPIQKDKTYSYLN KLPVKSEYPSIKLWEWQLQDDKNQSLFC EIPVQIVSHL .

[0044] Preferably, 548 has the following epitope, DFFVYADYHT (SEQ ID NO: 23), which corresponds to residues 502-511 of human NPC1.

[0045] Preferably, the monoclonal antibody or bispecific antibody is a human antibody or humanized antibody. "Humanized" forms of non-human (e.g., murine) antibodies are chimeric antibodies that contain minimal sequence derived from non-human immunoglobulin. In one embodiment, a humanized antibody is a human immunoglobulin (recipient antibody) in which residues from a hypervariable region (HVR) of the recipient are replaced by residues from a HVR of a non-human species (donor antibody) such as mouse, rat, rabbit, or nonhuman primate having the desired specificity, affinity, and/or capacity. In some instances, FR residues of the human immunoglobulin variable domain are replaced by corresponding non-human residues. These modifications may be made to further refine antibody performance. Furthermore, in a specific embodiment, humanized antibodies may comprise residues that are not found in the recipient antibody or in the donor antibody. In an embodiment, the humanized antibodies do not comprise residues that are not found in the recipient antibody or in the donor antibody. In general, a humanized antibody will comprise substantially all of at least one, and typically two, variable domains, in which all or substantially all of the hypervariable loops correspond to those of a non-human immunoglobulin, and all or substantially all of the FRs are those of a human immunoglobulin sequence. The humanized antibody optionally will also comprise at least a portion of an immunoglobulin constant region (Fc), typically that of a human immunoglobulin. See, e.g., Jones et al, Nature 321 :522-525 (1986); Riechmann et al., Nature 332:323-329 (1988); Presta, Curr. Op. Struct. Biol. 2:593-596 (1992); Vaswani and Hamilton, Ann. Allergy, Asthma & Immunol. 1 : 105-115 (1998); Harris, Biochem. Soc. Transactions 23: 1035-1038 (1995); Hurle and Gross, Curr. Op. Biotech. 5:428-433 (1994); and U.S. Pat. Nos. 6,982,321 and 7,087,409, the contents of each of which references and patents are hereby incorporated by reference in their entirety. In one embodiment where the humanized antibodies do comprise residues that are not found in the recipient antibody or in the donor antibody, the Fc regions of the antibodies are modified as described in WO 99/58572, the content of which is hereby incorporated by reference in its entirety.

[0046] Techniques to humanize a monoclonal antibody are described in U.S. Pat. Nos. 4,816,567; 5,807,715; 5,866,692; 6,331,415; 5,530,101; 5,693,761; 5,693,762; 5,585,089; and 6,180,370, the content of each of which is hereby incorporated by reference in its entirety.

[0047] A number of "humanized" antibody molecules comprising an antigen-binding site derived from a non-human immunoglobulin have been described, including antibodies having rodent or modified rodent V regions and their associated complementarity determining regions (CDRs) fused to human constant domains. See, for example, Winter et al. Nature 349: 293-299 (1991), Lobuglio et al. Proc. Nat. Acad. Sci. USA 86: 4220-4224 (1989), Shaw et al. J. Immunol. 138: 4534-4538 (1987), and Brown et al. Cancer Res. 47: 3577-3583 (1987), the content of each of which is hereby incorporated by reference in its entirety. Other references describe rodent hypervariable regions or CDRs grafted into a human supporting framework region (FR) prior to fusion with an appropriate human antibody constant domain. See, for example, Riechmann et al. Nature 332: 323-327 (1988), Verhoeyen et al. Science 239: 1534-1536 (1988), and Jones et al. Nature 321 : 522-525 (1986), the content of each of which is hereby incorporated by reference in its entirety. Another reference describes rodent CDRs supported by recombinantly veneered rodent framework regions - European Patent Publication No. 0519596 (incorporated by reference in its entirety). These "humanized" molecules are designed to minimize unwanted immunological response toward rodent anti-human antibody molecules which limits the duration and effectiveness of therapeutic applications of those moieties in human recipients. The antibody constant region can be engineered such that it is immunologically inert (e.g., does not trigger complement lysis). See, e.g. PCT Publication No. W099/58572; UK Patent Application No. 9809951.8. Other methods of humanizing antibodies that may also be utilized are disclosed by Daugherty et al., Nucl. Acids Res. 19: 2471-2476 (1991) and in U.S. Pat. Nos. 6,180,377; 6,054,297; 5,997,867; 5,866,692; 6,210,671; and 6,350,861; and in PCT Publication No. WO 01/27160 (each incorporated by reference in their entirety).

[0048] Other forms of humanized antibodies have one or more CDRs (CDR LI, CDR L2, CDR L3, CDR HI, CDR H2, or CDR H3) which are altered with respect to the original antibody, which are also termed one or more CDRs "derived from" one or more CDRs from the original antibody.

[0049] The invention also provises pharmaceutical compositions for treating or preventing or reducing or inhibiting a filovirus infection in a subject comprising any of the bispecific antibodies or bispecific fusion constructs disclosed herein and a pharmaceutically acceptable carrier. Examples of acceptable pharmaceutical carriers include, but are not limited to, additive solution-3 (AS-3), saline, phosphate buffered saline, Ringer's solution, lactated Ringer's solution, Locke-Ringer's solution, Krebs Ringer's solution, Hartmann's balanced saline solution, and heparinized sodium citrate acid dextrose solution. The bispecific antibody or bispecific fusion construct can be administered to the subject in a pharmaceutical composition comprising a pharmaceutically acceptable carrier. The pharmaceutically acceptable carrier used can depend on the route of administration. The pharmaceutical composition can be formulated for administration by any method known in the art, including but not limited to, oral administration, parenteral administration, intravenous administration and administration through an osmotic mini-pump.

[0050] The invention further provides an isolated nucleic acid encoding any of the bispecific antibodies or bispecific fusion constructs disclosed herein. The isolated nucleic acid can be, or comprise, a cDNA.

[0051] This invention will be better understood from the Experimental Details, which follow. However, one skilled in the art will readily appreciate that the specific methods and results discussed are merely illustrative of the invention as described more fully in the claims that follow thereafter.

EXPERIMENTAL DETAILS

[0052] Introduction and Overview: Ebola virus (EBOV) and related filoviruses are associated with sporadic outbreaks of highly lethal hemorrhagic fever in Middle and West Africa. The ongoing regional EBOV epidemic in West Africa has underscored the urgent need for antiviral treatments and demonstrated the potential of passive immunotherapy to reverse advanced filovirus disease. However, existing monoclonal antibody (mAb) cocktails such as ZMapp are limited by a narrow spectrum of antiviral action, which stems from viral strain-specific neutralization of the highly variable entry glycoprotein, GP, by most mAbs. Accordingly, set forth herein is an immunotherapeutic strategy targeting the broadly required and highly conserved filovirus entry receptor Niemann-Pick CI (NPCl) instead of GP. Unfortunately, anti-NPCl mAbs that efficiently blocked GP-NPC1 interaction in vitro failed to neutralize viral infection in cells, presumably because of an unusual feature of filovirus receptor recognition— the GP-NPC1 interaction can occur only in cellular endosomes where both virus and receptor are likely protected from extracellular antibodies. To overcome this limitation, bispecific Abs (bsAbs) were generated that combine both antiviral and anti-receptor specificities in the same molecule. These bsAbs potently neutralized EBOV infection in a manner that required their engagement of both GP and NPCl, indicating a two-step mechanism of action in which they exploit extracellular virus particles to gain access to NPCl -containing endosomes. bsAbs that combine broadly- reactive (but non-neutralizing) anti-GP Abs with anti-NPCl Abs are expected to afford broad-spectrum protection against filovirus infection in vivo.

[0053] NPCl is required for Ebola virus infection and in vivo pathogenesis. Niemann- Pick CI (NPCl), a ubiquitous multi-pass membrane protein localized to late endosomes is required for cytoplasmic entry and infection by all filoviruses. The filovirus spike glycoprotein, GP, must engage NPCl 's second luminal domain (Fig. 1A, domain C) to drive viral membrane fusion and cytoplasmic escape (Miller, EMBO J 2012). As shown in Fig. IB, NPCl -knockout mice (NPCl ^"7" ), but not their WT littermates (NPC1 ^+/+ ), are completely resistant to infection and killing by EBOV and MARV (Carette, Nature 2011 ; Herbert, submitted).

[0054] Mouse hybridomas were raised that express mAbs specific for human NPCl domain C (data not shown). Screens identified two mAbs (548 and 952) that bound with high affinity to NPCl and potently blocked its interaction with the EBOV glycoprotein GP in vitro. A third mAb (401) had much more modest receptor-blocking activity. Fig. 3 shows the capacity of three anti-NPCl mAbs to block GP-NPC1 binding in an ELISA. A soluble, flag-tagged form of NPCl domain C was pre-incubated with the indicated concentrations of each mAb, and the protein-mAb mixtures were then added to plates coated with vesicular stomatitis virus (VSV) particles bearing EBOV GP. Bound domain C was detected with an anti-flag antibody. Irrelevant isotype-matched antibodies had no effect on GP-NPC1 domain C binding (not shown).

[0055] As exemplified in Fig. 4, anti-NPCl domain C mAbs neutralize VSV-EBOV GP infection weakly or not at all. VSV-GP particles were exposed to U20S human osteosarcoma cells in the presence of increasing concentrations of NPCl -specific mAbs or a control GP-specific neutralizing mAb (KZ52). The number of infected cells was determined at 16 h post-infection, and normalized to that obtained in the absence of antibody (set to 100%). The NPCl-specific mAbs 548 and 401 neutralized VSV-EBOV GP infection weakly or not at all, respectively. By contrast, the previously described GP-specific neutralizing mAb KZ52 was highly effective at neutralization.

[0056] Fig. 5 illustrates engineering antibodies with dual binding specificities for GP and NPCl. In Fig. 5 A, constructs encoding bsAbs were generated by fusing NPCl-specific or control IgG sequences to a single-chain variable fragment (scFv) sequence derived from the EBOV GP-specific mAb KZ52. Fusions were done in four configurations: to the N- or C-terminus of the IgG heavy chain [HC] or light chain [LC]. In Fig. 5B, bsAbs and IgGs were expressed in 293-Freestyle cells by co-transfection of HC and LC expression vectors, and purified by protein A affinity chromatography. In Fig. 5C, it is shown that bsAbs, but not NPCl-specific IgGs, can simultaneously bind to GP and NPCl domain C. bsAbs and IgGs were captured onto GP-coated ELISA plates and then incubated with NPCl domain C- flag. Bound NPCl domain C was detected with an anti-flag antibody.

[0057] It was postulated that the NPCl -specific mAbs do not neutralize EBOV entry because they cannot efficiently access NPCl -containing late endosomes. To overcome this obstacle, an exploratory panel of bispecific antibodies (bsAbs) containing both NPC1- binding (401 or 548) and GP-binding (KZ52) specificities was generated. It was reasoned that one or more of these bsAbs might be able to 'ride' virus particles into late endosomes and then bind to NPCl, thereby inhibiting viral entry by both GP -blocking and receptor- blocking mechanisms. It was found that all four purified bsAbs could bind simultaneously to EBOV GP and NPCl (e.g., see Fig. 5C), setting the stage for infection neutralization studies.

[0058] As shown in Fig. 6, GP-NPC1 bsAbs neutralize VSV-EBOV GP infection with enhanced potency. In the left panel, GP-NPC1 bsAbs show enhanced potency at viral neutralization relative to control GP-only bsAbs. VSV-EBOV GP particles were pre- incubated with the indicated concentrations of each bsAb for 1 hour at room temperature and then allowed to infect U20S cells. Infected cells were quantified at 16 hours postinfection and normalized to a no-antibody control (100%). In the right panel, the neutralization curves were fit to a logistic equation to extract the concentration of half- maximal neutralization (IC5 ₀ ± 95% confidence intervals). The GP-NPC1 bsAbs were 10- 60 fold more potent at neutralizing EBOV entry than control bsAbs in which an irrelevant (non-NPCl binding) IgG was fused to the same scFv, derived from the GP-specific mAb, KZ52. These results indicate that the bsAbs possess an enhanced capacity to block GP- NPC1 binding within endosomes, relative to the NPCl-specific IgGs.

[0059] As shown in Fig. 6, the enhanced neutralization potency of GP-NPC1 bsAbs can be attributed to their receptor blocking activity. In the left panel, VSV-EBOV GP particles were pre-incubated with the indicated concentrations of each bsAb for 1 hour at room temperature and then allowed to infect matched U20S cell lines expressing endogenous levels of NPCl or over-expressing NPCl (NPCl ^hl ). Infected cells were quantified at 16 hours post-infection and normalized to a no-antibody control (100%). In the right panel, the neutralization curves were fit to a logistic equation to extract the concentration of half- maximal neutralization (IC5 ₀ ± 95% confidence intervals). The ratio of IC5 ₀ S for each bsAb in NPCl ^hl vs. WT U20S cells is shown. It was reasoned that the neutralization potency of GP-NPC1 bsAbs should be reduced in cells engineered to overexpress NPCl, because the excess NPCl should titrate the available NPCl -binding sites in the bsAbs and thereby increase the probability of productive GP-NPC1 binding. This is precisely what was observed. The bsAb 548-HCN' suffered a 24-fold loss in potency in U20S-NPCl ^hi cells, relative to WT U20S cells. By contrast, the potency of the Ctrl-HCN' bsAb, which cannot bind to NPCl, was reduced much more modestly (~2-fold). These findings provide strong evidence that the bsAbs neutralize viral infection by deploying both GP- and NPCl -binding specificities within cells.

[0060] An additional strategy to deliver NPCl -specific IgGs to NPCl -positive late endosomes was also developed. The IgGs were fused to NPC2, a late endosomal/lysosomal host protein that interacts with NPCl and that can autonomously traffic to late endosomes when added to cells (Naureckiene, Science 2000). As shown in Fig. 8, fusion of a receptor- blocking IgG to a late endosome-localizing protein facilitates viral neutralization. bsAbs comprising an NPCl-specific IgG (401 or 548) fused to NPC2, a soluble late endosomal/lysosomal protein, were expressed and purified, and examined for their capacity to neutralize VSV-EBOV GP infection, as described in Figs. 6-7. Fusion of the potent receptor-blocking IgG 548 to NPC2 afforded neutralization of infection, whereas fusion of the weak receptor-blocking IgG 401 to NPC2 did not. Therefore, direct, cellular protein- mediated delivery of NPCl -specific IgGs to NPC1+ endosomes is also a feasible strategy for targeting NPCl and preventing NPCl -mediated filovirus entry into cells.

[0061] The NPCl -targeting bispecific antibody approaches described herein provide an unprecedented opportunity for anti-filovirus therapeutics with increased antiviral breadth.

REFERENCES

Brummelkamp TR, Chandran K. Ebola virus entry requires the host-programmed recognition of an intracellular receptor. EMBO J 31 : 1947-1960, 2012.

Carette, J.E., Raaben, M., Wong, A.C., Herbert, A.S., Obernosterer, G, Mulherkar, N., Kuehne, A.I., Kranzusch, P. J., Griffin, A.M., Ruthel, G., et al. Ebola virus entry requires the cholesterol transporter Niemann-Pick CI. Nature 477: 340-343, 2011.

Davies JP, Ioannou YA. Topological analysis of Niemann-Pick CI protein reveals that the membrane orientation of the putative sterol-sensing domain is identical to those of 3- hydroxy-3-methylglutaryl-CoA reductase and sterol regulatory element binding protein cleavage-activating protein. J Biol Chem. 275(32):24367-74, 2000. Keck ZY, et al. Macaque Monoclonal Antibodies Targeting Novel Conserved Epitopes within Filovirus Glycoprotein. J Virol 90:279-91, 2015.

Miller EH, Obernosterer G, Raaben M, Herbert AS, Deffieu M, Krishnan A, Ndungo E, Sandesara RG, Carette JE, Kuehne Al, Ruthel G, Pfeffer SR, Dye JM, Whelan SP, Naureckiene S, Sleat DE, Lackland H, Fensom A, Vanier MT, Wattiaux R, Jadot M, Lobel P. Identification of HE1 as the second gene of Niemann-Pick C disease. Science. 290:2298- 2301, 2000.

Negredo A, Palacios G, Vazquez-Moron S, Gonzalez F, Dopazo H, Molero F, Juste J, Quetglas J, Savji N, de la Cruz Martinez M, Herrera JE, Pizarro M, Hutchison SK, Echevarria JE, Lipkin WI, Tenorio A. Discovery of an ebolavirus-like filovirus in europe. PLoS Pathog. 7:el002304, 2011.

Ng M, Ndungo E, Jangra RK, Cai Y, Postnikova E, Radoshitzky SR, Dye JM, Ramirez de Arellano E, Negredo A, Palacios G, Kuhn JH, Chandran K. 2014. Cell entry by a novel European filovirus requires host endosomal cysteine proteases and Niemann-Pick CI . Virology. 468-470:637-646, 2014.

Qiu X, Wong G, Audet J, Bello A, Fernando L, Alimonti JB, Fausther-Bovendo H, Wei H, Aviles J, Hiatt E, Johnson A, Morton J, Swope K, Bohorov O, Bohorova N, Goodman C, Kim D, Pauly MH, Velasco J, Pettitt J, Olinger GG, Whaley K, Xu B, Strong JE, Zeitlin L, Kobinger GP. 2014. Reversion of advanced Ebola virus disease in nonhuman primates with ZMapp. Nature. 514:47-53, 2014.

Wu C, Ying H, Grinnell C, Bryant S, Miller R, Clabbers A et al. Simultaneous targeting of multiple disease mediators by a dual variable-domain immunoglobulin. Nat Biotechnol 25: 1290-1297, 2007.