HIV-1-NEUTRALIZING ANTIBODY POTENCY AND BREADTH VIA CELL RECEPTOR ANCHORING USING BISPECIFIC ANTIBODIES WITH NATIVE ARCHITECTURE

This application claims priority to and the benefit of, and incorporates herein by reference in its entirety, U.S. Provisional Patent Application No. 61/910,685, entitled IMPROVED HIV-1-NEUTRALIZING ANTIBODY POTENCY AND BREADTH VIA CELL RECEPTOR ANCHORING USING BISPECIFIC ANTIBODIES WITH NATIVE ARCHITECTURE, which was filed on Dec. 2, 2013.

In various embodiments, the present invention relates generally to using bispecific antibodies in the prevention and treatment of HIV.

Passive immunization with antibodies (Abs) is a recognized method of prophylaxis and treatment of infectious diseases. This approach may involve preparing human immunoglobulins from donors who recovered from an infectious disease and utilizing such preparations, containing Abs specific for the infectious organism, to protect a recipient against the same disease. Alternatively, therapeutic antibodies can be made by immunizing mice with an antigen, and then engineering/humanizing the mouse Ab into a human version. Monoclonal antibodies (mAbs) are homogeneous in terms of physical characteristics and immunochemical reactivity, and so offer the possibility of absolute specific activity.

That specificity can ultimately be a limitation for some targets, so practitioners have developed “bispecific” mAbs composed of fragments of two different mAbs and which bind to two different types of antigen. This facilitates binding to antigens expressed only weakly, for example. Some bispecific mAbs can stimulate strong immune responses, limiting their clinical application. One recent approach to ameliorating this effect is “CrossMab” methodology, a bispecific antibody format that adopts a more native antibody-like structure.

The prospects for generating a highly potent bispecific or bivalent antibody against a pathogen, such as HIV, for clinical use involves many uncertainties. The low spike density and spike structure on HIV may impede bivalent binding of antibodies to HIV, for example, and the geometry and spatial relationship of cell surface anchoring are not well-characterized. Nor is it known whether sufficient epitope accessibility on the HIV envelope exists. CrossMab bispecific antibodies that are anchored to a host cell membrane offer the possibility of improved local antibody concentration, targeting of sequential and/or interdependent entry steps, and compensating for monovalent binding.

In one aspect, the present invention pertains to a bispecific antibody for neutralizing HIV. The bispecific antibody includes portions of a first and a second antibody, in which the first antibody binds to a HIV envelope protein. In certain embodiments, the first antibody is selected from PGT145, PG9, PGT128, PGT121, 10-1074, 3BNC117, VRC01, PGT151, 4E10, 10E8 and a variant thereof. In certain embodiments, the bispecific antibody includes portions of a second antibody, in which the second antibody binds to a cell membrane protein. For example, the second antibody may binds to a cell receptor protein or a cell membrane co-receptor protein. In an embodiment, the second antibody is selected from a CD4 antibody, a CCR5 antibody and a CXCR4 antibody, such as Pro 140, ibalizumab, 515H7, or a variant thereof. In various embodiments, the bispecific antibody has a CrossMab format.

In another aspect, the present invention provides a bispecific antibody including portions of a first antibody and a second antibody, wherein the first antibody binds to a HIV envelope protein and the second antibody binds to a cell membrane protein. In various embodiments, the bispecific antibody has a CrossMab format.

In various embodiments, pharmaceutical compositions including the bispecific antibodies disclosed herein are also provided. The pharmaceutical composition may be formulated for oral, intranasal, pulmonary, intradermal, transdermal, subcutaneous, intramuscular, intraperitoneal, or intravenous delivery.

In a further aspect, methods for neutralizing HIV are provided. The methods include the steps of contacting an antigen binding site with a bispecific antibody that binds a HIV envelope protein and contacting another antigen binding site with a bispecific antibody that binds a cell membrane protein.

In another aspect, methods for treating a patient infected with HIV are also provided. The methods include administering to the patient any of the bispecific antibodies or pharmaceutical compositions as disclosed herein. In an embodiment, the patient is human.

Embodiments of the present invention provide for inhibition of HIV. In various implementations, bispecific antibodies are formed, each including heavy chain and light chain components from two different parent antibodies. One parent antibody specifically binds HIV, for example, the HIV envelope protein Env. The other parent antibody specifically binds a cell membrane protein, for example CD4 and CCR5. In a bispecific antibody, a heavy chain and light chain from each of two parental antibodies are combined, providing an antibody in which the antigen binding sites of fragment antigen-binding 1 (Fab1) and Fab2 have different binding specificities. In certain embodiments, the bispecific antibody is a CrossMab format antibody, as shown in FIG. 1. In a CrossMab format, one heavy chain includes a “knob” structure and the other heavy chain includes a corresponding “hole” structure, and the positions of the constant domains (i.e., CL and CH1) from one parental antibody are switched, which together ensure correct pairing of heavy chains and light chains during assembly.

Various mAbs have been shown to block HIV infection by targeting and binding to the HIV envelope protein Env (FIGS. 2B and 10). These mAbs include, for example, PGT145, PG9, PGT128, PGT121, 10-1074, 3BNC117, VRC01, PGT151, 4E10, and 10E8. In addition, monoclonal antibodies Pro 140 (“P140”), Ibalizumab (“iMab”) and 515H7 have been shown to block HIV infection by targeting and binding to CCR5, CD4 and CXCR4 human cell membrane proteins, respectively (FIG. 2A). Specifically, FIG. 2A shows how iMab targets CD4, the primary receptor for HIV-1 entry that is expressed on human T-cells; and how Pro 140 targets CCR5, a co-receptor for HIV-1 entry by CCR5 tropic HIV-1. FIG. 2B (adapted from http://www.scripps.edu/news/press/2014/20140424hiv.html) illustrates how the mAb PGT145 targets the V1/V2 epitope on the HIV viral envelope gp120; how mAb PGT128 targets the glycan on the V3 stem region of HIV gp120; how mAb 3BNC117 targets the CD4 binding site of HIV gp120; how mAb 10E8 targets the membrane proximal external region (MPER) of HIV gp41; and how mAb PGT151 targets an epitope on both HIV gp120 and HIV gp41. In various embodiments, the bispecific antibody (e.g., a HIV CrossMab antibody) of the present invention has the natural architecture of an IgG molecule, but with bispecificity.

Although the ensuing discussion focuses on the use of bispecific antibodies directed to Env and the cell membrane proteins CD4 and CCR5, it is to be understood that this is solely for ease of presentation, and that any suitable antibody directed to any HIV epitope and any suitable antibody directed to any suitable cell membrane protein may be used and are within the scope of the invention.

Accordingly, in various embodiments, the present invention provides bispecific antibodies that target and bind to the HIV Env protein as well as the cell membrane proteins CCR5, CD4 and/or CXCR4. In certain embodiments, the bispecific antibodies include sequences (for example, heavy and light chain sequences) derived from, but not limited to, the PGT145, PG9, PGT128, PGT121, 10-1074, 3BNC117, VRC01, PGT151, 4E10, and/or 10E8 antibodies and variants thereof.

The amino acid sequences defining the heavy and light chains of the PGT145 antibody can be found, for example, at http://www.ncbi.nlm.nih.gov/protein/3U1S_H and http://www.ncbi.nlm.nih.gov/protein/3U1S_L, respectively, the entire contents of which are incorporated herein by reference.

The amino acid sequences defining the heavy and light chains of the PG9 antibody can be found, for example, at http://www.ncbi.nlm.nih.gov/protein/3U4E_H and http://www.ncbi.nlm.nih.gov/protein/3MUH_L, respectively, the entire contents of which are incorporated herein by reference.

The amino acid sequences defining the heavy and light chains of the PGT128 antibody can be found, for example, at http://www.ncbi.nlm.nih.gov/protein/3TYG_H and http://www.ncbi.nlm.nih.gov/protein/3TYG_L, respectively, the entire contents of which are incorporated herein by reference.

The amino acid sequences defining the heavy and light chains of the PGT121 antibody can be found, for example, at http://www.ncbi.nlm.nih.gov/protein/4FQC_H and http://www.ncbi.nlm.nih.gov/protein/4FQC_L, respectively, the entire contents of which are incorporated herein by reference.

The amino acid sequences defining the heavy and light chains of the 10-1074 antibody can be found, for example, in Mouquet H., et al., (2012) PNAS, 109(47):E3268-77 (including supplementary information), the entire contents of which are incorporated herein by reference.

The amino acid sequences defining the heavy and light chains of the 3BNC117 antibody can be found, for example, at http://www.ncbi.nlm.nih.gov/protein/4LSV_H and http://www.ncbi.nlm.nih.gov/protein/4LSV_L, respectively, the entire contents of which are incorporated herein by reference.

The amino acid sequences defining the heavy and light chains of the VRC01 antibody can be found, for example, at http://www.ncbi.nlm.nih.gov/protein/4LST_H and http://www.ncbi.nlm.nih.gov/protein/4LST_L, respectively, the entire contents of which are incorporated herein by reference.

The amino acid sequences defining the heavy and light chains of the PGT151 antibody can be found, for example, at http://www.ncbi.nlm.nih.gov/protein/4NUG_H and http://www.ncbi.nlm.nih.gov/protein/4NUG_L, respectively, the entire contents of which are incorporated herein by reference.

The amino acid sequences defining the heavy and light chains of the 4E10 antibody can be found, for example, at http://www.ncbi.nlm.nih.gov/protein/4LLV_H and http://www.ncbi.nlm.nih.gov/protein/4LLV_L, respectively, the entire contents of which are incorporated herein by reference.

The amino acid sequences defining the heavy and light chains of the 10E8 antibody can be found, for example, at http://www.ncbi.nlm.nih.gov/protein/4G6F_B and http://www.ncbi.nlm.nih.gov/protein/4G6F_D, respectively, the entire contents of which are incorporated herein by reference.

In certain embodiments, the bispecific antibodies include sequences (for example, heavy and light chain sequences) derived from, but not limited to, the P140, iMab (or the MV1 variant) and/or 515H7 antibodies and variants thereof. The heavy and light chain sequences of the Pro 140, Ibalizumab (or its MV1 variant), and 515H7 antibodies are further described, for example, in Olson, W. C. et al., (1999) J Virol., 73(5):4145-55, Trkola, A. et al., (2001) J Virol., 75(2):579-88, U.S. Pat. No. 7,122,185, Burkly L. C. et al., (1992) J Immunol., 149(5):1779-87, Moore J. P. et al., (1992) J Virol., 66(8):4784-93, Reimann K. A., et al., (1997) AIDS Res Hum Retroviruses, 13(11):933-43, International Patent Publication No. WO2014100139, and European Patent Publication No. EP2246364, the entire contents of all of which are incorporated herein by reference.

As used herein, an antibody “variant” refers to an antibody which has an amino acid sequence which differs from the amino acid sequence of a parent antibody from which it is derived. In various embodiments, the variant has one or more amino acid alterations with respect to the parent antibody.

An embodiment of a bispecific antibody includes a heavy and light chain sequence from the PGT145, PG9, PGT128, PGT121, 10-1074, 3BNC117, VRC01, PGT151, 4E10, or 10E8 antibody or a variant thereof and a heavy and light chain sequence from the P140, iMab (or the MV1 variant), or 515H7 antibody or a variant thereof.

In exemplary embodiments, a series of HIV CrossMab antibodies have been constructed including but not limited to, for example, 145/MV1, 117/MV1, 128/MV1, 10E8/MV1, 145/P140, 128/P140, 117/P140, 10E8/P140, 10E8/alpha-Her2, 10E8/X19, and 4E10/P140. PGT145 (“145”), 3BNC117 (“117”), PGT128 (“128”), and 10E8 are four different HIV envelope antibodies. Pro 140 (“P140”) is a mAb that binds to the cell surface receptor CCR5. MV1 is a CD4 antibody that is a modified variant of the mAb Ibalizumab (see, for example, International Patent Publication No. WO2014100139, incorporated herein by reference in its entirety). X19 is one of the antibody variants of the anti-cell surface receptor CXCR4 (see, for example, U.S. Pat. No. 8,329,178, incorporated herein by reference in its entirety) that does not bind to cells expressing CXCR4 (and is therefore used as a non-surface binding control). Alpha-Her2 is a mAb that binds to the Her2 receptor expressed on cells. Many of these CrossMab antibodies increase the breadth of HIV neutralization as compared to their parental antibodies (i.e., monoclonal antibodies MV1, 145, 117 or 10E8). In addition, many of these antibodies also significantly increase the potency of neutralization against HIV as compared to their parental antibodies.

The amino acid sequences defining the heavy and light chains of various HIV CrossMab antibodies are shown below.

In various embodiments, at least one of the heavy chain and/or light chain sequences derived from the PGT145, PG9, PGT128, PGT121, 10-1074, 3BNC117, VRC01, PGT151, 4E10, 10E8, P140, iMab (or the MV1 variant), 515H7 antibodies and variants thereof are paired together to form a bispecific antibody (e.g., a HIV CrossMab antibody). In an exemplary embodiment, at least one of the disclosed heavy and light chains selected from SEQ ID NOs: 1-36 are paired together to form a bispecific antibody (e.g., a HIV CrossMab antibody).

In various embodiments, the amino acid sequence of the bispecific antibody (e.g., HIV CrossMab antibody) further includes an amino acid analog, an amino acid derivative, or other non-classical amino acids.

In various embodiments, the bispecific antibody (e.g., HIV CrossMab antibody) comprises a sequence that is at least 60% identical to a wild-type heavy or light chain sequence of the PGT145, PG9, PGT128, PGT121, 10-1074, 3BNC117, VRC01, PGT151, 4E10, or 10E8 antibody. In various embodiments, the bispecific antibody (e.g., HIV CrossMab antibody) comprises a sequence that is at least 60% identical to a wild-type heavy chain or light chain sequence of the P140, iMab (or the MV1 variant), or 515H7 antibody. In exemplary embodiments, the bispecific antibody (e.g., HIV CrossMab antibody) comprises a sequence that is at least 60% identical to SEQ ID NOs: 1-36.

In various embodiments, the bispecific antibody (e.g., HIV CrossMab antibody) may comprise a sequence that is at least about 60%, at least about 61%, at least about 62%, at least about 63%, at least about 64%, at least about 65%, at least about 66%, at least about 67%, at least about 68%, at least about 69%, at least about 70%, at least about 71%, at least about 72%, at least about 73%, at least about 74%, at least about 75%, at least about 76%, at least about 77%, at least about 78%, at least about 79%, at least about 80%, at least about 81%, at least about 82%, at least about 83%, at least about 84%, at least about 85%, at least about 86%, at least about 87%, at least about 88%, at least about 89%, at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, or 100% identical to a wild-type heavy chain or light chain sequence of the PGT145, PG9, PGT128, PGT121, 10-1074, 3BNC117, VRC01, PGT151, 4E10, or 10E8 antibody.

In various embodiments, the bispecific antibody (e.g., HIV CrossMab antibody) may comprise a sequence that is at least about 60%, at least about 61%, at least about 62%, at least about 63%, at least about 64%, at least about 65%, at least about 66%, at least about 67%, at least about 68%, at least about 69%, at least about 70%, at least about 71%, at least about 72%, at least about 73%, at least about 74%, at least about 75%, at least about 76%, at least about 77%, at least about 78%, at least about 79%, at least about 80%, at least about 81%, at least about 82%, at least about 83%, at least about 84%, at least about 85%, at least about 86%, at least about 87%, at least about 88%, at least about 89%, at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, or 100% identical to a wild-type heavy chain or light chain sequence of the P140, iMab (or the MV1 variant), or 515H7 antibody.

In exemplary embodiments, the bispecific antibody (e.g., HIV CrossMab antibody) may comprise a sequence that is at least about 60%, at least about 61%, at least about 62%, at least about 63%, at least about 64%, at least about 65%, at least about 66%, at least about 67%, at least about 68%, at least about 69%, at least about 70%, at least about 71%, at least about 72%, at least about 73%, at least about 74%, at least about 75%, at least about 76%, at least about 77%, at least about 78%, at least about 79%, at least about 80%, at least about 81%, at least about 82%, at least about 83%, at least about 84%, at least about 85%, at least about 86%, at least about 87%, at least about 88%, at least about 89%, at least about 90%, at least about 91%, at least about 92%, at least about 93%, at least about 94%, at least about 95%, at least about 96%, at least about 97%, at least about 98%, at least about 99%, or 100% identical to SEQ ID NOs: 1-36.

Homology or identity may be determined in various ways that are within the skill in the art, for instance, using publicly available computer software such as BLAST, BLAST-2, ALIGN or Megalign (DNASTAR) software. BLAST (Basic Local Alignment Search Tool) analysis using the algorithm employed by the programs blastp, blastn, blastx, tblastn and tblastx (Karlin et al., (1990) PROC. NATL. ACAD. SCI. USA 87, 2264-2268; Altschul, (1993) J. MOL. EVOL. 36, 290-300; Altschul et al., (1997) NUCLEIC ACIDS RES. 25, 3389-3402, incorporated by reference) are tailored for sequence similarity searching. The approach used by the BLAST program is to first consider similar segments between a query sequence and a database sequence, then to evaluate the statistical significance of all matches that are identified and finally to summarize only those matches which satisfy a preselected threshold of significance. For a discussion of basic issues in similarity searching of sequence databases see Altschul et al., (1994) NATURE GENETICS 6, 119-129 which is fully incorporated by reference. Those skilled in the art can determine appropriate parameters for measuring alignment, including any algorithms needed to achieve maximal alignment over the full length of the sequences being compared. The search parameters for histogram, descriptions, alignments, expect (i.e., the statistical significance threshold for reporting matches against database sequences), cutoff, matrix and filter are at the default settings. The default scoring matrix used by blastp, blastx, tblastn, and tblastx is the BLOSUM62 matrix (Henikoff et al., (1992) PROC. NATL. ACAD. SCI. USA 89, 10915-10919, fully incorporated by reference). Four blastn parameters may be adjusted as follows: Q=10 (gap creation penalty); R=10 (gap extension penalty); wink=1 (generates word hits at every wink.sup.th position along the query); and gapw=16 (sets the window width within which gapped alignments are generated). The equivalent Blastp parameter settings may be Q=9; R=2; wink=1; and gapw=32. Searches may also be conducted using the NCBI (National Center for Biotechnology Information) BLAST Advanced Option parameter (e.g.: -G, Cost to open gap: default=5 for nucleotides/11 for proteins; -E, Cost to extend gap: default=2 for nucleotides/1 for proteins; -q, Penalty for nucleotide mismatch: default=-3; -r, reward for nucleotide match: default=1; -e, expect value: default=10; -W, wordsize: default=11 for nucleotides/28 for megablast/3 for proteins; -y, Dropoff (X) for blast extensions in bits: default=20 for blastn/7 for others; -X, X dropoff value for gapped alignment (in bits): default=15 for all programs, not applicable to blastn; and -Z, final X dropoff value for gapped alignment (in bits): 50 for blastn, 25 for others). ClustalW for pairwise protein alignments may also be used (default parameters may include, e.g., Blosum62 matrix and Gap Opening Penalty=10 and Gap Extension Penalty=0.1). A Bestfit comparison between sequences, available in the GCG package version 10.0, uses DNA parameters GAP=50 (gap creation penalty) and LEN=3 (gap extension penalty) and the equivalent settings in protein comparisons are GAP=8 and LEN=2.

In various embodiments, the bispecific antibody (e.g., HIV CrossMab antibody) comprises a sequence that includes at least one amino acid alteration with respect to a wild-type heavy or light chain sequence of the PGT145, PG9, PGT128, PGT121, 10-1074, 3BNC117, VRC01, PGT151, 4E10, or 10E8 antibody. In various embodiments, the bispecific antibody (e.g., HIV CrossMab antibody) comprises a sequence that includes at least one amino acid alteration with respect to a wild-type heavy or light chain sequence of the P140, iMab (or the MV1 variant), 515H7 antibody. In exemplary embodiments, the bispecific antibody (e.g., HIV CrossMab antibody) comprises a sequence that includes at least one amino acid alteration with respect to SEQ ID NOs: 1-36.

In various embodiments, the bispecific antibody (e.g., HIV CrossMab antibody) comprises a sequence that includes at least about 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 40, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, or 80 amino acid alterations with respect to a wild-type heavy or light chain sequence of the PGT145, PG9, PGT128, PGT121, 10-1074, 3BNC117, VRC01, PGT151, 4E10, or 10E8 antibody.

In various embodiments, the bispecific antibody (e.g., HIV CrossMab antibody) comprises a sequence that includes at least about 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 40, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, or 80 amino acid alterations with respect to a wild-type heavy or light chain sequence of the P140, iMab (or the MV1 variant), or 515H7 antibody.

In exemplary embodiments, the bispecific antibody (e.g., HIV CrossMab antibody) comprises a sequence that includes at least about 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 40, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, or 80 amino acid alterations with respect to SEQ ID NOs: 1-36.

The amino acid alteration can be an amino acid deletion, insertion, substitution, or modification. In one embodiment, the amino acid alteration is an amino acid deletion. In another embodiment, the amino acid alteration is an amino acid substitution.

In various embodiments, the amino acid alteration may be in the Complementarity Determining Regions (CDRs) of the bispecific antibody (e.g., the CDR1, CDR2 or CDR3 regions). In another embodiment, the amino acid alteration may be in the framework regions (FWs) of the bispecific antibody (e.g., the FW1, FW2, FW3, or FW4 regions). In a further embodiment, the amino acid alteration may be in the joining regions (J regions) of the bispecific antibody (e.g., the J1, J2, J3, J4, J5, J6, or J7 regions).

Also provided herein are chimeric antibody derivatives of the bispecific antibodies, i.e., antibody molecules in which a portion of the heavy and/or light chain is identical with or homologous to corresponding sequences in antibodies derived from a particular species or belonging to a particular antibody class or subclass, while the remainder of the chain(s) is identical with or homologous to corresponding sequences in antibodies derived from another species or belonging to another antibody class or subclass, as well as fragments of such antibodies, so long as they exhibit the desired biological activity. For example, the bispecific antibody may include a heavy and/or light chain in which one or more CDRs or FWs derived from an antibody selected from a PGT145, PG9, PGT128, PGT121, 10-1074, 3BNC117, VRC01, PGT151, 4E10, 10E8, P140, iMab (or the MV1 variant), or 515H7 antibody are replaced with one or more CDRs or FWs derived from a different antibody selected from a PGT145, PG9, PGT128, PGT121, 10-1074, 3BNC117, VRC01, PGT151, 4E10, 10E8, P140, iMab (or the MV1 variant), or 515H7 antibody.

Modification of the amino acid sequence of recombinant binding protein is achieved using any known technique in the art e.g., site-directed mutagenesis or PCR based mutagenesis. Such techniques are described, for example, in Sambrook et al., Molecular Cloning: A Laboratory Manual, Cold Spring Harbor Press, Plainview, N.Y., 1989 and Ausubel et al., Current Protocols in Molecular Biology, John Wiley & Sons, New York, N.Y., 1989.

Methods for producing antibodies, such as those disclosed herein, are known in the art. For example, DNA molecules encoding light chain variable regions and/or heavy chain variable regions can be chemically synthesized using the sequence information provided herein. Synthetic DNA molecules can be ligated to other appropriate nucleotide sequences, including, e.g., expression control sequences, to produce conventional gene expression constructs encoding the desired antibodies. Production of defined gene constructs is within routine skill in the art. Alternatively, the sequences provided herein can be cloned out of hybridomas by conventional hybridization techniques or polymerase chain reaction (PCR) techniques, using synthetic nucleic acid probes whose sequences are based on sequence information provided herein, or prior art sequence information regarding genes encoding the heavy and light chains.

Nucleic acids encoding desired antibodies can be incorporated (ligated) into expression vectors, which can be introduced into host cells through conventional transfection or transformation techniques. Exemplary host cells are E. coli cells, Chinese hamster ovary (CHO) cells, human embryonic kidney 293 (HEK 293) cells, HeLa cells, baby hamster kidney (BHK) cells, monkey kidney cells (COS), human hepatocellular carcinoma cells (e.g., Hep G2), and myeloma cells that do not otherwise produce IgG protein. Transformed host cells can be grown under conditions that permit the host cells to express the genes that encode the immunoglobulin light and/or heavy chain variable regions. Specific expression and purification conditions will vary depending upon the expression system employed.

In various embodiments, the bispecific antibodies of the present invention (e.g., HIV CrossMab antibodies) are used in therapy. For example, the bispecific antibody (e.g., HIV CrossMab antibody) can be used to neutralize HIV in a mammal (e.g., a human patient). For example, antibodies of the invention can bind to HIV so as to partially or completely inhibit one or more biological activities of the virus. In an embodiment, the bispecific antibody (e.g., HIV CrossMab antibody) neutralizes a R5-tropic HIV. In another embodiment, the bispecific antibody (e.g., HIV CrossMab antibody) neutralizes a X4-tropic HIV. In a further embodiment, the bispecific antibody (e.g., HIV CrossMab antibody) neutralizes a R5X4 dual-tropic HIV. In some embodiments, use of the antibody to neutralize HIV in a mammal comprises administering to the mammal a therapeutically effective amount of the antibody.

Generally, a therapeutically effective amount of active component is in the range of, for example, about 0.1 mg/kg to about 100 mg/kg, e.g., about 1 mg/kg to about 100 mg/kg, e.g., about 1 m/kg to about 10 mg/kg of the body weight of the patient. In various embodiments, a therapeutically effective amount of active component is in a range of about 0.01 mg/kg to about 10 mg/kg of the body weight of the patient, for example, about 0.01 mg/kg, about 0.02 mg/kg, about 0.03 mg/kg, about 0.04 mg/kg, about 0.05 mg/kg, about 0.06 mg/kg, about 0.07 mg/kg, about 0.08 mg/kg, about 0.09 mg/kg, about 0.1 mg/kg, about 0.2 mg/kg, about 0.3 mg/kg, about 0.4 mg/kg, about 0.5 mg/kg, about 0.6 mg/kg, about 0.7 mg/kg, about 0.8 mg/kg, about 0.9 mg/kg, about 1 mg/kg, about 1.1 mg/kg, about 1.2 mg/kg, about 1.3 mg/kg, about 1.4 mg/kg, about 1.5 mg/kg, about 1.6 mg/kg, about 1.7 mg/kg, about 1.8 mg/kg, 1.9 mg/kg, about 2 mg/kg, about 3 mg/kg, about 4 mg/kg, about 5 mg/kg, about 6 mg/kg, about 7 mg/kg, about 8 mg/kg, about 9 mg/kg, about 10 mg/kg body weight, inclusive of all values and ranges therebetween.

The amount administered will depend on variables such as the type and extent of disease or indication to be treated, the overall health of the patient, the in vivo potency of the antibody, the pharmaceutical formulation, and the route of administration. The initial dosage can be increased beyond the upper level in order to rapidly achieve the desired blood-level or tissue level. Alternatively, the initial dosage can be smaller than the optimum, and the dosage may be progressively increased during the course of treatment. Human dosage can be optimized, e.g., in a conventional Phase I dose escalation study designed to run from, for example, 0.5 mg/kg to 20 mg/kg. Dosing frequency can vary, depending on factors such as route of administration, dosage amount and the disease being treated. Exemplary dosing frequencies are more than once daily, about once per day, about twice a day, about three times a day, about four times a day, about five times a day, about every other day, about every third day, about once a week, about once every two weeks, about once every month, about once every two months, about once every three months, about once every six months, or about once every year. Formulation of antibody-based drugs is within ordinary skill in the art.

For therapeutic use, an antibody may be combined with a pharmaceutically acceptable carrier. As used herein, “pharmaceutically acceptable carrier” means buffers, carriers, and excipients suitable for use in contact with the tissues of human beings and animals without excessive toxicity, irritation, allergic response, or other problem or complication, commensurate with a reasonable benefit/risk ratio. The carrier(s) should be “acceptable” in the sense of being compatible with the other ingredients of the formulations and not deleterious to the recipient. Pharmaceutically acceptable carriers include buffers, solvents, dispersion media, coatings, isotonic and absorption delaying agents, and the like, that are compatible with pharmaceutical administration. The use of such media and agents for pharmaceutically active substances is known in the art.

Pharmaceutical compositions containing antibodies, such as those disclosed herein, can be presented in a dosage unit form and can be prepared by any suitable method. A pharmaceutical composition should be formulated to be compatible with its intended route of administration. Examples of routes of administration are intravenous (IV), intradermal, inhalation, transdermal, topical, transmucosal, and rectal administration. In an embodiment, the route of administration for antibodies of the invention is IV infusion.

Useful formulations can be prepared by methods well known in the pharmaceutical art. For example, see Remington's Pharmaceutical Sciences, 18th ed. (Mack Publishing Company, 1990). Formulation components suitable for parenteral administration include a sterile diluent such as water for injection, saline solution, fixed oils, polyethylene glycols, glycerine, propylene glycol or other synthetic solvents; antibacterial agents such as benzyl alcohol or methyl paraben; antioxidants such as ascorbic acid or sodium bisulfite; chelating agents such as EDTA; buffers such as acetates, citrates or phosphates; and agents for the adjustment of tonicity such as sodium chloride or dextrose. For intravenous administration, suitable carriers include physiological saline, bacteriostatic water, Cremophor ELTM (BASF, Parsippany, N.J.) or phosphate buffered saline (PBS). The carrier should be stable under the conditions of manufacture and storage, and should be preserved against microorganisms. The carrier can be a solvent or dispersion medium containing, for example, water, ethanol, polyol (for example, glycerol, propylene glycol, and liquid polyetheylene glycol), and suitable mixtures thereof.

Pharmaceutical formulations preferably are sterile. Sterilization can be accomplished, for example, by filtration through sterile filtration membranes. Where the composition is lyophilized, filter sterilization can be conducted prior to or following lyophilization and reconstitution.

FIGS. 13 and 14 demonstrate that some iMab-based CrossMabs have greater potency and breadth than parental Abs. Except otherwise stated, all iMab-based bispecific antibodies were constructed using the MV1 variant. IC80, the antibody concentration that confers 80% neutralization of viral infectivity, is one method to evaluate antibody potency against HIV. The lower the IC80 number (indicated in the y-axis of the graphs in term of antibody concentration (μm/ml)), the more potent the antibody is at neutralizing a particular HIV strain or isolate. IC50, the antibody concentration that confers 50% neutralization of viral infectivity, is another method to evaluate antibody potency against HIV. The lower the IC50 number (indicated in the y-axis of the graphs in term of antibody concentration (μm/ml)), the more potent the antibody is at neutralizing a particular HIV strain or isolate.

Various sets of antibodies were tested against a large panel of HIV-1 pseudoviruses (118 different HIV viral isolates) representative of HIV envelope diversity by geography, Glade, tropism, and stage of infection. IC80 and IC50 were used to evaluate the strength of antiviral potency and breadth. The bottom right panels in FIGS. 13 and 14 clearly demonstrate that, as compared to the parental antibodies iMab and 10E8, the bispecific CrossMab of the two together (10E8/iMab) neutralizes almost all HIV viruses (each virus is indicated as a dot) more potently. The other antibody sets (used to make 145/iMab, 117/iMab, 128/iMab and 151/iMab) sometimes enhance HIV potency compared to their parental components and sometimes do not.

As shown in FIG. 15, the antibody iMab is also relatively potent in cell-cell neutralizing assays. PGT145, 3BNC117, 10E8, PGT128 and PGT151 are relatively potent at neutralizing cell-free viral infection, but are poor in neutralizing viruses in cell-cell transmission assays. Creating bispecific antibodies including PGT145, 3BNC117, 10E8, PGT128 and PGT151 with iMab makes these chimeric antibodies active at neutralizing viruses in a cell-cell transmission assay. It can be seen that 10E8/iMab is the most potent antibody in these comparative studies. It is also found that 10E8/iMab is most active in preventing cell-cell transmission in vitro.

As illustrated in FIG. 16, the improved potency of 10E8/iMab is statistically significant. FIG. 3 shows that improved potency requires covalent linkage of the antibody, i.e., the CrossMab format (since co-administration of two parental antibodies, iMab and 10E8, provides a lower MPI than the fused and physically linked bispecific 10E8/iMab antibody). FIGS. 10-14 provide further evidence of the improved potency of iMab-derived CrossMab antibodies over its parental antibodies.

In summary, it is found that, for the iMab-based CrossMabs (fused with PGT145, 3BNC117, PGT151, PGT128 and 10E8), 117/iMab improves breadth but not potency; 145/iMab, 151/iMab and 128/iMab improve breadth and potency; and 10E8/iMab markedly improves breadth and potency. In terms of epitope location/accessibility and potential models of neutralization, 10E8/iMab appears to exhibit pre- and post-attachment neutralization; 145/iMab, 151/iMab and 117/iMab appear to exhibit pre-attachment neutralization; and 117/iMab may show signs of steric restriction and potentially reduced potency for some viruses. 10E8/iMab also exhibits potent activity against HIV cell-to-cell transmission.

As also shown in the top panels and bottom left panels of FIGS. 3 and 14, Pro 140-based CrossMab activities are sometimes weaker than their parental antibodies and corresponding iMab-based CrossMabs, as shown by the high concentrations required to reach IC80 and IC50. Anchoring of these four mAbs to the host cell receptor CCR5 via another host cell receptor-binding antibody called Pro 140 does not improve the antiviral potency or breadth (as measured by IC80 against a large panel of HIV isolates) compared to their respective parental antibodies. These panels indicate that Pro140-based CrossMabs for these four antibodies are weaker than their corresponding iMab-based CrossMabs (IC50 and IC80 comparisons of Pro140-based vs. iMab-based CrossMabs).

As shown in the bottom right panels of FIGS. 13 and 14, 10E8/P140, a fifth Pro 140-based CrossMab, is more potent than its parental antibodies and 10E8/iMab CrossMab. These panels illustrate a comparison of the potency (IC80 or IC50) of parental mAb Pro140 (right-most column of data points in these panels), bispecific CrossMab 10E8/P140 (second from right column of data points in these panels), and parental mAb 10E8 (center column of data points in these panels) against a large panel of HIV isolates. These panels also illustrate a comparison of the potency (IC80 or IC50) of parental mAb iMab (left-most column of data points in these panels), bispecific CrossMab 10E8/iMab (second from left column of data points in these panels), and parental mAb 10E8 (center column of data points in these panels) against a large panel of HIV isolates. The second from left and second from right columns of data points in these panels illustrate a comparison of the potency (IC80 or IC50) of the bispecific CrossMabs 10E8/iMab and 10E8/P140 against a large panel of HIV isolates.

Pro 140 is known to not have activity against X4 HIV viruses, as X4 viruses use CXCR4 as a co-receptor for HIV-1 entry, and Pro 140 binds to CCR5. 10E8 alone has very weak activity against X4 viruses. However, the bispecific CrossMab 10E8/Pro 140 can neutralize all X4 viruses tested to date better than either of the parent antibodies. The panels of FIG. 4 illustrate the effectiveness of 10E8, Pro 140, and 10E8/P140 bispecific CrossMab antibody in inhibition of various strains of HIV.

As shown in FIG. 5, 10E8/Pro140 CrossMab is a more potent inhibitor of various strains of HIV than the co-administration of the two parental antibodies, demonstrating a synergistic, not merely additive, enhancement of potency with this particular bispecific antibody.

As shown in FIG. 6, a CrossMab of 10E8 fused to a non-membrane bound antibody (X19) does not provide enhanced potency, as can be seen when compared to membrane bound 10E8/P140. Thus, the potency of the 10E8/P140 CrossMab appears to require anchoring of 10E8 to the cell membrane. However, membrane binding alone does not afford the enhanced potency of these CrossMabs. FIG. 7 shows that anchoring 10E8 on HER2 does not provide substantial potency enhancement as compared to anchoring 10E8 on CCR5. Anchoring of 10E8 to a viral receptor specifically (in this case CCR5 via Pro 140 or CD4 via iMab) provides enhanced antiviral activity.

Δ10E8 is a mutant version of the 10E8 mAb that has a one amino acid deletion in the light chain FR3. Compared to 10E8, Δ10E8 has a much weaker epitope binding activity, as illustrated in FIG. 8A and panels A and B in FIG. 8B. However, once the Δ10E8 was anchored on a cell receptor (by combining Δ10E8 and iMab in a CrossMab antibody—iMab specifically binds cell receptor CD4), FIG. 8B, panels C and D, show that its inhibition activity is improved. These data suggest the contribution of specific cell receptor anchoring, i.e., anchoring on a viral receptor or a viral co-receptor, in enhancing the activity of this HIV antibody. Still, while Δ10E8/P140 CrossMab has improved antiviral activity over Δ10E8, it is still not as potent as 10E8/P140 CrossMab. Δ10E8/P140 CrossMab is comparatively more effective in neutralizing R5 viruses than it is in neutralizing X4 viruses.

4E10 is an anti-gp41 MPER mAb known to be less potent than the anti-gp41 MPER mAb 10E8. Similar to the results for Δ10E8, FIGS. 20 and 21 show that anchoring 4E10 on co-receptor CCR5 (via Pro 140 in a CrossMab antibody) enhanced antiviral activity of 4E10 significantly. Taken together, this suggests that the anchoring of a number of anti-gp41 MPER Abs to either CCR5 or CD4 (via combining the MPER Abs with P140 or iMab in a CrossMab antibody bispecific) can greatly improve the potency and breadth of the respective anti-gp41 MPER Ab.

Multiple parameters contribute to enhanced activity of certain bispecific CrossMabs against HIV, including parental Ab potency, affinity, and pre- and post-attachment neutralization abilities. In particular, the 10E8/Pro140 CrossMab represents an effective combination in terms of overcoming energetic, spatial and temporal constraints, targeting sequential/interdependent steps in the entry process, epitope location/accessibility, binding affinity, pre- and post-attachment neutralization, and binding geometry. As shown in FIG. 20, 4E10/Pro140 has a greater binding affinity for MPER than Δ10E8/Pro140 and 10E8/Pro140. FIG. 9 shows the inhibition potency of 10E8/Pro140, Δ10E8/Pro140 and 4E10/Pro140, and their parental antibodies 10E8, Δ10E8 and 4E10 against various strains of HIV. FIGS. 10 and 13-17 provide additional evidence of the greater potency of CrossMab antibodies as compared to their parental antibodies individually and the parental antibodies in combination.

The enhanced antiviral coverage of 10E8/iMab and 10E8/Pro140 CrossMabs is illustrated in FIG. 10, which depict the potency and breadth of several antibodies against HIV. The x-axis indicates the concentration of a particular antibody, the y-axis indicates the percent of a large panel of HIV viral isolates neutralized by a particular antibody at a specific concentration, and each line indicates a different antibody evaluated. The left-most lines along the x-axis and those that can closely approach or reach 100% on the y-axis indicate a highly potent and broad antibody against HIV. 10E8/P140 CrossMab and 10E8/iMab CrossMab are among the most effective antibodies with respect to both viral coverage and potency, and are significantly more effective than their parental antibodies.

FIGS. 18 and 19 show the potency of the CrossMab 10E8/515H7 antibody as compared to its parental antibodies and previously discussed antibodies. The potency of a CrossMab antibody does not appear to correlate directly with the density of cell membrane protein targets, as the density of CCR5 (the target of Pro140) is less than that of CD4 (the target of ibalizumab), yet the potency of 10E8/Pro140-derived CrossMab antibody is greater than that of 10E8/iMab-derived CrossMab antibody.

As shown in FIG. 22, the lack of single, sharp peaks in size exclusion chromatography indicates a type of instability indicative of multiple molecular species for 10E8 and 10E8-derived CrossMab antibodies. Table 1 recites various process and formulation modifications used to resolve the 10E8 instability. However, as indicated by the “X” in the SEC or Size Exclusion Chromatography column, the modifications were unsuccessful in providing a single, sharp peak.

Pairing the 10E8 heavy chain with the light chain of 4E10 resolves the instability issue, as shown in FIG. 23, producing a functional, though less potent, antibody. This result indicates that the instability of 10E8 is due to the light chain. Various modifications of the 10E8 light chain, shown in the center and bottom panels of FIG. 24, such as removal of a C-terminal serine, engineering a lambda-variable region and kappa-constant region chimera, engineering an additional disulfide bond between the 10E8 heavy and light chains, or genetically grafting kappa light chain regions of non-10E8 antibodies onto the 10E8 light chain do not fully resolve 10E8 instability. As shown in FIGS. 25 and 26, the instability is likely due to a combination of the Complementarity Determining Regions (“CDRs”) and the framework regions (“FWs”) of 10E8. Using 10E8-HC/4E10-LC, each 10E8 light chain CDR was grafted into 4E10LC individually or in concert, as shown in FIG. 25. Addition of 10E8 LC CDR2 and CDR3 are well tolerated, but addition of 10E8 LC CDR1 disrupts the single peak. When all 10E8 CDRs are grafted onto 4E10, the peak is broad. Grafting 10E8 CDRs or frameworks onto its germline light chain λ results in a single peak, as shown in FIG. 26, but effectiveness in MPER binding and HIV neutralization is decreased. Table 2 summarizes the 10E8 light chain variants tested and the efficacy thereof

Variant H6L10 of 10E8 antibody was found to be active, non-autoreactive, and stable by size exclusion chromatography, as shown in FIG. 27. H6L10/Pro 140 and its parental antibodies were found to have comparable pharmacokinetics profiles in mice, as shown in FIG. 28. However, as shown in FIG. 29, the H6L10 variant of 10E8 (referred to as 10E8_{v 1.0}) combined with P140 in a bispecific antibody is substantially less potent than 10E8/P140 when tested against a large panel of HIV strains. The H6L10 variant of 10E8 (referred to as 10E8_{v 1.0}) combined with iMab in a bispecific antibody retains the same relative amount of potency as compared to 10E8/iMab when tested against a large panel of HIV strains, but 10E8_{v 1.0}/iMab possesses the same instability as 10E8/iMab as determined by size exclusion chromatography and indicated by an X in Table 3. In an embodiment, the H6L10 variant may further include a S74W mutation.

Table 3 below lists exemplary variants, their activities, size exclusion chromatography results, and pharmacokinetics (“PK”) results.

As indicated above, 10E8_V1.0is a somatic variant of 10E8 known as H6L10. As a mAb, H6L10 has a single peak by SEC but reduced activity compared to 10E8. H6L10/Pro140 CrossMab has single SEC peak and good mouse PK, but reduced anti-HIV activity. H6L10/iMab CrossMab has double SEC peaks and poor mouse PK, but its activity against HIV is roughly the same as 10E8/iMab. 10E8_V1.1includes a single point mutation in H6L10. When paired with Pro 140 in a CrossMab bispecific, this construct has single SEC peak and good mouse PK. Its activity against HIV is improved as compared to 10E8V1.0/Pro140, but still slightly less than that of 10E8/Pro140. When paired with iMab in a CrossMab bispecific, this construct has double SEC peaks and poor mouse PK, and its activity against HIV is still roughly the same as 10E8/iMab and 10E8V1.0/iMab. 10E8_V2.0is a chimeric antibody variant of 10E8 in which the FW1, CDR1 and part of FW2 are from 10E8_V1.0and in which the remaining part of FW2, CDR2, FW3, CDR3 and FW4 are from 10E8. When paired with Pro140 in a CrossMab bispecific, this construct has double SEC peaks and has reduced activity against HIV as compared to 10E8/Pro140. When paired with iMab in a CrossMab bispecific, this construct has a single SEC peak, good PK, and activity against HIV that is improved over 10E8/iMab. 10E8_V3.0is a somatic variant of 10E8 known as H11L1. H11L1/Pro140 CrossMab has a single SEC peak and better anti-HIV activity than any other 10E8/Pro140 construct (including the original one identified), but has poor mouse PK due to autoreactivity. H11L1/iMab CrossMab has a single SEC peak and anti-HIV activity that is better than the original 10E8/iMab identified and roughly equivalent activity to that observed for 10E8V2.0/iMab, but has poor mouse PK due to autoreactivity.

The variant of 10E8 that produced a single SEC peak in the context of a particular CrossMab bispecific was different when paired with Pro140 or iMab. It appears that the stability of the 10E8 arm of these CrossMab bispecific antibodies is context dependent and will vary depending of what antibody it is paired with. Thus, one variant (10E8_V1.1) was identified that was stable by SEC and with good mouse PK and good anti-HIV activity when paired with Pro140. Another variant (10E8_V2.0) was also identified that was stable by SEC with good mouse PK and with better anti-HIV activity than the originally identified 10E8/iMab.

Table 4 below describes the autoreactivity of tested variants, where “ANA” refers to anti-nuclear activity and “ACA” refers to anti-cardiolipin activity.

FIG. 30 depicts the pharmacokinetics of 10E8 and CrossMab antibodies derived from several 10E8 variants and iMab or P140 in a mouse model. As shown in FIGS. 31 and 32, 10E8_v1.1/P140 and 10E8_v2.0/iMab improve anti-HIV activity and stability, and have good stability when stored in PBS at 4° C. FIG. 33 depicts a native mass spectroscopy analysis of 10E8_v2.0/iMab (N297A). FIG. 34 compares the activity of 10E8_v1.1/P140 and 10E8_v2.0/iMab on a HIV Clade C panel, and compares their IC50 and IC80 efficacy. FIGS. 35 and 36 compare the potency of 10E8_v1.1/P140, 10E8_v2.0/iMab, and various monoclonal antibodies against HIV.

The terms and expressions employed herein are used as terms and expressions of description and not of limitation, and there is no intention, in the use of such terms and expressions, of excluding any equivalents of the features shown and described or portions thereof. In addition, having described certain embodiments of the invention, it will be apparent to those of ordinary skill in the art that other embodiments incorporating the concepts disclosed herein may be used without departing from the spirit and scope of the invention. Accordingly, the described embodiments are to be considered in all respects as only illustrative and not restrictive.