MICE THAT MAKE VL BINDING PROTEINS
20220177606 · 2022-06-09
Inventors
- Lynn Macdonald (Harrison, NY)
- Sean Stevens (Del Mar, CA)
- Cagan Gurer (Chappaqua, NY)
- Karolina A. Meagher (Yorktown Heights, NY, US)
- Andrew J. Murphy (Croton-on-Hudson, NY)
Cpc classification
A01K67/0275
HUMAN NECESSITIES
A01K2267/01
HUMAN NECESSITIES
C07K2317/14
CHEMISTRY; METALLURGY
C07K16/00
CHEMISTRY; METALLURGY
C12N15/8509
CHEMISTRY; METALLURGY
C07K16/461
CHEMISTRY; METALLURGY
C07K2317/92
CHEMISTRY; METALLURGY
A01K67/0278
HUMAN NECESSITIES
A01K2217/072
HUMAN NECESSITIES
C12N2015/8518
CHEMISTRY; METALLURGY
International classification
C07K16/00
CHEMISTRY; METALLURGY
Abstract
Genetically modified mice and methods for making an using them are provided, wherein the mice comprise a replacement of all or substantially all immunoglobulin heavy chain V gene segments, D gene segments, and J gene segments with at least one light chain V gene segment and at least one light chain J gene segment. Mice that make binding proteins that comprise a light chain variable domain operably linked to a heavy chain constant region are provided. Binding proteins that contain an immunoglobulin light chain variable domain, including a somatically hypermutated light chain variable domain, fused with a heavy chain constant region, are provided. Modified cells, embryos, and mice that encode sequences for making the binding proteins are provided.
Claims
1. A mouse whose germline genome comprises an endogenous immunoglobulin (Ig) heavy chain locus modified to comprise a human genomic germline kappa (κ) sequence comprising (i) unrearranged functional human Ig light chain variable κ (hVκ) gene segments and (ii) all five unrearranged functional human Ig light chain joining κ (hJκ1-hJκ5) gene segments, wherein the human genomic germline κ sequence (A) replaces at the endogenous Ig heavy chain locus an endogenous genomic sequence comprising endogenous immunoglobulin heavy chain V gene segments, all endogenous immunoglobulin heavy chain D gene segments, and all endogenous immunoglobulin heavy chain J gene segments, and (B) rearranges in a B cell during B cell development to form a rearranged Ig hVκ/hJκ gene sequence operably linked to the endogenous Ig heavy chain constant region (C.sub.H) nucleic acid sequence at the endogenous Ig heavy chain locus, and wherein the mouse comprises a CD19.sup.+ B cell comprising the rearranged Ig hVκ/hJκ gene sequence operably linked to the endogenous Ig C.sub.H nucleic acid sequence.
2.-14. (canceled)
15. A mouse that expresses from its germline an immunoglobulin comprising a first polypeptide comprising a first human light chain variable region sequence fused with an immunoglobulin heavy chain constant region, and a second polypeptide comprising a second human light chain variable region fused with an immunoglobulin light chain constant region.
16.-21. (canceled)
22. A method for making a binding protein comprising obtaining a nucleotide sequence encoding a Vκ domain from a gene encoding a Vκ domain fused to a C.sub.H region from a cell of the mouse of claim 1, cloning the nucleotide sequence encoding the Vκ domain sequence in frame with a gene encoding a human C.sub.H region to form a human binding protein sequence, and expressing the human binding protein sequence in a suitable cell.
23. An antigen-binding protein comprising a first dimer that comprises: (i) a first polypeptide comprising a first human immunoglobulin (Ig) light chain variable domain (V.sub.L1) fused to a first Ig heavy chain constant region, and (ii) a second polypeptide comprising a second human Ig light chain variable domain (V.sub.L2) fused to a first Ig light chain constant region, wherein the V.sub.L1 of (i) and the V.sub.L2 of (ii) are cognate and not identical, and wherein first Ig heavy chain constant region of (i) and first Ig light chain constant region of (ii) associate such that the first dimer comprises a first Fab which includes the V.sub.L1 and the V.sub.L2, and wherein the first Fab specifically binds an antigen of interest.
Description
BRIEF DESCRIPTION OF THE FIGURES
[0132]
[0133]
[0134]
[0135]
[0136]
[0137]
[0138]
[0139]
[0140]
[0141]
[0142]
[0143]
[0144]
[0145]
[0146]
[0147]
[0148]
[0149]
[0150]
[0151]
[0152]
DETAILED DESCRIPTION
[0153] The phrase “bispecific binding protein” includes a binding protein capable of selectively binding two or more epitopes. Bispecific binding proteins comprise two different polypeptides that comprise a first light chain variable domain (V.sub.L1) fused with a first C.sub.H region and a second light chain variable domain (V.sub.L2) fused with a second C.sub.H region. In general, the first and the second C.sub.H regions are identical, or they differ by one or more amino acid substitutions (e.g., as described herein). V.sub.L1 and V.sub.L2 specifically binding different epitopes—either on two different molecules (e.g., antigens) or on the same molecule (e.g., on the same antigen). If a bispecific binding protein selectively binds two different epitopes (a first epitope and a second epitope), the affinity of V.sub.L1 for the first epitope will generally be at least one to two or three or four orders of magnitude lower than the affinity of V.sub.L1 for the second epitope, and vice versa with respect to V.sub.L2. The epitopes recognized by the bispecific binding protein can be on the same or a different target (e.g., on the same or a different antigen). Bispecific binding proteins can be made, for example, by combining a V.sub.L1 and a V.sub.L2 that recognize different epitopes of the same antigen. For example, nucleic acid sequences encoding V.sub.L sequences that recognize different epitopes of the same antigen can be fused to nucleic acid sequences encoding different C.sub.H regions, and such sequences can be expressed in a cell that expresses an immunoglobulin light chain, or can be expressed in a cell that does not express an immunoglobulin light chain. A typical bispecific binding protein has two heavy chains each having three light chain CDRs, followed by (N-terminal to C-terminal) a C.sub.H1 domain, a hinge, a C.sub.H2 domain, and a C.sub.H3 domain, and an immunoglobulin light chain that either does not confer antigen-binding specificity but that can associate with each heavy chain, or that can associate with each heavy chain and that can bind one or more of the epitopes bound by V.sub.L1 and/or V.sub.L2, or that can associate with each heavy chain and enable binding or assist in binding of one or both of the heavy chains to one or both epitopes.
[0154] Therefore, two general types of bispecific binding proteins are (1) V.sub.L1-C.sub.H(dimer), and (2) V.sub.L1-C.sub.H:light chain+V.sub.L2-C.sub.H:light chain, wherein the light chain is the same or different. In either case, the C.sub.H (i.e., the heavy chain constant region) can be differentially modified (e.g., to differentially bind protein A, to increase serum half-life, etc.) as described herein, or can be the same.
[0155] The term “cell,” when used in connection with expressing a sequence, includes any cell that is suitable for expressing a recombinant nucleic acid sequence. Cells include those of prokaryotes and eukaryotes (single-cell or multiple-cell), bacterial cells (e.g., strains of E. coli, Bacillus spp., Streptomyces spp., etc.), mycobacteria cells, fungal cells, yeast cells (e.g., S. cerevisiae, S. pombe, P. pastoris, P. methanolica, etc.), plant cells, insect cells (e.g., SF-9, SF-21, baculovirus-infected insect cells, Trichoplusia ni, etc.), non-human animal cells, human cells, B cells, or cell fusions such as, for example, hybridomas or quadromas. In some embodiments, the cell is a human, monkey, ape, hamster, rat, or mouse cell. In some embodiments, the cell is eukaryotic and is selected from the following cells: CHO (e.g., CHO K1, DXB-11 CHO, Veggie-CHO), COS (e.g., COS-7), retinal cell, Vero, CV1, kidney (e.g., HEK293, 293 EBNA, MSR 293, MDCK, HaK, BHK), HeLa, HepG2, WI38, MRC 5, Colo205, HB 8065, HL-60, (e.g., BHK21), Jurkat, Daudi, A431 (epidermal), CV-1, U937, 3T3, L cell, C127 cell, SP2/0, NS-0, MMT 060562, Sertoli cell, BRL 3A cell, HT1080 cell, myeloma cell, tumor cell, and a cell line derived from an aforementioned cell. In some embodiments, the cell comprises one or more viral genes, e.g. a retinal cell that expresses a viral gene (e.g., a PER.C6™ cell).
[0156] The term “cognate,” when used in the sense of “cognate with,” e.g., a first V.sub.L domain that is “cognate with” a second V.sub.L domain, is intended to include reference to the relation between two V.sub.L domains from a same binding protein made by a mouse in accordance with the invention. For example, a mouse that is genetically modified in accordance with an embodiment of the invention, e.g., a mouse having a heavy chain locus in which V.sub.H, D.sub.H, and J.sub.H regions are replaced with V.sub.L and J.sub.L regions, makes antibody-like binding proteins that have two identical polypeptide chains made of the same mouse C.sub.H region (e.g., an IgG isotype) fused with a first human V.sub.L domain, and two identical polypeptide chains made of the same mouse C.sub.L region fused with a second human V.sub.L domain. During clonal selection in the mouse, the first and the second human V.sub.L domains were selected by the clonal selection process to appear together in the context of a single antibody-like binding protein. Thus, first and second V.sub.L domains that appear together, as the result of the clonal selection process, in a single antibody-like molecule are referred to as being “cognate.” In contrast, a V.sub.L domain that appears in a first antibody-like molecule and a V.sub.L domain that appears in a second antibody-like molecule are not cognate, unless the first and the second antibody-like molecules have identical heavy chains (i.e., unless the V.sub.L domain fused to the first human heavy chain region and the V.sub.L domain fused to the second human heavy chain region are identical).
[0157] The phrase “complementarity determining region,” or the term “CDR,” includes an amino acid sequence encoded by a nucleic acid sequence of an organism's immunoglobulin genes that normally (i.e., in a wild-type animal) appears between two framework regions in a variable region of a light or a heavy chain of an immunoglobulin molecule (e.g., an antibody or a T cell receptor). A CDR can be encoded by, for example, a germline sequence or a rearranged or unrearranged sequence, and, for example, by a naïve or a mature B cell or a T cell. In some circumstances (e.g., for a CDR3), CDRs can be encoded by two or more sequences (e.g., germline sequences) that are not contiguous (e.g., in an unrearranged nucleic acid sequence) but are contiguous in a B cell nucleic acid sequence, e.g., as the result of splicing or connecting the sequences (e.g., V-D-J recombination to form a heavy chain CDR3).
[0158] The phrase “gene segment,” or “segment” includes reference to a V (light or heavy) or D or J (light or heavy) immunoglobulin gene segment, which includes unrearranged sequences at immunoglobulin loci (in e.g., humans and mice) that can participate in a rearrangement (mediated by, e.g., endogenous recombinases) to form a rearranged V/J or V/D/J sequence. Unless indicated otherwise, the V, D, and J segments comprise recombination signal sequences (RSS) that allow for V/J recombination or V/D/J recombination according to the 12/23 rule. Unless indicated otherwise, the segments further comprise sequences with which they are associated in nature or functional equivalents thereof (e.g., for V segments promoter(s) and leader(s)).
[0159] The phrase “heavy chain,” or “immunoglobulin heavy chain” includes an immunoglobulin heavy chain constant region sequence from any organism, and unless otherwise specified includes a heavy chain variable domain (V.sub.H). V.sub.H domains include three heavy chain CDRs and four framework (FR) regions, unless otherwise specified. Fragments of heavy chains include CDRs, CDRs and FRs, and combinations thereof. A typical heavy chain consists essentially of, following the variable domain (from N-terminal to C-terminal), a C.sub.H1 domain, a hinge, a C.sub.H2 domain, a C.sub.H3 domain, and optionally a C.sub.H4 domain (e.g., in the case of IgM or IgE) and a transmembrane (M) domain (e.g., in the case of membrane-bound immunoglobulin on lymphocytes). A heavy chain constant region is a region of a heavy chain that extends (from N-terminal side to C-terminal side) from outside FR4 to the C-terminal of the heavy chain. Heavy chain constant regions with minor deviations, e.g., truncations of one, two, three or several amino acids from the C-terminal, would be encompassed by the phrase “heavy chain constant region,” as well as heavy chain constant regions with sequence modifications, e.g., 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 amino acid substitutions. Amino acid substitutions can be made at one or more positions selected from, e.g. (with reference to EU numbering of an immunoglobulin constant region, e.g., a human IgG constant region), 228, 233, 234, 235, 236, 237, 238, 239, 241, 248, 249, 250, 252, 254, 255, 256, 258, 265, 267, 268, 269, 270, 272, 276, 278, 280, 283, 285, 286, 289, 290, 292, 293, 294, 295, 296, 297, 298, 301, 303, 305, 307, 308, 309, 311, 312, 315, 318, 320, 322, 324, 326, 327, 328, 329, 330, 331, 332, 333, 334, 335, 337, 338, 339, 340, 342, 344, 356, 358, 359, 360, 361, 362, 373, 375, 376, 378, 380, 382, 383, 384, 386, 388, 389, 398, 414, 416, 419, 428, 430, 433, 434, 435, 437, 438, and 439.
[0160] For example, and not by way of limitation, a heavy chain constant region can be modified to exhibit enhanced serum half-life (as compared with the same heavy chain constant region without the recited modification(s)) and have a modification at position 250 (e.g., E or Q); 250 and 428 (e.g., L or F); 252 (e.g., L/Y/F/W or T), 254 (e.g., S or T), and 256 (e.g., S/R/Q/E/D or T); or a modification at 428 and/or 433 (e.g., L/R/SI/P/Q or K) and/or 434 (e.g., H/F or Y); or a modification at 250 and/or 428; or a modification at 307 or 308 (e.g., 308F, V308F), and 434. In another example, the modification can comprise a 428L (e.g., M428L) and 434S (e.g., N434S) modification; a 428L, 259I (e.g., V259I), and a 308F (e.g., V308F) modification; a 433K (e.g., H433K) and a 434 (e.g., 434Y) modification; a 252, 254, and 256 (e.g., 252Y, 254T, and 256E) modification; a 250Q and 428L modification (e.g., T250Q and M428L); a 307 and/or 308 modification (e.g., 308F or 308P).
[0161] The phrase “light chain” includes an immunoglobulin light chain constant (C.sub.L) region sequence from any organism, and unless otherwise specified includes human κ and λ light chains. Light chain variable (V.sub.L) domains typically include three light chain CDRs and four framework (FR) regions, unless otherwise specified. Generally, a full-length light chain (V.sub.L +C.sub.L) includes, from amino terminus to carboxyl terminus, a V.sub.L domain that includes FR1-CDR1-FR2-CDR2-FR3-CDR3-FR4, and a C.sub.L region. Light chains (V.sub.L+C.sub.L) that can be used with this invention include those, e.g., that do not selectively bind either a first or second (in the case of bispecific binding proteins) epitope selectively bound by the binding protein (e.g., the epitope(s) selectively bound by the V.sub.L domain fused with the C.sub.H domain). V.sub.L domains that do not selectively bind the epitope(s) bound by the V.sub.L that is fused with the C.sub.H domain include those that can be identified by screening for the most commonly employed light chains in existing antibody libraries (wet libraries or in silico), wherein the light chains do not substantially interfere with the affinity and/or selectivity of the epitope binding domains of the binding proteins. Suitable light chains include those that can bind (alone or in combination with its cognate V.sub.L fused with the C.sub.H region) an epitope that is specifically bound by the V.sub.L fused to the C.sub.H region.
[0162] The phrase “micromolar range” is intended to mean 1-999 micromolar; the phrase “nanomolar range” is intended to mean 1-999 nanomolar; the phrase “picomolar range” is intended to mean 1-999 picomolar.
[0163] The term “non-human animals” is intended to include any vertebrate such as cyclostomes, bony fish, cartilaginous fish such as sharks and rays, amphibians, reptiles, mammals, and birds. Suitable non-human animals include mammals. Suitable mammals include non-human primates, goats, sheep, pigs, dogs, cows, and rodents. Suitable non-human animals are selected from the rodent family including rat and mouse. In one embodiment, the non-human animals are mice.
[0164] Mice, Nucleotide Sequences, and Binding Proteins
[0165] Binding proteins are provided that are encoded by elements of immunoglobulin loci, wherein the binding proteins comprise immunoglobulin heavy chain constant regions fused with immunoglobulin light chain variable domains. Further, multiple strategies are provided to genetically modify an immunoglobulin heavy chain locus in a mouse to encode binding proteins that contain elements encoded by immunoglobulin light chain loci. Such genetically modified mice represent a source for generating unique populations of binding proteins that have an immunoglobulin structure, yet exhibit an enhanced diversity over traditional antibodies.
[0166] Binding protein aspects described herein include binding proteins that are encoded by modified immunoglobulin loci, which are modified such that gene segments that normally (i.e., in a wild-type animal) encode immunoglobulin light chain variable domains (or portions thereof) are operably linked to nucleotide sequences that encode heavy chain constant regions. Upon rearrangement of the light chain gene segments, a rearranged nucleotide sequence is obtained that comprises a sequence encoding a light chain variable domain fused with a sequence encoding a heavy chain constant region. This sequence encodes a polypeptide that has an immunoglobulin light chain variable domain fused with a heavy chain constant region. Thus, in one embodiment, the polypeptide consists essentially of, from N-terminal to C-terminal, a V.sub.L domain, a C.sub.H1 region, a hinge, a C.sub.H2 region, a C.sub.H3 region, and optionally a C.sub.H4 region.
[0167] In modified mice described herein, such binding proteins are made that also comprise a cognate light chain, wherein in one embodiment the cognate light chain pairs with the polypeptide described above to make a binding protein that is antibody-like, but the binding protein comprises a V.sub.L region—not a V.sub.H region—fused to a C.sub.H region.
[0168] In various embodiments, the modified mice make binding proteins that comprise a V.sub.L region fused with a C.sub.H region (a hybrid heavy chain), wherein the V.sub.L region of the hybrid heavy chain exhibits an enhanced degree of somatic hypermutation. In these embodiments, the enhancement is over a V.sub.L region that is fused with a C.sub.L region (a light chain). In some embodiments, a V.sub.L region of a hybrid heavy chain exhibits about 1.5-fold, 2-fold, 2.5-fold, 3-fold, 3.5-fold, 4-fold, 4.5-fold, or 5-fold or more somatic hypermutations than a V.sub.L region fused with a C.sub.L region. In some embodiments, the modified mice in response to an antigen exhibit a population of binding proteins that comprise a V.sub.L region of a hybrid heavy chain, wherein the population of binding proteins exhibits an average of about 1.5-fold, 2-fold, 2.5-fold, 3-fold, 3.5-fold, 4-fold, 4.5-fold, 5-fold or more somatic hypermutations in the V.sub.L region of the hybrid heavy chain than is observed in a wild-type mouse in response to the same antigen. In one embodiment, the somatic hypermutations in the V.sub.L region of the hybrid heavy chain comprise one or more or two or more N additions in a CDR3.
[0169] In various embodiments, the binding proteins comprise variable domains encoded by immunoglobulin light chain sequences that comprise a larger number of N additions than observed in nature for light chains rearranged from an endogenous light chain locus, e.g., a binding protein comprising a mouse heavy chain constant region fused with a variable domain derived from human light chain V gene segments and human (light or heavy) J gene segments, wherein the human V and human J segments rearrange to form a rearranged gene that comprises 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 or more N additions.
[0170] In various embodiments, the mice of the invention make binding proteins that are on average smaller than wild-type antibodies (i.e., antibodies that have a V.sub.H domain), and possess advantages associated with smaller size. Smaller size is realized at least in part through the absence of an amino acid sequence encoded by a D.sub.H region, normally present in a V.sub.H domain. Smaller size can also be realized in the formation of a CDR3 that is derived, e.g., from a Vκ region and a Jκ region.
[0171] In another aspect, a mouse and a method is provided for providing a population of binding proteins having somatically hypermutated V.sub.L domains, e.g., somatically mutated human Vκ domains, and, e.g., human Vκ domains encoded by rearranged κ variable genes that comprise 1-10 or more N additions. In one embodiment, in the absence of a V.sub.H region for generating antibody diversity, a mouse of the invention will generate binding proteins, e.g., in response to challenge with an antigen, whose V domains are only or substantially V.sub.L domains. The clonal selection process of the mouse therefore is limited to selecting only or substantially from binding proteins that have V.sub.L domains, rather than V.sub.H domains. Somatic hypermutation of the V.sub.L domains will be as frequent, or substantially more frequent (e.g., 2- to 5-fold higher, or more), than in wild-type mice (which also mutate V.sub.L domains with some frequency). The clonal selection process in a mouse of the invention will generate high affinity binding proteins from the modified immunoglobulin locus, including binding proteins that specifically bind an epitope with an affinity in the nanomolar or picomolar range. Sequences that encode such binding proteins can be used to make therapeutic binding proteins containing human variable regions and human constant regions using an appropriate expression system.
[0172] In other embodiments, a mouse according to the invention can be made wherein the mouse heavy chain and/or light chain immunoglobulin loci are disabled, rendered non-functional, or knocked out, and fully human or chimeric human-mouse transgenes can be placed in the mouse, wherein at least one of the transgenes contains a modified heavy chain locus (e.g., having light chain gene segments operably linked to one or more heavy chain gene sequences). Such a mouse may also make a binding protein as described herein.
[0173] In one aspect, a method is provided for increasing the diversity, including by somatic hypermutation or by N additions in a V.sub.L domain, comprising placing an unrearranged light chain V gene segment and an unrearranged J gene segment in operable linkage with a mouse C.sub.H gene sequence, exposing the animal to an antigen of interest, and isolating from the animal a rearranged and somatically hypermutated V(light)/J gene sequence of the animal, wherein the rearranged V(light)/J gene sequence is fused with a nucleotide sequence encoding an immunoglobulin C.sub.H region.
[0174] In one embodiment, the immunoglobulin heavy chain fused with the hypermutated V.sub.L is an IgM; in another embodiment, an IgG; in another embodiment, an IgE; in another embodiment, an IgA.
[0175] In one embodiment, the somatically hypermutated and class-switched V.sub.L domain contains about 2- to 5-fold or more of the somatic hypermutations observed for a rearranged and class-switched antibody having a V.sub.L sequence that is operably linked to a C.sub.L sequence. In one embodiment, the observed somatic hypermutations in the somatically hypermutated V.sub.L domain are about the same in number as observed in a V.sub.H domain expressed from a V.sub.H gene fused to a C.sub.H region.
[0176] In one aspect, a method for making a high-affinity human V.sub.L domain is provided, comprising exposing a mouse of the invention to an antigen of interest, allowing the mouse to develop an immune response to the antigen of interest, and isolating a somatically mutated, class-switched human V.sub.L domain from the mouse that specifically binds the antigen of interest with high affinity.
[0177] In one embodiment, the K.sub.D of a binding protein comprising the somatically mutated, class-switched human V.sub.L domain is in the nanomolar or picomolar range.
[0178] In one embodiment, the binding protein consists essentially of a polypeptide dimer, wherein the polypeptide consists essentially of the somatically mutated, class-switched binding protein comprising a human V.sub.L domain fused with a human C.sub.H region.
[0179] In one embodiment, the binding protein consists essentially of a polypeptide dimer and two light chains, wherein the polypeptide consists essentially of the somatically mutated, class-switched binding protein having a human V.sub.L domain fused with a human C.sub.H region; and wherein each polypeptide of the dimer is associated with a cognate light chain comprising a cognate light chain V.sub.L domain and a human C.sub.L region.
[0180] In one aspect, a method is provided for somatically hypermutating a human V.sub.L gene sequence, comprising placing a human V.sub.L gene segment and a human J.sub.L gene segment in operable linkage with an endogenous mouse C.sub.H gene at an endogenous mouse heavy chain immunoglobulin locus, exposing the mouse to an antigen of interest, and obtaining from the mouse a somatically hypermutated human V.sub.L gene sequence that binds the antigen of interest.
[0181] In one embodiment, the method further comprises obtaining from the mouse a V.sub.L gene sequence from a light chain that is cognate to the human somatically hypermutated human V.sub.L gene sequence that binds the antigen of interest.
[0182] V.sub.L Binding Proteins with D.sub.H Sequences
[0183] In various aspects, mice comprising an unrearranged immunoglobulin light chain V gene segment and an unrearranged (e.g., light or heavy) J gene segment also comprise an unrearranged DH gene segment that is capable of recombining with the J segment to form a rearranged D/J sequence, which in turn is capable of rearranging with the light chain V segment to form a rearranged variable sequence derived from (a) the light chain V segment, (b) the DH segment, and (c) the (e.g., light or heavy) J segment; wherein the rearranged variable sequence is operably linked to a heavy chain constant sequence (e.g., selected from CH1, hinge, CH2, CH3, and a combination thereof; e.g., operably linked to a mouse or human CH1, a hinge, a CH2, and a CH3).
[0184] In various aspects, mice comprising unrearranged human light chain V segments and J segments that also comprise a human D segment are useful, e.g., as a source of increased diversity of CDR3 sequences. Normally, CDR3 sequences arise in light chains from V/J recombination, and in heavy chains from V/D/J recombination. Further diversity is provided by nucleotide additions that occur during recombination (e.g., N additions), and also as the result of somatic hypermutation. Binding characteristics conferred by CDR3 sequences are generally limited to those conferred by the light chain CDR3 sequence, the heavy chain CDR3 sequence, and a combination of the light and the heavy chain CDR3 sequence, as the case may be. In mice as described herein, however, an added source of diversity is available due to binding characteristics conferred as the result of a combination of a first light chain CDR3 (on the heavy chain polypeptide) and a second light chain CDR3 (on the light chain polypeptide). Further diversity is possible where the first light chain CDR3 may contain a sequence derived from a D gene segment, as from a mouse as described herein that comprises an unrearranged V segment from a light chain V region operably linked to a D segment and operably linked to a J segment (light or heavy), employing the RSS engineering as taught here.
[0185] Another source of diversity is the N and/or P additions that can occur in the V(light)/J or V(light)/D/J recombinations that are possible in mice as described. Thus, mice described herein not only provide a different source of diversity (light chain-light chain) but also a further source of diversity due to the addition of, e.g., 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 or more N additions in a rearranged V(light)/J or a rearranged V(light)/D/J gene in a mouse as described herein.
[0186] In various aspects the use of a D gene segment operably linked to a J gene segment and a light chain V gene segment provides an enhanced diversity. Operable linkage of a DH segment in this instance will require that that D segment is capable of recombining with the J segment with which it is recited. Thus, the D segment will require to have juxtaposed a downstream RSS that is matched to the RSS juxtaposed upstream of the J segment such that the D segment and the J segment may rearrange. Further, the D segment will require an appropriate RSS juxtaposed upstream that is matched to the RSS juxtaposed downstream of the V segment such that the rearranged D/J segment and the V segment may rearrange to form a gene encoding a variable domain.
[0187] An RSS, or a recombination signal sequence, comprises a conserved nucleic acid heptamer sequence separated, by 12 base pairs (bp) or 23 base pairs (bp) of unconserved sequence, from a conserved nucleic acid nonamer sequence. RSS's are used by recombinases to achieve joining of immunoglobulin gene segments during the rearrangement process following the 12/23 rule. According to the 12/23 rule, a gene segment juxtaposed with an RSS having a 12 bp (unconserved) spacer rearranges with a gene segment juxtaposed with an RSS having a 23 bp (unconserved) spacer; i.e., rearrangements between gene segments each having an RSS with a 12 bp spacer, or each having an RSS with a 23 bp spacer, are generally not observed.
[0188] In the case of the λ light chain locus, variable gene segments (Vλ gene segments) are flanked downstream (with respect to the direction of transcription of the V sequence) with an RSS having a 23-mer spacer, and joining gene segments (Jλ gene segments) are flanked upstream (with respect to the direction of transcription of the J sequence) with an RSS having a 12-mer spacer. Thus, Vλ and Jλ segments are flanked with RSS's that are compatible under the 12/23 rule, and therefore are capable of recombine during rearrangement.
[0189] At the κ locus in a wild-type organism, however, each functional Vκ segment is flanked downstream with an RSS having a 12-mer spacer. Jκ segments, therefore, have 23-mer spaces juxtaposed on the upstream side of the Jκ segment. At the heavy chain locus, V.sub.H gene segments are juxtaposed downstream with an RSS having a 23-mer spacer, followed by D.sub.H segment juxtaposed upstream and downstream with a 12-mer spacer, and J.sub.H segments each with a 23-mer segment juxtaposed on the upstream side of the J.sub.H segment. At the heavy chain locus, D/J recombination occurs first, mediated by the downstream D.sub.H RSS with the 12-mer spacer and the upstream J.sub.H RSS with the 23-mer spacer, to yield an intermediate rearranged D-J sequence having an RSS juxtaposed on the upstream side that has an RSS with a 12-mer spacer. The rearranged D-J segment having the RSS with the 12-mer juxtaposed on the upstream side then rearranges with the V.sub.H segment having the RSS with the 23-mer juxtaposed on its downstream side to form a rearranged V/D/J sequence.
[0190] In one embodiment, a Vλ segment is employed at the heavy chain locus with a J gene segment that is a Jκ segment, wherein the Vλ segment comprises an RSS juxtaposed on the downstream side of the Vλ sequence, and the RSS comprises a 23-mer spacer, and the J segment is a Jλ segment with an RSS juxtaposed on its upstream side having a 12-mer spacer (e.g., as found in nature).
[0191] In one embodiment, a Vλ segment is employed at the heavy chain locus with a J gene segment that is a Jκ or a J.sub.H gene segment, wherein the Vλ sequence has juxtaposed on its downstream side an RSS comprising a 23-mer spacer, and the Jκ or J.sub.H segment has juxtaposed on its upstream side an RSS comprising a 12-mer spacer.
[0192] In one embodiment, a Vλ segment is employed at the heavy chain locus with a D.sub.H gene segment and a J gene segment. In one embodiment, the Vλ segment comprises an RSS juxtaposed on the downstream side of the Vλ sequence with an RSS having a 23-mer spacer; the D.sub.H segment comprises an RSS juxtaposed on the upstream side and on the downstream side of the D.sub.H sequence with an RSS having a 12-mer spacer; and a J segment having an RSS juxtaposed on its upstream side having a 23-mer spacer, wherein the J segment is selected from a Jλ, a Jκ, and a J.sub.H.
[0193] In one embodiment, a Vκ segment is employed at the heavy chain locus with a J gene segment (with no intervening D segment), wherein the Vκ segment has an RSS juxtaposed on the downstream side of the Vκ segment that comprises a 12-mer spaced RSS, and the J segment has juxtaposed on its upstream side a 23-mer spaced RSS, and the Jκ segment is selected from a Jκ segment, a Jλ segment, and a J.sub.H segment. In one embodiment, the V segment and/or the J segment are human.
[0194] In one embodiment, the Vκ segment is employed at the heavy chain locus with a D segment and a J segment, wherein the Vκ segment has an RSS juxtaposed on the downstream side of the Vκ segment that comprises a 12-mer spaced RSS, the D segment has juxtaposed on its upstream and downstream side a 23-mer spaced RSS, and the J segment has juxtaposed on its upstream side a 12-mer spaced RSS. In one embodiment, the J segment is selected from a Jκ segment, a Jλ segment, and a J.sub.H segment. In one embodiment, the V segment and/or the J segment are human.
[0195] A Jλ segment with an RSS having a 23-mer spacer juxtaposed at its upstream end, or a Jκ or J.sub.H segment with an RSS having a 12-mer spacer juxtaposed at its upstream end, is made using any suitable method for making nucleic acid sequences that is known in the art. A suitable method for making a J segment having an RSS juxtaposed upstream wherein the RSS has a selected spacer (e.g., either 12-mer or 23-mer) is to chemically synthesize a nucleic acid comprising the heptamer, the nonamer, and the selected spacer and fuse it to a J segment sequence that is either chemically synthesized or cloned from a suitable source (e.g., a human sequence source), and employ the fused J segment sequence and RSS in a targeting vector to target the RSS-J to a suitable site.
[0196] A D segment with a 23-mer spaced RSS juxtaposed upstream and downstream can be made by any method known in the art. One method comprises chemically synthesizing the upstream 23-mer RSS and D segment sequence and the downstream 23-mer RSS, and placing the RSS-flanked D segment in a suitable vector. The vector may be directed to replace one or more mouse D segments with a human D segment with 12-mer RSS sequences juxtaposed on the upstream and downstream sides, or directed to be inserted into, e.g., a humanized locus at a position between a human V segment and a human or mouse J segment.
[0197] Suitable nonamers and heptamers for RSS construction are known in the art (e.g., see Janeway's Immunobiology, 7th ed., Murphy et al., (2008, Garland Science, Taylor & Francis Group, LLC) at page 148,
[0198] Bispecific-Binding Proteins
[0199] The binding proteins described herein, and nucleotide sequences encoding them, can be used to make multispecific binding proteins, e.g., bispecific binding proteins. In this aspect, a first polypeptide consisting essentially of a first V.sub.L domain fused with a C.sub.H region can associate with a second polypeptide consisting essentially of a second V.sub.L domain fused with a C.sub.H region. Where the first V.sub.L domain and the second V.sub.L domain specifically bind a different epitope, a bispecific-binding molecule can be made using the two V.sub.L domains. The C.sub.H region can be the same or different. In one embodiment, e.g., one of the C.sub.H regions can be modified so as to eliminate a protein A binding determinant, whereas the other heavy chain constant region is not so modified. This particular arrangement simplifies isolation of the bispecific binding protein from, e.g., a mixture of homodimers (e.g., homodimers of the first or the second polypeptides).
[0200] In one aspect, the methods and compositions described herein are used to make bispecific-binding proteins. In this aspect, a first V.sub.L that is fused to a C.sub.H region and a second V.sub.L that is fused to a C.sub.H region are each independently cloned in frame with a human IgG sequence of the same isotype (e.g., a human IgG1, IgG2, IgG3, or IgG4). The first V.sub.L specifically binds a first epitope, and the second V.sub.L specifically binds a second epitope. The first and second epitopes may be on different antigens, or on the same antigen.
[0201] In one embodiment, the IgG isotype of the C.sub.H region fused to the first V.sub.L and the IgG isotype of the C.sub.H region fused to the second V.sub.L are the same isotype, but differ in that one IgG isotype comprises at least one amino acid substitution. In one embodiment, the at least one amino acid substitution renders the heavy chain bearing the substitution unable or substantially unable to bind protein A as compared with the heavy chain that lacks the substitution.
[0202] In one embodiment, the first C.sub.H region comprises a first C.sub.H3 domain of a human IgG selected from IgG1, IgG2, and IgG4; and the second C.sub.H region comprises a second C.sub.H3 domain of a human IgG selected from IgG1, IgG2, and IgG4, wherein the second C.sub.H3 domain comprises a modification that reduces or eliminates binding of the second C.sub.H3 domain to protein A.
[0203] In one embodiment, the second C.sub.H3 domain comprises a 435R modification, numbered according to the EU index of Kabat. In another embodiment, the second C.sub.H3 domain further comprises a 436F modification, numbered according to the EU index of Kabat.
[0204] In one embodiment, the second C.sub.H3 domain is that of a human IgG1 that comprises a modification selected from the group consisting of D356E, L358M, N384S, K392N, V397M, and V422I, numbered according to the EU index of Kabat.
[0205] In one embodiment, the second C.sub.H3 domain is that of a human IgG2 that comprises a modification selected from the group consisting of N384S, K392N, and V422I, numbered according to the EU index of Kabat.
[0206] In one embodiment, the second C.sub.H3 domain is that of a human IgG4 comprising a modification selected from the group consisting of Q355R, N384S, K392N, V397M, R409K, E419Q, and V422I, numbered according to the EU index of Kabat.
[0207] In one embodiment, the binding protein comprises C.sub.H regions having one or more modifications as recited herein, wherein the constant region of the binding protein is nonimmunogenic or substantially nonimmunogenic in a human. In a specific embodiment, the C.sub.H regions comprise amino acid sequences that do not present an immunogenic epitope in a human. In another specific embodiment, the binding protein comprises a C.sub.H region that is not found in a wild-type human heavy chain, and the C.sub.H region does not comprise a sequence that generates a T-cell epitope.
EXAMPLES
[0208] The following examples are provided so as to describe how to make and use methods and compositions of the invention, and are not intended to limit the scope of what the inventors regard as their invention. Unless indicated otherwise, temperature is indicated in Celsius, and pressure is at or near atmospheric.
Example I
Introduction of Light Chain Gene Segments Into A Heavy Chain Locus
[0209] Various targeting constructs were made using VELOCIGENE® genetic engineering technology (see, e.g., U.S. Pat. No. 6,586,251 and Valenzuela, D. M., Murphy, A. J., Frendewey, D., Gale, N. W., Economides, A. N., Auerbach, W., Poueymirou, W. T., Adams, N. C., Rojas, J., Yasenchak, J., Chernomorsky, R., Boucher, M., Elsasser, A. L., Esau, L., Zheng, J., Griffiths, J. A., Wang, X., Su, H., Xue, Y., Dominguez, M. G., Noguera, I., Torres, R., Macdonald, L. E., Stewart, A. F., DeChiara, T. M., Yancopoulos, G. D. (2003). High-throughput engineering of the mouse genome coupled with high-resolution expression analysis. Nat Biotechnol 21, 652-659) to modify mouse genomic Bacterial Artificial Chromosome (BAC) libraries. Mouse BAC DNA was modified by homologous recombination to inactivate the endogenous mouse heavy chain locus through targeted deletion of V.sub.H, D.sub.H and J.sub.H gene segments for the ensuing insertion of unrearranged human germline κ light chain gene sequences (top of
[0210] Briefly, the mouse heavy chain locus was deleted in two successive targeting events using recombinase-mediated recombination. The first targeting event included a targeting at the 5′ end of the mouse heavy chain locus using a targeting vector comprising from 5′ to 3′ a 5′ mouse homology arm, a recombinase recognition site, a neomycin cassette and a 3′ homology arm. The 5′ and 3′ homology arms contained sequence 5′ of the mouse heavy chain locus. The second targeting event included a targeting at the 3′ end of the mouse heavy chain locus in the region of the J.sub.H gene segments using a second targeting vector that contained from 5′ to 3′ a 5′ mouse homology arm, a 5′ recombinase recognition site, a second recombinase recognition site, a hygromycin cassette, a third recombinase recognition site, and a 3′ mouse homology arm. The 5′ and 3′ homology arms contained sequence flanking the mouse J.sub.H gene segments and 5′ of the intronic enhancer and constant regions. Positive ES cells containing a modified heavy chain locus targeted with both targeting vectors (as described above) were confirmed by karyotyping. DNA was then isolated from the double-targeted ES cells and subjected to treatment with a recombinase thereby mediating the deletion of genomic DNA of the mouse heavy chain locus between the 5′ recombinase recognition site in the first targeting vector and the 5′ recombinase recognition site in the second targeting vector, leaving a single recombinase recognition site and the hygromycin cassette flanked by two recombinase recognition sites (see top of
[0211] Four separate targeting vectors were engineered to progressively insert 40 human Vκ gene segments and five human Jκ gene segments into the inactivated mouse heavy chain locus (described above) using standard molecular techniques recognized in the art (
[0212] A ˜110,499 bp human genomic fragment containing the first six human Vκ gene segments and five human Jκ gene segments was engineered to contain a PI-SceI site 431 bp downstream (3′) of the human Jκ5 gene segment. Another PI-SceI site was engineered at the 5′ end of a ˜7,852 bp genomic fragment containing the mouse heavy chain intronic enhancer, the IgM switch region (Sμ) and the IgM gene of the mouse heavy chain locus. This mouse fragment was used as a 3′ homology arm by ligation to the ˜110.5 kb human fragment, which created a 3′ junction containing, from 5′ to 3′, ˜110.5 kb of genomic sequence of the human κ light chain locus containing the first six consecutive Vκ gene segments and five Jκ gene segments, a PI-SceI site, ˜7,852 bp of mouse heavy chain sequence containing the mouse intronic enhancer, Sμ and the mouse IgM constant gene. Upstream (5′) from the human Vκ1-6 gene segment was an additional 3,710 bp of human κ sequence before the start of the 5′ mouse homology arm, which contained 19,752 bp of mouse genomic DNA corresponding to sequence 5′ of the mouse heavy chain locus. Between the 5′ homology arm and the beginning of the human κ sequence was a neomycin cassette flanked by three recombinase recognition sites (see Targeting Vector 1,
TABLE-US-00001 TABLE 1 Targeting Size of Human κ Gene Segments Added Vector Human κ Sequence Vκ Jκ 1 ~110.5 kb 4-1, 5-2, 7-3, 2-4, 1-5, 1-6 1-5 2 ~140 kb 3-7, 1-8, 1-9, 2-10, 3-11, 1-12, 1-13, 2-14, 3-15, 1-16 — 3 ~161 kb 1-17, 2-18, 2-19, 3-20, 6-21, 1-22, 1-23, 2-24, 3-25, 2-26, — 1-27, 2-28, 2-29, 2-30 4 ~90 kb 3-31, 1-32, 1-33, 3-34, 1-35, 2-36, 1-37, 2-38, 1-39, 2-40 —
[0213] Introduction of ten additional human Vκ gene segments into a hybrid heavy chain locus. A second targeting vector was engineered for introduction of 10 additional human Vκ gene segments to the modified mouse heavy chain locus described above (see
[0214] Introduction of fourteen additional human Vκ gene segments into a hybrid heavy chain locus. A third targeting vector was engineered for introduction of 14 additional human Vκ gene segments to the modified mouse heavy chain locus described above (see
[0215] Introduction of ten additional human Vκ gene segments into a hybrid heavy chain locus. A fourth targeting vector was engineered for introduction of 10 additional human Vκ gene segments to the modified mouse heavy chain locus described above (see
[0216] Using a similar approach as described above, other combinations of human light chain variable domains in the context of mouse heavy chain constant regions are constructed. Additional light chain variable domains may be derived from human Vλ and Jλ gene segments (
[0217] The human λ light chain locus extends over 1,000 kb and contains over 80 genes that encode variable (V) or joining (J) segments. Among the 70 Vλ gene segments of the human λ light chain locus, anywhere from 30-38 appear to be functional gene segments according to published reports. The 70 Vλ sequences are arranged in three clusters, all of which contain different members of distinct V gene family groups (clusters A, B and C). Within the human λ light chain locus, over half of all observed Vλ domains are encoded by the gene segments 1-40, 1-44, 2-8, 2-14, and 3-21. There are seven Jλ gene segments, only four of which are regarded as generally functional Jλ gene segments—Jλ1, Jλ2, Jλ3, and Jλ7. In some alleles, a fifth Cλ gene segment pair is reportedly a pseudo gene (Cλ6). Incorporation of multiple human Jλ gene segments into a hybrid heavy chain locus, as described herein, is constructed by de novo synthesis. In this way, a genomic fragment containing multiple human Jλ gene segments in germline configuration is engineered with multiple human Vλ gene segments and allow for normal V-J recombination in the context of a heavy chain constant region.
[0218] Coupling light chain variable domains with heavy chain constant regions represents a potentially rich source of diversity for generating unique V.sub.L binding proteins with human V.sub.L regions in non-human animals. Exploiting this diversity of the human λ light chain locus (or human κ locus as described above) in mice results in the engineering of unique hybrid heavy chains and gives rise to another dimension of binding proteins to the immune repertoire of genetically modified animals and their subsequent use as a next generation platform for the generation of therapeutics.
[0219] Additionally, human D.sub.H and J.sub.H (or Jκ) gene segments can be incorporated with either human Vκ or Vλ gene segments to construct novel hybrid loci that will give rise, upon recombination, to novel engineered variable domains (
[0220] Thus, using the strategy outlined above for incorporation of human κ light chain gene segments into an endogenous heavy chain locus allows for the use of other combinations of human λ light chain gene segments as well as specific human heavy chain gene segments (e.g., D.sub.H and J.sub.H) and combinations thereof.
Example II
Identification of Targeted ES cells Bearing Human Light Chain Gene Segments at an Endogenous Heavy Chain Locus
[0221] The targeted BAC DNA made in the foregoing Examples was used to electroporate mouse ES cells to created modified ES cells for generating chimeric mice that express V.sub.L binding proteins (i.e., human κ light chain gene segments operably linked to mouse heavy chain constant regions). ES cells containing an insertion of unrearranged human κ light chain gene segments were identified by the quantitative PCR assay, TAQMAN® (Lie and Petropoulos, 1998. Curr. Opin. Biotechnology 9:43-48). Specific primers sets and probes were design for insertion of human κ sequences and associated selection cassettes, loss of mouse heavy chain sequences and retention of mouse sequences flanking the endogenous heavy chain locus.
[0222] ES cells bearing the human κ light chain gene segments can be transfected with a construct that expresses a recombinase in order to remove any undesired selection cassette introduced by the insertion of the targeting construct containing human κ gene segments. Optionally, the selection cassette may be removed by breeding to mice that express the recombinase (e.g., U.S. Pat. No. 6,774,279). Optionally, the selection cassette is retained in the mice.
Example III
Generation and Analysis of Mice Expressing V.SUB.L .Binding Proteins
[0223] Targeted ES cells described above were used as donor ES cells and introduced into an 8-cell stage mouse embryo by the VELOCIMOUSE® method (see, e.g., U.S. Pat. No. 7,294,754 and Poueymirou, W. T., Auerbach, W., Frendewey, D., Hickey, J. F., Escaravage, J. M., Esau, L., Dore, A. T., Stevens, S., Adams, N. C., Dominguez, M. G., Gale, N. W., Yancopoulos, G. D., DeChiara, T. M., Valenzuela, D. M. (2007). F0 generation mice fully derived from gene-targeted embryonic stem cells allowing immediate phenotypic analyses. Nat Biotechnol 25, 91-99). VELOCIMICE® (F0 mice fully derived from the donor ES cell) independently bearing human κ gene segments at the mouse heavy chain locus were identified by genotyping using a modification of allele assay (Valenzuela et al., supra) that detected the presence of the unique human κ gene segments at the endogenous heavy chain locus (supra). Pups are genotyped and a pup heterozygous for the hybrid heavy chain gene locus is selected for characterizing expression of V.sub.L binding proteins.
[0224] Flow Cytometry. The introduction of human κ light chain gene segments into the mouse heavy chain locus was carried out in an F1 ES line (F1H4; Valenzuela et al. 2007, supra) derived from 129S6/SvEvTac and C57BL/6NTac heterozygous embryos that further contained an in situ replacement of the mouse κ light chain gene segments with human κ light chain gene segments (U.S. Pat. No. 6,596,541). The human κ light chain germline variable gene segments are targeted to the 129S6 allele, which carries the IgM.sup.a haplotype, whereas the unmodified mouse C576BL/6N allele bears the IgM.sup.b haplotype. These allelic forms of IgM can be distinguished by flow cytometry using antibodies specific to the polymorphisms found in the IgM.sup.a or IgM.sup.b alleles. Heterozygous mice bearing human κ light chain gene segments at the endogenous heavy chain locus as described in Example I were evaluated for expression of human V.sub.L binding proteins using flow cytometry.
[0225] Briefly, blood was drawn from groups of mice (n=6 per group) and grinded using glass slides. C57BL/6 and Balb/c mice were used as control groups. Following lysis of red blood cells (RBCs) with ACK lysis buffer (Lonza Walkersville), cells were resuspended in BD Pharmingen FACS staining buffer and blocked with anti-mouse CD16/32 (BD Pharmingen). Lymphocytes were stained with anti-mouse IgM.sup.b-FITC (BD Pharmingen), anti-mouse IgM.sup.a-PE (BD Pharmingen), anti-mouse CD19 (Clone 1D3; BD Biosciences), and anti-mouse CD3 (17A2; BIOLEGEND®) followed by fixation with BD CYTOFIX™ all according to the manufacturer's instructions. Final cell pellets were resuspended in staining buffer and analyzed using a BD FACSCALIBUR™ and BD CELLQUEST PRO™ software. Table 2 sets forth the average percent values for B cells (CD19.sup.+), T cells (CD3.sup.+), hybrid heavy chain (CD19.sup.+IgM.sup.a+), and wild type heavy chain (CD19.sup.+IgM.sup.b+) expression observed in groups of animals bearing each genetic modification.
[0226] In a similar experiment, B cell contents of the spleen, blood and bone marrow compartments from mice homozygous for six human Vκ and five human Jκ gene segments operably linked to the mouse heavy chain constant region (described in Example I,
[0227] Briefly, two groups (n=3 each, 8 weeks old females) of wild type and mice homozygous for six human Vκ and five human Jκ gene segments operably linked to the mouse heavy chain constant region were sacrificed and blood, spleens and bone marrow were harvested. Blood was collected into microtainer tubes with EDTA (BD Biosciences). Bone marrow was collected from femurs by flushing with complete RPMI medium (RPMI medium supplemented with fetal calf serum, sodium pyruvate, Hepes, 2-mercaptoethanol, non-essential amino acids, and gentamycin). RBCs from spleen and bone marrow preparations were lysed with ACK lysis buffer (Lonza Walkersville), followed by washing with complete RPMI medium.
[0228] Cells (1×10.sup.6) were incubated with anti-mouse CD16/CD32 (2.4G2, BD) on ice for ten minutes, followed by labeling with the following antibody cocktail for thirty minutes on ice: anti-mouse FITC-CD43 (1B11, BIOLEGEND®), PE-ckit (2B8, BIOLEGEND®), PeCy7-IgM (II/41, EBIOSCIENCE®), PerCP-Cy5.5-IgD (11-26c.2a, BIOLEGEND®), APC-eFluor 780-B220 (RA3-6B2, EBIOSCIENCE®), APC-CD19 (MB19-1, EBIOSCIENCE®). Bone marrow: immature B cells (B220.sup.intIgM.sup.+), mature B cells (B220.sup.hiIgM.sup.+), pro B cells (CD19.sup.+ckit.sup.+CD43.sup.+), pre B cells (CD19.sup.+ckit.sup.−CD43.sup.−), pre-B cells (CD19.sup.+CD43.sup.intIgM.sup.+/−), immature B cells (CD19.sup.+CD43.sup.−IgM.sup.+/−). Blood and spleen: B cells (CD19.sup.+), mature B cells (CD19.sup.+IgM.sup.intIgD.sup.hi), transitional/immature B cells (CD19.sup.+IgM.sup.hiIgD.sup.int).
[0229] Following staining, cells were washed and fixed in 2% formaldehyde. Data acquisition was performed on a LSRII flow cytometer and analyzed with FLOWJO™ software (Tree Star, Inc.).
[0230] In a similar experiment, B cell contents of the spleen, blood and bone marrow compartments from mice homozygous for thirty human Vκ and five human Jκ gene segments operably linked to the mouse heavy chain constant region (described in Example I,
[0231] Briefly, two groups (N=3 each, 6 week old females) of mice containing a wild-type heavy chain locus and a replacement of the endogenous Vκ and Jκ gene segments with human Vκ and Jκ gene segments (WT) and mice homozygous for thirty hVκ and five Jκ gene segments and a replacement of the endogenous Vκ and Jκ gene segments with human Vκ and Jκ gene segments (30hVκ-5hJκ HO) were sacrificed and spleens and bone marrow were harvested. Bone marrow and splenocytes were prepared for staining with various cell surface markers (as described above).
[0232] Cells (1×10.sup.6) were incubated with anti-mouse CD16/CD32 (2.4G2, BD Biosciences) on ice for ten minutes, followed by labeling with bone marrow or splenocyte panels for thirty minutes on ice. Bone marrow panel: anti-mouse FITC-CD43 (1B11, BIOLEGEND®), PE-ckit (2B8, BIOLEGEND®), PeCy7-IgM (II/41, EBIOSCIENCE®), APC-CD19 (MB19-1, EBIOSCIENCE®). Bone marrow and spleen panel: anti-mouse FITC-Igκ (187.1 BD Biosciences), PE-Igλ (RML-42, BIOLEGEND®), PeCy7-IgM (11/41, EBIOSCIENCE®), PerCP-Cy5.5-IgD (11-26c.2a, BIOLEGEND®), Pacific Blue-CD3 (17A2, BIOLEGEND®), APC-B220 (RA3-6B2, EBIOSCIENCE®), APC-H7-CD19 (1D3, BD). Bone marrow: immature B cells (B220.sup.intIgM.sup.+), mature B cells (B220.sup.hiIgM.sup.+), pro B cells (CD19.sup.+ckit.sup.+CD43.sup.+), pre B cells (CD19+ckit-CD43−), immature Igκ.sup.+ B cells (B220.sup.intIgM.sup.+IgM.sup.+Igλ.sup.−), immature Igλ.sup.+ B cells (B220.sup.intIgM.sup.+Igκ.sup.−Igλ.sup.+), mature Igκ.sup.+ B cells (B220.sup.hiIgM.sup.+Igκ.sup.+Igλ.sup.−), mature Igλ.sup.+ B cells (B220.sup.hiIgM.sup.+Igκ.sup.−Igλ.sup.+). Spleen: B cells (CD19.sup.+), mature B cells (CD19.sup.+IgD.sup.hiIgM.sup.int), transitional/immature B cells (CD19.sup.+IgD.sup.intIgM.sup.hi). Bone marrow and spleen: Igκ.sup.+ B cells (CD19.sup.+Igκ.sup.+Igλ.sup.−), Igλ.sup.+ B cells (CD19.sup.+ Igκ.sup.−Igλ.sup.+).
[0233] Following staining, cells were washed and fixed in 2% formaldehyde. Data acquisition was performed on a LSRII flow cytometer and analyzed with FLOWJO™ software (Tree Star, Inc.). The results demonstrated similar staining patterns and cell populations for all three compartments as compared to mice homozygous for six human Vκ and five human Jκ gene segments (described above). However, these mice demonstrated a loss in endogenous λ light chain expression in both the splenic and bone marrow compartments (
[0234] Isotype Expression. Total and surface (i.e., membrane bound) immunoglobulin M (IgM) and immunoglobulin G1 (IgG1) was determined for mice homozygous for human heavy and κ light chain variable gene loci (VELCOIMMUNE® Humanized Mice, see U.S. Pat. No. 7,105,348) and mice homozygous for six human Vκ and 5 human Jκ gene segments engineered into the endogenous heavy chain locus (6hVκ-5hJκ HO) by a quantitative PCR assay using TAQMAN® probes (as described above in Example II).
[0235] Briefly, CD19.sup.+ B cells were purified from the spleens of groups of mice (n=3 to 4 mice per group) using mouse CD19 Microbeads (Miltenyi Biotec) according to manufacturer's instructions. Total RNA was purified using the RNEASY™ Mini kit (Qiagen). Genomic RNA was removed using an RNase-free DNase on-column treatment (Qiagen). About 200 ng mRNA was reverse-transcribed into cDNA using the First Stand cDNA Synthesis kit (Invitrogen) and then amplified with the TAQMAN® Universal PCR Master Mix (Applied Biosystems) using the ABI 7900 Sequence Detection System (Applied Biosystems). Unique primer/probe combinations were employed to specifically determine expression of total, surface (i.e., transmembrane) and secreted forms of IgM and IgG1 isotypes (Table 3). Relative expression was normalized to the mouse κ constant region (mCκ).
TABLE-US-00002 TABLE 2 Mouse Genotype % CD3 % CD19 % IgMa % IgMb C57BL/6 22 63 0 100 Balb/c 11 60 100 0 6hVκ-5hJκ HET 43 30 7 85 16hVκ-5hJκ HET 33 41 7 81
TABLE-US-00003 TABLE 3 Isotype Sequence (5′-3′) SEQ ID NOs: Surface IgM sense: GAGAGGACCG TGGACAAGTC 1 antisense: TGACGGTGGT GCTGTAGAAG 2 probe: ATGCTGAGGA GGAAGGCTTT GAGAACCT 3 Total IgM sense: GCTCGTGAGC AACTGAACCT 4 antisense: GCCACTGCAC ACTGATGTC 5 probe: AGTCAGCCAC AGTCACCTGC CTG 6 Surface IgG1 sense: GCCTGCACAA CCACCATAC 7 antisense: GAGCAGGAAG AGGCTGATGA AG 8 probe: AGAAGAGCCT CTCCCACTCT CCTGG 9 Total IgG1 sense: CAGCCAGCGG AGAACTACAA G 10 antisense: GCCTCCCAGT TGCTCTTCTG 11 probe: AACACTCAGC CCATCATGGA CACA 12 Cκ sense: TGAGCAGCAC CCTCACGTT 13 antisense: GTGGCCTCAC AGGTATAGCT GTT 14 probe: ACCAAGGACG AGTATGAA 15
[0236] The results from the quantitative TAQMAN® PCR assay demonstrated a decrease in total IgM and total IgG1. However, the ratio of secreted versus surface forms of IgM and IgG1 appeared normal as compared to VELCOIMMUNE® humanized mice (data not shown).
[0237] Human κ gene segment usage and Vκ-Jκ junction analysis. Naïve mice homozygous for thirty hVκ and five Jκ gene segments and a replacement of the endogenous Vκ and Jκ gene segments with human Vκ and Jκ gene segments (30hVκ-5hJκ HO) were analyzed for unique human Vκ-Jκ rearrangements on mouse heavy chain (IgG) by reverse transcription polymerase chain reaction (RT-PCR) using RNA isolated from splenocytes.
[0238] Briefly, spleens were harvested and perfused with 10 mL RPMI-1640 (Sigma) with 5% HI-FBS in sterile disposable bags. Each bag containing a single spleen was then placed into a STOMACHER™ (Seward) and homogenized at a medium setting for 30 seconds. Homogenized spleens were filtered using a 0.7 μm cell strainer and then pelleted with a centrifuge (1000 rpm for 10 minutes) and RBCs were lysed in BD PHARM LYSE™ (BD Biosciences) for three minutes. Splenocytes were diluted with RPMI-1640 and centrifuged again, followed by resuspension in 1 mL of PBS (Irvine Scientific). RNA was isolated from pelleted splenocytes using standard techniques known in the art.
[0239] RT-PCR was performed on splenocyte RNA using primers specific for human hVκ gene segments and the mouse IgG. The mouse IgG primer was designed such that it was capable of amplifying RNA derived from all mouse IgG isotypes. PCR products were gel-purified and cloned into pCR2.1-TOPO TA vector (Invitrogen) and sequenced with primers M13 Forward (GTAAAACGAC GGCCAG; SEQ ID NO:16) and M13 Reverse (CAGGAAACAG CTATGAC; SEQ ID NO:17) located within the vector at locations flanking the cloning site. Human Vκ and Jκ gene segment usage among twelve selected clones are shown in Table 4.
[0240] As shown in this Example, mice homozygous for six human Vκ and five human Jκ gene segments or homozygous for thirty human Vκ and five human Jκ gene segments operably linked to the mouse heavy chain constant region demonstrated expression human light chain variable regions from a modified heavy chain locus containing light chain variable gene segments in their germline configuration. Progression through the various stages of B cell development was observed in these mice, indicating multiple productive recombination events involving light chain variable gene segments from an endogenous heavy chain locus and expression of such hybrid heavy chains (i.e., human light chain variable region linked to a heavy chain constant region) as part of the antibody repertoire.
TABLE-US-00004 TABLE 4 Hybrid Heavy Chain Clone Vκ Jκ C.sub.H SEQ ID NO: 1E 1-5 4 IgG2A/C 18 1G 1-9 4 IgG2A/C 19 1A 1-16 5 IgG3 20 2E 1-12 2 IgG1 21 1C 1-27 4 IgG2A/C 22 2H 2-28 1 IgG1 23 3D 3-11 4 IgG1 24 3A 3-20 4 IgG2A/C 25 4B 4-1 5 IgG2A/C 26 4C 4-1 2 IgG3 27 5A 5-2 2 IgG2A/C 28 5D 5-2 1 IgG1 29
Example IV
Propagation of Mice Expressing V.SUB.L .Binding Proteins
[0241] To create a new generation of V.sub.L binding proteins, mice bearing the unrearranged human κ gene segments can be bred to another mouse containing a deletion of the other endogenous heavy chain allele. In this manner, the progeny obtained would express only hybrid heavy chains as described in Example I. Breeding is performed by standard techniques recognized in the art and, alternatively, by commercial companies, e.g., The Jackson Laboratory. Mouse strains bearing a hybrid heavy chain locus are screened for presence of the unique hybrid heavy chains and absence of traditional mouse heavy chains.
[0242] Alternatively, mice bearing the unrearranged human κ gene segments at the mouse heavy chain locus can be optimized by breeding to other mice containing one or more deletions in the mouse light chain loci (κ and λ). In this manner, the progeny obtained would express unique human κ heavy chain only antibodies as described in Example I. Breeding is similarly performed by standard techniques recognized in the art and, alternatively, by commercial companies, e.g., The Jackson Laboratory. Mouse strains bearing a hybrid heavy chain locus and one or more deletions of the mouse light chain loci are screened for presence of the unique hybrid heavy chains containing human κ light chain domains and mouse heavy chain constant domains and absence of endogenous mouse light chains.
[0243] Mice bearing an unrearranged hybrid heavy chain locus are also bred with mice that contain a replacement of the endogenous mouse κ light chain variable gene locus with the human κ light chain variable gene locus (see U.S. Pat. No. 6,596,541, Regeneron Pharmaceuticals, The VELOCIMMUNE® Humanized Mouse Technology). The VELOCIMMUNE® Humanized Mouse includes, in part, having a genome comprising human κ light chain variable regions operably linked to endogenous mouse κ light chain variable constant region loci such that the mouse produces antibodies comprising a human κ light chain variable domain and a mouse heavy chain constant domain in response to antigenic stimulation. The DNA encoding the variable regions of the light chains of the antibodies can be isolated and operably linked to DNA encoding the human light chain constant regions. The DNA can then be expressed in a cell capable of expressing the fully human light chain of the antibody. Upon a suitable breeding schedule, mice bearing a replacement of the endogenous mouse κ light chain with the human κ light chain locus and an unrearranged hybrid heavy chain locus is obtained. Unique V.sub.L binding proteins containing somatically mutated human Vκ domains can be isolated upon immunization with an antigen of interest.
Example V
Generation of V.SUB.L .Binding Proteins
[0244] After breeding mice that contain the unrearranged hybrid heavy chain locus to various desired strains containing modifications and deletions of other endogenous Ig loci (as described in Example IV), selected mice can be immunized with an antigen of interest.
[0245] Generally, a VELOCIMMUNE® humanized mouse containing at least one hybrid heavy chain locus is challenged with an antigen, and cells (such as B-cells) are recovered from the animal (e.g., from spleen or lymph nodes). The cells may be fused with a myeloma cell line to prepare immortal hybridoma cell lines, and such hybridoma cell lines are screened and selected to identify hybridoma cell lines that produce antibodies containing hybrid heavy chains specific to the antigen used for immunization. DNA encoding the human Vκ regions of the hybrid heavy chains may be isolated and linked to desirable constant regions, e.g., heavy chain and/or light chain. Due to the presence of human Vκ gene segments fused to the mouse heavy chain constant regions, a unique antibody-like repertoire is produced and the diversity of the immunoglobulin repertoire is dramatically increased as a result of the unique antibody format created. This confers an added level of diversity to the antigen specific repertoire upon immunization. The resulting cloned antibody sequences may be subsequently produced in a cell, such as a CHO cell. Alternatively, DNA encoding the antigen-specific V.sub.L binding proteins or the variable domains may be isolated directly from antigen-specific lymphocytes (e.g., B cells).
[0246] Initially, high affinity V.sub.L binding proteins are isolated having a human Vκ region and a mouse constant region. As described above, the V.sub.L binding proteins are characterized and selected for desirable characteristics, including affinity, selectivity, epitope, etc. The mouse constant regions are replaced with a desired human constant region to generate unique fully human V.sub.L binding proteins containing somatically mutated human Vκ domains from an unrearranged hybrid heavy chain locus of the invention. Suitable human constant regions include, for example wild type or modified IgG1 or IgG4 or, alternatively Cκ or Cλ.
[0247] Separate cohorts of mice containing a replacement of the endogenous mouse heavy chain locus with six human Vκ and five human Jκ gene segments (as described in Example I) and a replacement of the endogenous Vκ and Jκ gene segments with human Vκ and Jκ gene segments were immunized with a human cell-surface receptor protein (Antigen X). Antigen X is administered directly onto the hind footpad of mice with six consecutive injections every 3-4 days. Two to three micrograms of Antigen X are mixed with 10 μg of CpG oligonucleotide (Cat # tlrl-modn-ODN1826 oligonucleotide; InVivogen, San Diego, Calif.) and 25 μg of Adju-Phos (Aluminum phosphate gel adjuvant, Cat# H-71639-250; Brenntag Biosector, Frederikssund, Denmark) prior to injection. A total of six injections are given prior to the final antigen recall, which is given 3-5 days prior to sacrifice. Bleeds after the 4th and 6th injection are collected and the antibody immune response is monitored by a standard antigen-specific immunoassay.
[0248] When a desired immune response is achieved splenocytes are harvested and fused with mouse myeloma cells to preserve their viability and form hybridoma cell lines. The hybridoma cell lines are screened and selected to identify cell lines that produce Antigen X-specific V.sub.L binding proteins. Using this technique several anti-Antigen X-specific V.sub.L binding proteins (i.e., binding proteins possessing human Vκ domains in the context of mouse heavy and light chain constant domains) are obtained.
[0249] Alternatively, anti-Antigen X V.sub.L binding proteins are isolated directly from antigen-positive B cells without fusion to myeloma cells, as described in U.S. 2007/0280945A1, herein specifically incorporated by reference in its entirety. Using this method, several fully human anti-Antigen X V.sub.L binding proteins (i.e., antibodies possessing human Vκ domains and human constant domains) were obtained.
[0250] Human κ Gene Segment Usage. To analyze the structure of the anti-Antigen X V.sub.L binding proteins produced, nucleic acids encoding the human Vκ domains (from both the heavy and light chains of the V.sub.L binding protein) were cloned and sequenced using methods adapted from those described in US 2007/0280945A1 (supra). From the nucleic acid sequences and predicted amino acid sequences of the antibodies, gene usage was identified for the hybrid heavy chain variable region of selected V.sub.L binding proteins obtained from immunized mice (described above). Table 5 sets for the gene usage of human Vκ and Jκ gene segments from selected anti-Antigen X V.sub.L binding proteins, which demonstrates that mice according to the invention generate antigen-specific V.sub.L binding proteins from a variety of human Vκ and Jκ gene segments, due to a variety of rearrangements at the endogenous heavy chain and κ light chain loci both containing unrearranged human Vκ and Jκ gene segments. Human Vκ gene segments rearranged with a variety of human Jκ segments to yield unique antigen-specific V.sub.L binding proteins.
TABLE-US-00005 TABLE 5 Hybrid Heavy Chain Light Chain Antibody Vκ Jκ Vκ Jκ A 4-1 3 3-20 1 B 4-1 3 3-20 1 C 4-1 3 3-20 1 D 4-1 3 3-20 1 E 4-1 3 3-20 1 F 4-1 3 3-20 1 G 4-1 3 3-20 1 H 4-1 3 3-20 1 I 4-1 3 3-20 1 J 1-5 3 1-33 3 K 4-1 3 3-20 1 L 4-1 3 1-9 3 M 4-1 1 1-33 4 N 4-1 1 1-33 3 O 1-5 1 1-9 2 P 1-5 3 1-16 4 Q 4-1 3 3-20 1 R 4-1 3 3-20 1 S 1-5 1 1-9 2 T 1-5 1 1-9 2 U 5-2 2 1-9 3 V 1-5 2 1-9 2 W 4-1 1 1-33 4
[0251] Enzyme-linked immunosorbent assay (ELISA). Human V.sub.L binding proteins raised against Antigen X were tested for their ability to block binding of Antigen X's natural ligand (Ligand Y) in an ELISA assay.
[0252] Briefly, Ligand Y was coated onto 96-well plates at a concentration of 2 μg/mL diluted in PBS and incubated overnight followed by washing four times in PBS with 0.05% Tween-20. The plate was then blocked with PBS (Irvine Scientific, Santa Ana, Calif.) containing 0.5% (w/v) BSA (Sigma-Aldrich Corp., St. Louis, Mo.) for one hour at room temperature. In a separate plate, supernatants containing anti-Antigen X V.sub.L binding proteins were diluted 1:10 in buffer. A mock supernatant with the same components of the V.sub.L binding proteins was used as a negative control. The extracellular domain (ECD) of Antigen X was conjugated to the Fc portion of mouse IgG2a (Antigen X-mFc). Antigen X-mFc was added to a final concentration of 0.150 nM and incubated for one hour at room temperature. The V.sub.L binding protein/Antigen X-mFc mixture was then added to the plate containing Ligand Y and incubated for one hour at room temperature. Detection of Antigen X-mFc bound to Ligand Y was determined with Horse-Radish Peroxidase (HRP) conjugated to anti-Penta-His antibody (Qiagen, Valencia, Calif.) and developed by standard colorimetric response using tetramethylbenzidine (TMB) substrate (BD Biosciences, San Jose, Calif.) neutralized by sulfuric acid. Absorbance was read at OD450 for 0.1 sec. Background absorbance of a sample without Antigen X was subtracted from all samples. Percent blocking was calculated for >250 (three 96 well plates) Antigen X-specific V.sub.L binding proteins by division of the background-subtracted MFI of each sample by the adjusted negative control value, multiplying by 100 and subtracting the resulting value from 100.
[0253] The results showed that several V.sub.L binding proteins isolated from mice immunized with Antigen X specifically bound the extracellular domain of Antigen X fused to the Fc portion of mouse IgG2a (data not shown).
[0254] Affinity Determination. Equilibrium dissociation constants (K.sub.D) for selected Antigen X-specific V.sub.L binding protein supernatants were determined by SPR (Surface Plasmon Resonance) using a BIACORE™ T100 instrument (GE Healthcare). All data were obtained using HBS-EP (10 mM HEPES, 150 mM NaCl, 0.3 mM EDTA, 0.05% Surfactant P20, pH 7.4) as both the running and sample buffers, at 25° C.
[0255] Briefly, V.sub.L binding proteins were captured from crude supernatant samples on a CM5 sensor chip surface previously derivatized with a high density of anti-human Fc antibodies using standard amine coupling chemistry. During the capture step, supernatants were injected across the anti-human Fc surface at a flow rate of 3 μL/min, for a total of 3 minutes. The capture step was followed by an injection of either running buffer or analyte at a concentration of 100 nM for 2 minutes at a flow rate of 35 μL/min. Dissociation of antigen from the captured V.sub.L binding protein was monitored for 6 minutes. The captured V.sub.L binding protein was removed by a brief injection of 10 mM glycine, pH 1.5. All sensorgrams were double referenced by subtracting sensorgrams from buffer injections from the analyte sensorgrams, thereby removing artifacts caused by dissociation of the V.sub.L binding protein from the capture surface. Binding data for each V.sub.L binding protein was fit to a 1:1 binding model with mass transport using BIAcore T100 Evaluation software v2.1.
[0256] The binding affinities of thirty-four selected V.sub.L binding proteins varied, with all exhibiting a K.sub.D in the nanomolar range (1.5 to 130 nM). Further, about 70% of the selected V.sub.L binding proteins (23 of 34) demonstrated single digit nanomolar affinity. T.sup.1/2 measurements for these selected V.sub.L binding proteins demonstrated a range of about 0.2 to 66 minutes. Of the thirty-four V.sub.L binding proteins, six showed greater than 3 nM affinity for Antigen X (1.53, 2.23, 2.58, 2.59, 2.79, and 2.84). The affinity data is consistent with the V.sub.L binding proteins resulting from the combinatorial association of rearranged human light chain variable domains linked to heavy and light chain constant regions (described in Table 4) being high-affinity, clonally selected, and somatically mutated. The V.sub.L binding proteins generated by the mice described herein comprise a collection of diverse, high-affinity unique binding proteins that exhibit specificity for one or more epitopes on Antigen X.
[0257] In another experiment, selected human V.sub.L binding proteins raised against Antigen X were tested for their ability to block binding of Antigen X's natural ligand (Ligand Y) to Antigen X in a LUMINEX® bead-based assay (data not shown). The results demonstrated that in addition to specifically binding the extracellular domain of Antigen X with affinities in the nanomolar range (described above), selected V.sub.L binding proteins were also capable of binding Antigen X from cynomolgus monkey (Macaca fascicularis).