ANIMAL MODELS AND THERAPEUTIC MOLECULES

Abstract

The invention discloses methods for the generation of chimaeric human-non-human antibodies and chimaeric antibody chains, antibodies and antibody chains so produced, and derivatives thereof including fully humanised antibodies; compositions comprising said antibodies, antibody chains and derivatives, as well as cells, non-human mammals and vectors, suitable for use in said methods.

Claims

1. A mouse having a genome comprising a homozygous recombinant immunoglobulin light chain locus, said homozygous recombinant immunoglobulin light chain locus comprising unrearranged human V light chain gene segments positioned (i) at an endogenous mouse immunoglobulin light chain locus comprising an endogenous light chain enhancer and (ii) upstream of a constant region, said homozygous recombinant light chain locus being functional to rearrange to express an Ig light chain comprising a human V region, wherein said homozygous recombinant immunoglobulin light chain locus comprises human Vλ gene segments and Jλ gene segments operatively linked to an endogenous kappa light chain enhancer at an endogenous kappa locus, wherein said human Vλ and Jλ gene segments comprise coordinates 22,375,609 to 23,327,884 from human chromosome 22, with reference to the GRCH37/hg19 sequence database of a human λ light chain locus; wherein said mouse expresses endogenous λ light chain, wherein said mouse comprises B cells expressing an immunoglobulin chain comprising a human Vλ region and B cells comprising an immunoglobulin chain comprising an endogenous Vλ region, and among said human and endogenous Vλ regions, said human Vλ region predominates, wherein at least 80% of the immunoglobulin light chains that comprise 2 variable regions expressed in said mouse comprise human λ variable regions, wherein said mouse comprises IgG comprising human Vλ region.

2. The mouse of claim 1, wherein said endogenous light chain enhancer is in germline order relative to an endogenous constant region.

3. The mouse of claim 1, wherein said endogenous kappa light chain enhancer comprises an iEκ or 3′ Eκ enhancer or both.

4. The mouse of claim 1, wherein the recombinant immunoglobulin light chain locus comprising human Vλ gene segments and Jλ gene segments is positioned upstream of an endogenous constant kappa region.

5. The mouse of claim 1, wherein said IgG comprising human Vλ regions comprises antigen specific human Vλ.

6. The mouse of claim 1, wherein said IgG comprising human Vλ regions comprises a repertoire of human λ variable regions.

7. The mouse of claim 1, wherein said mouse produces subsequent generation mice comprising unrearranged human Vλ and Jλ light chain gene segments positioned (i) at an endogenous mouse immunoglobulin light chain locus comprising an endogenous light chain enhancer and (ii) upstream of a constant region, said recombinant light chain locus being functional to rearrange to express an Ig light chain comprising a human Vλ region.

8. The mouse of claim 1, wherein endogenous VK expression is inactive.

9. The mouse of claim 1, wherein said mouse expresses λ immunoglobulin light chains comprising a plurality of a human λ variable regions encoded by a human Vλ and Jλ gene segments, and wherein said wherein said human Vλ and Jλ gene segments comprise coordinates 22,375,609 to 23,327,884 from human chromosome 22, with reference to the GRCH37/hg19 sequence database of a human λ light chain locus; and wherein said human Vλ and Jλ gene segments are included in said human λ locus DNA.

10. A mouse having a genome comprising a homozygous recombinant immunoglobulin light chain locus, said homozygous recombinant immunoglobulin light chain locus comprising unrearranged human V light chain gene segments positioned (i) at an endogenous mouse immunoglobulin light chain locus comprising an endogenous light chain enhancer and (ii) upstream of a constant region, said homozygous recombinant light chain locus being functional to rearrange to express an Ig light chain comprising a human V region, wherein said homozygous recombinant immunoglobulin light chain locus comprises human Vλ A gene segments and Jλ gene segments operatively linked to an endogenous kappa light chain enhancer at an endogenous kappa locus, wherein said human Vλ and Jλ gene segments comprise coordinates 22,375,609 to 23,327,884 from human chromosome 22, with reference to the GRCH37/hg19 sequence database of a human λ light chain locus, wherein 80% of total splenic B cells comprise immunoglobulin light chains that comprise λ variable regions expressed in said mouse comprise human λ variable regions, and wherein said mouse comprises IgG comprising human Vλ, wherein said mouse expresses endogenous λ light chain; wherein said mouse comprises B cells expressing an immunoglobulin chain comprising a human Vλ and B cells comprising an immunoglobulin chain comprising an endogenous Vλ, and among said human and endogenous Vλ said human Vλ predominates, wherein the recombinant immunoglobulin light chain locus comprising human Vλ gene segments and Jλ gene segments is positioned upstream of an endogenous constant kappa region.

Description

BRIEF DESCRIPTION OF THE FIGURES

[0788] FIGS. 1, 2, 3, 4, 5, 6, 7 and 8 show an iterative process for insertion of a series of human BACs into a mouse Ig locus

[0789] FIGS. 9, 10, 11, 12, 13, 14, 15, 16, 17 and 18 show in more detail the process of FIGS. 1-8 for the IgH and kappa locus

[0790] FIGS. 19 and 20 show the principles behind antibody generation in chimaeric mice

[0791] FIG. 21 shows a possible insertion site for the human DNA in a mouse chromosome

[0792] FIGS. 22, 23, 24, 25 and 26 disclose an alternative iterative process for insertion of a series of human BACs into a mouse Ig locus

[0793] FIGS. 27, 28 and 29 illustrate a mechanism for inversion of the host VDJ region

[0794] FIG. 30 illustrates proof of principle for insertion of a plasmid using an RMCE approach

[0795] FIG. 31 illustrates sequential RMCE—Integration into Landing Pad

[0796] FIG. 32 illustrates confirmation of Successful Insertion into Landing Pad

[0797] FIG. 33 illustrates PCR Confirmation of 3′ End Curing

[0798] FIG. 34 illustrates insertion of BAC #1 and PCR Diagnostics

[0799] FIG. 35 illustrates JH and JK usage

[0800] FIG. 36 illustrates DH usage

[0801] FIG. 37 illustrates the distribution of CDR-H3 length in human VDJCμ transcripts from chimera mice

[0802] FIG. 38 illustrates the distribution of nucleotide numbers of deletion and insertion in IGH-VDI or IGK-VJ junctions

[0803] FIG. 39 illustrates Distribution of JH Usage Within Each VHs

[0804] FIGS. 40A-40B illustrate Distribution of DH Usage Within Each VHs

[0805] FIGS. 41A-41B illustrate Nucleotide Gain or Loss at VJ Joints Generates IGK Variants

[0806] FIGS. 42A-42B illustrate Hypermutation in J Regions Generates IGK Variants

[0807] FIG. 43 illustrates Joint Diversity Produces Functional CDS

[0808] FIG. 44 illustrates a plot of identity of J.sub.H gene segment use a 5-RACE Cμ-specific library generated from the splenic B lymphocytes of transgenic mice according to the invention in which endogenous gene segment use has been inactivated by inversion

[0809] FIG. 45 illustrates the ratio of mouse VH to human VH usage as determined from antibody sequences from splenic B lymphocytes of transgenic mice according to the invention in which endogenous gene segment use has been inactivated by inversion

[0810] FIG. 46 illustrates inversion strategy schematic

[0811] FIG. 47 illustrates targeting construct R57 for inversion

[0812] FIG. 48 illustrates sequence analysis from a Cμ-specific 5-RACE library of splenic B lymphocytes of S1.sup.inv1 (one human IGH BAC (ie, multiple human VH, all functional human D and JH) with an inverted endogenous IGH locus) mouse shows that practically all the transcripts came from rearranged human V.sub.H-D-J.sub.H gene segments

[0813] FIG. 49 illustrates that the S1.sup.inv1 mouse shows a similar usage of both D and J.sub.H gene segments to human

[0814] FIG. 50 illustrates that mouse V.sub.H usage is further significantly reduced following insertion of the 2.sup.∞ human BAC into the endogenous heavy chain locus

[0815] FIG. 51 illustrates a gel showing that normal class-switching (to IgG-type) was observed in transcripts from mice of the invention. The rearranged transcripts were detected using RT-PCR with human VH-specific and mouse Cy-specific primers for amplification from peripheral blood cells of immunized transgenic mice

[0816] FIG. 52 illustrates sequence analysis amplified fragments demonstrate hypermutation occurred within the human variable regions of these IGγ chains from mice of the invention

[0817] FIG. 53 illustrates Flow cytometric analysis showing normal B-cell compartments in transgenic mice of the invention

[0818] FIGS. 54A-54D illustrate normal IgH isotypes in transgenic mice (H1) immunised with 100 μg Cholera Toxin B subunit. FIGS. 54E-54H illustrate normal IgH isotypes in transgenic mice (S1) immunised with 100 μg Cholera Toxin B subunit,

[0819] FIGS. 55A and 55B illustrate normal IgH isotypes and serum levels are obtained in transgenic H1 and S1 animals, respectively, of the invention following immunisation with antigens.

[0820] FIG. 56, part 1 illustrates the first and second BACs used for insertion into mouse endogenous light chain loci. The human DNA in each BAC is shown. Part 2 of FIG. 56 shows the insertion point of human lambda Ig locus DNA into the mouse endogenous kappa chain locus. Part 3 of FIG. 56 shows the insertion point of human lambda Ig locus DNA into the mouse endogenous lambda chain locus.

[0821] FIG. 57 shows the results of FACS analysis to determine mouse and human Cλ expression (and thus correspondingly mouse and human variable region expression) in B220.sup.+ splenic B cells from P1 homozygous mice (P1/P1) compared to wild-type mice (WT).

[0822] FIG. 58A shows the results of FACS analysis to determine mouse Cκ and Cλ expression in B220.sup.+ splenic B cells from P2 homozygous mice (P2/P2) compared to wild-type mice (WT). No detectable mouse Cκ expression was seen.

[0823] FIG. 58B shows the results of FACS analysis to determine human Cλ expression (and thus correspondingly human variable region expression) in B220.sup.+ splenic B cells from P2 homozygous mice (P2/P2) compared to wild-type mice (WT).

[0824] FIG. 59 shows human Vλ usage in P2 homozygous mice (P2/P2) and typical Vλ usage in humans (inset)

[0825] FIG. 60 shows human Jλ usage in P2 homozygous mice (P2/P2) and typical Jλ usage in humans (inset)

[0826] FIG. 61 shows Vλ usage is very high in P2 homozygous mice (P2/P2).

[0827] FIG. 62 shows the distribution of mouse Vκ and human Vλ gene segment usage from the chimaeric kappa locus in P2 homozygous mice (P2/P2).

[0828] FIG. 63 illustrates RSS arrangement in the lambda and kappa loci.

[0829] FIG. 64A shows the results of FACS analysis to determine mouse and human Cλ expression (and thus correspondingly mouse and human variable region expression) in B220.sup.+ splenic B cells from L2 homozygous mice in which endogenous kappa chain expression has been inactivated (L2/L2; KA/KA) compared to mice having no human lambda DNA inserted and in which endogenous kappa chain expression has been inactivated (KA/KA). Very high human Vλ usage was seen in the L2/L2; KA/KA) mice, almost to the exclusion of mouse VA use.

[0830] FIG. 64B: Splenic B-Cell Compartment Analysis. This figure shows the results of FACS analysis on splenic B-cells from transgenic L2/L2; KA/KA mice (L2 homozygotes; homozygous for human lambda gene segment insertion into endogenous lambda loci; endogenous kappa chain expression having been inactivated) compared with splenic B-cells from mice expressing only mouse antibodies (KA/KA mice). The results show that the splenic B-cell compartments in the mice of the invention are normal (ie, equivalent to the compartments of mice expressing only mouse antibody chains).

[0831] FIG. 65: B-cell development and markers in the bone marrow and splenic compartments.

[0832] FIG. 66A: Splenic B-Cell Compartment Analysis. This figure shows the results of FACS analysis on splenic B-cells from transgenic SIF/HA, KA/+ mice of the invention expressing heavy chain variable regions which are all human (where endogenous heavy chain expression has been inactivated by inversion), compared with splenic B-cells from mice expressing only mouse antibodies. The results show that the splenic B-cell compartments in the mice of the invention are normal (ie, equivalent to the compartments of mice expressing only mouse antibody chains).

[0833] S1F/HA, +/KA=(i) S1F—first endogenous heavy chain allele has one human heavy chain locus DNA insertion, endogenous mouse VDJ region has been inactivated by inversion and movement upstream on the chromosome; (ii) HA—second endogenous heavy chain allele has been inactivated (by insertion of an endogenous interrupting sequence); (iii) +—first endogenous kappa allele is a wild-type kappa allele; and (iv) KA—the second endogenous kappa allele has been inactivated (by insertion of an endogenous interrupting sequence). This arrangement encodes exclusively for heavy chains from the first endogenous heavy chain allele.

[0834] FIG. 66B: Splenic B-Cell Compartment Analysis. This figure shows the results of FAGS analysis on splenic B-cells from transgenic S1F/HA, K2/KA mice of the invention expressing heavy chain variable regions which are all human (where endogenous heavy chain expression has been inactivated by inversion) and human kappa chain variable regions, compared with splenic B-cells from +/HA, K2/KA mice. The results show that the splenic B-cell compartments in the mice of the invention are normal.

[0835] S1F/HA, K2/KA=(i) K2—the first endogenous kappa allele has two kappa chain locus DNA insertions between the most 3′ endogenous Jκ and the mouse Cκ, providing an insertion of 14 human Vκ and Jκ1-Jκ5; and (ii) KA—the second endogenous kappa allele has been inactivated (by insertion of an endogenous interrupting sequence). This arrangement encodes exclusively for heavy chains comprising human variable regions and substantially kappa light chains from the first endogenous kappa allele.

[0836] +/HA, K2/KA—this arrangement encodes for mouse heavy chains and human kappa chains.

[0837] FIG. 67A: Bone marrow B progenitor compartment analysis. This figure shows the results of FACS analysis on bone marrow (BM) B-cells from transgenic SIF/HA, KA/+ mice of the invention expressing heavy chain variable regions which are all human (where endogenous heavy chain expression has been inactivated by inversion), compared with BM B-cells from mice expressing only mouse antibodies. The results show that the BM B-cell compartments in the mice of the invention are normal (ie, equivalent to the compartments of mice expressing only mouse antibody chains).

[0838] FIG. 67B: Bone marrow B progenitor compartment analysis. This figure shows the results of FACS analysis on bone marrow (BM) B-cells from transgenic S1F/HA, K2/KA mice of the invention expressing heavy chain variable regions which are all human (where endogenous heavy chain expression has been inactivated by inversion) and human kappa chain variable regions, compared with BM B-cells from +/HA, K2/KA mice. The results show that the BM B-cell compartments in the mice of the invention are normal.

[0839] FIG. 68: shows Ig quantification for subtype and total Ig in various mice: S1F/HA, KA/+=(i) S1F—first endogenous heavy chain allele has one human heavy chain locus DNA insertion, endogenous mouse VDJ region has been inactivated by inversion and movement upstream on the chromosome; (ii) HA—second endogenous heavy chain allele has been inactivated (by insertion of an endogenous interrupting sequence); (iii) KA—the first endogenous kappa allele has been inactivated (by insertion of an endogenous interrupting sequence); and (iv)+—second endogenous kappa allele is a wild-type kappa allele. This arrangement encodes exclusively for heavy chains from the first endogenous heavy chain allele.

[0840] S1F/HA, K2/KA=(i) K2—the first endogenous kappa allele has two kappa chain locus DNA insertions between the most 3′ endogenous Jκ and the mouse Cκ, providing an insertion of 14 human Vκ and Jλ1-Jκ5; and (ii) KA—the second endogenous kappa allele has been inactivated (by insertion of an endogenous interrupting sequence). This arrangement encodes exclusively for heavy chains comprising human variable regions and substantially kappa light chains from the first endogenous kappa allele.

[0841] +/HA, K2/+—this arrangement encodes for mouse heavy chains and both mouse and human kappa chains.

[0842] +/HA, +/KA—this arrangement encodes for mouse heavy and kappa chains. In this figure, “Sum Ig” is the sum of IgG and IgM isotypes.

[0843] FIG. 69: shows Ig quantification for subtype and total Ig in various mice: S1F/HA, K2/KA (n=15) and 12 mice expressing only mouse antibody chains (+/HA, +/KA (n=6) and wild-type mice (WT; n=6)).

SEQUENCES

[0844] SEQ ID No 1 is a Rat switch sequence
SEQ ID No 2 is a landing pad targeting vector (long version)
SEQ ID No 3 is a landing pad targeting vector (shorter version)
SEQ ID No 4 is the mouse strain 129 switch
SEQ ID No 5 is the mouse strain C57 switch
SEQ ID No 6 is the 5′ homology arm of a landing pad
SEQ ID No 7 is oligo HV2-5
SEQ ID No 8 is oligo HV4-4
SEQ ID No 9 is oligo HV1-3
SEQ ID No 10 is oligo HV1-2
SEQ ID No 11 is oligo HV6-1
SEQ ID No 12 is oligo Cμ
SEQ ID No 13 is oligo KV1-9
SEQ ID No 14 is oligo KV1-8
SEQ ID No 15 is oligo KV1-6
SEQ ID No 16 is oligo KV1-5
SEQ ID No 17 is oligo Cκ
SEQ ID Nos 18-20 are rat switch sequences
SEQ ID No 21 is X.sub.1X.sub.2 T F G Q, where X.sub.1X.sub.2=PR, RT, or PW
SEQ ID No 22 is X.sub.1X.sub.2 T F G Q G T K V E I K R A D A, where X.sub.1X.sub.2=PR, RT, or PW;
SEQ ID No 23 is X.sub.3X.sub.4 T F G Q, where X.sub.3X.sub.4=PR or PW
SEQ ID No 24 is X.sub.3X.sub.4 T F G Q G T K V E I K R A D A, where X.sub.3X.sub.4=PR or PW

SEQ ID No 25 is Primer E1554

SEQ ID No 26 is Primer E1555

SEQ ID No 27 is Primer ELP1352_Cγ1

SEQ ID No 28 is Primer ELP1353_Cγ2b

SEQ ID No 29 is Primer ELP1354_Cγ2a

SEQ ID No 30 is Primer ELP1356_VH4-4

SEQ ID No 31 is Primer ELP1357_VH1-2,3

SEQ ID No 32 is Primer ELP1358_VH6-1

[0845] SEQ ID No 33 is Primer mIgG1_2 rev
SEQ ID No 34 is Primer mIgG2b rev
SEQ ID No 35 is Primer mIgG2a_2 rev
SEQ ID No 36 is Primer mCH1 unirev
SEQ ID No 37 is Primer mCH1 unirev_2
SEQ ID Nos 38-45 are CDRH3 sequences
SEQ ID Nos 46-50 is 3, 4, 5, 6 or more (up to 82) repeats of GGGCT
SEQ ID NOs 51-55 are heavy chain CDR1 sequences against CTB (cloned and reference)
SEQ ID NOs 56-60 are heavy chain CDR2 sequences against CTB (cloned and reference)
SEQ ID NOs 61-63 are heavy chain CDR3 sequences against CTB (cloned and reference)
SEQ ID NOs 64-68 are J Region sequences against CTB (cloned and reference)

DETAILED DESCRIPTION OF THE INVENTION

[0846] It will be understood that particular embodiments described herein are shown by way of illustration and not as limitations of the invention. The principal features of this invention can be employed in various embodiments without departing from the scope of the invention. Those skilled in the art will recognize, or be able to ascertain using no more than routine study, numerous equivalents to the specific procedures described herein. Such equivalents are considered to be within the scope of this invention and are covered by the claims.

[0847] The use of the word “a” or “an” when used in conjunction with the term “comprising” in the claims and/or the specification may mean “one,” but it is also consistent with the meaning of “one or more,” “at least one,” and “one or more than one.” The use of the term “or” in the claims is used to mean “and/or” unless explicitly indicated to refer to alternatives only or the alternatives are mutually exclusive, although the disclosure supports a definition that refers to only alternatives and “and/or.” Throughout this application, the term “about” is used to indicate that a value includes the inherent variation of error for the device, the method being employed to determine the value, or the variation that exists among the study subjects.

[0848] As used in this specification and claim(s), the words “comprising” (and any form of comprising, such as “comprise” and “comprises”), “having” (and any form of having, such as “have” and “has”), “including” (and any form of including, such as “includes” and “include”) or “containing” (and any form of containing, such as “contains” and “contain”) are inclusive or open-ended and do not exclude additional, unrecited elements or method steps

[0849] The term “or combinations thereof” as used herein refers to all permutations and combinations of the listed items preceding the term. For example, “A, B, C, or combinations thereof is intended to include at least one of: A, B, C, AB, AC, BC, or ABC, and if order is important in a particular context, also BA, CA, CB, CBA, BCA, ACB, BAC, or CAB. Continuing with this example, expressly included are combinations that contain repeats of one or more item or term, such as BB, AAA, MB, BBC, AAABCCCC, CBBAAA, CABABB, and so forth. The skilled artisan will understand that typically there is no limit on the number of items or terms in any combination, unless otherwise apparent from the context.

[0850] As a source of antibody gene segment sequences, the skilled person will also be aware of the following available databases and resources (including updates thereof) the contents of which are incorporated herein by reference:

[0851] The Kabat Database (G. Johnson and T. T. Wu, 2002; World Wide Web (www) kabatdatabase.com). Created by E. A. Kabat and T. T. Wu in 1966, the Kabat database publishes aligned sequences of antibodies, T-cell receptors, major histocompatibility complex (MHC) class I and II molecules, and other proteins of immunological interest. A searchable interface is provided by the Seghuntll

tool, and a range of utilities is available for sequence alignment, sequence subgroup classification, and the generation of variability plots. See also Kabat, E. A., Wu, T. T., Perry, H., Gottesman, K., and Foeller, C. (1991) Sequences of Proteins of Immunological Interest, 5th ed., NIH Publication No. 91-3242, Bethesda, Md., which is incorporated herein by reference, in particular with reference to human gene segments for use in the present invention.

[0852] KabatMan (A. C. R. Martin, 2002; World Wide Web (www) bioinf.org.uk/abs/simkab.html). This is a web interface to make simple queries to the Kabat sequence database.

[0853] IMGT (the International ImMunoGeneTics Information System®; M.-P. Lefranc, 2002; World Wide Web (www) imgt.cines.fr). IMGT is an integrated information system that specializes in antibodies, T cell receptors, and MHC molecules of all vertebrate species. It provides a common portal to standardized data that include nucleotide and protein sequences, oligonucleotide primers, gene maps, genetic polymorphisms, specificities, and two-dimensional (2D) and three-dimensional (3D) structures. IMGT includes three sequence databases (IMGT/LIGM-DB, IMGT/MHC-DB, IMGT/PRIMERDB), one genome database (IMGT/GENE-DB), one 3D structure database (IMGT/3Dstructure-DB), and a range of web resources (“/MGT Marie-Paule page”) and interactive tools.

[0854] V-BASE (I. M. Tomlinson, 2002; World Wide Web (www) mrc-cpe.cam.ac.uk/vbase). V-BASE is a comprehensive directory of all human antibody germline variable region sequences compiled from more than one thousand published sequences. It includes a version of the alignment software DNAPLOT (developed by Hans-Helmar Althaus and Werner Müller) that allows the assignment of rearranged antibody V genes to their closest germline gene segments.

[0855] Antibodies—Structure and Sequence (A. C. R. Martin, 2002; World Wide Web (www) bioinf.org.uk/abs). This page summarizes useful information on antibody structure and sequence. It provides a query interface to the Kabat antibody sequence data, general information on antibodies, crystal structures, and links to other antibody-related information. It also distributes an automated summary of all antibody structures deposited in the Protein Databank (PDB). Of particular interest is a thorough description and comparison of the various numbering schemes for antibody variable regions.

[0856] AAAAA (A Ho's Amazing Atlas of Antibody Anatomy; A. Honegger, 2001; World Wide Web (www) unizh.ch/˜antibody). This resource includes tools for structural analysis, modeling, and engineering. It adopts a unifying scheme for comprehensive structural alignment of antibody and T-cell-receptor sequences, and includes Excel macros for antibody analysis and graphical representation.

[0857] WAM (Web Antibody Modeling: N. Whitelegg and A. R. Rees, 2001; World Wide Web (www) antibody.bath.ac.uk). Hosted by the Centre for Protein Analysis and Design at the University of Bath, United Kingdom. Based on the AbM package (formerly marketed by Oxford Molecular) to construct 3D models of antibody Fv sequences using a combination of established theoretical methods, this site also includes the latest antibody structural information.

[0858] Mike's Immunoglobulin Structure/Function Page (M. R. Clark. 2001; World Wide Web (www) path.cam.ac.uk/˜mrc7/mikeimages.html) These pages provide educational materials on immunoglobulin structure and function, and are illustrated by many colour images, models, and animations. Additional information is available on antibody humanization and Mike Clark's Therapeutic Antibody Human Homology Project, which aims to correlate clinical efficacy and anti-immunoglobulin responses with variable region sequences of therapeutic antibodies.

[0859] The Antibody Resource Page (The Antibody Resource Page, 2000; World Wide Web (www) antibodyresource.com). This site describes itself as the “complete guide to antibody research and suppliers.” Links to amino acid sequencing tools, nucleotide antibody sequencing tools, and hybridoma/cell-culture databases are provided.

[0860] Humanization bY Design (J. Saldanha, 2000; World Wide Web (www) people.cryst.bbk.ac.uk/˜ubcg07s). This resource provides an overview on antibody humanization technology. The most useful feature is a searchable database (by sequence and text) of more than 40 published humanized antibodies including information on design issues, framework choice, framework back-mutations, and binding affinity of the humanized constructs.

[0861] See also Antibody Engineering Methods and Protocols, Ed. Benny K C Lo: Methods in Molecular Biology™, Human Press. Also at World Wide Web (www) blogsua.com/pdf/antibody-engineering-methods-and-protocolsantibody-engineering-methods-and-protocols.pdf

[0862] Any part of this disclosure may be read in combination with any other part of the disclosure, unless otherwise apparent from the context.

[0863] All of the compositions and/or methods disclosed and claimed herein can be made and executed without undue experimentation in light of the present disclosure. While the compositions and methods of this invention have been described in terms of preferred embodiments, it will be apparent to those of skill in the art that variations may be applied to the compositions and/or methods and in the steps or in the sequence of steps of the method described herein without departing from the concept, spirit and scope of the invention. All such similar substitutes and modifications apparent to those skilled in the art are deemed to be within the spirit, scope and concept of the invention as defined by the appended claims.

[0864] The following examples are put forth so as to provide those of ordinary skill in the art with a complete disclosure and description of how to make and use the invention, and are not intended to limit the scope of what the inventors regard as their invention.

EXAMPLES

Example 1

BAC Recombineering

[0865] Overall strategy: A mouse model of the invention can be achieved by inserting ˜960 kb of the human heavy chain locus containing all the V, D and J-regions upstream of the mouse constant region and 473 kb of the human kappa region upstream of the mouse constant region. Alternatively, or in tandem, the human lambda region is inserted upstream of the mouse constant region. This insertion is achieved by gene targeting in ES cells using techniques well known in the art.

[0866] High fidelity insertion of intact V-D-J regions into each locus in their native (wild-type) configuration is suitably achieved by insertion of human bacterial artificial chromosomes (BACs) into the locus. Suitably the BACs are trimmed so that in the final locus no sequence is duplicated or lost compared to the original. Such trimming can be carried out by recombineering.

[0867] The relevant human BACs, suitably trimmed covering these loci are on average 90 kb in size.

[0868] In one approach the full complement of human D and J-elements as well as seven or eight human V-regions are covered by the first BACs to be inserted in the experimental insertion scheme described below. The first BACs to be inserted in the IgH and IgK loci may contain the following V-regions. IgH: V6-1, VII-1-1, V1-2, VIII-2-1, V1-3, V4-4, V2-5 and IgK: V4-1, V5-2, V7-3, V2-4, V1-5, V1-6, V3-7, V1-8.

[0869] Suitably the performance of each locus is assessed after the first BAC insertion using chimaeric mice and also after each subsequent BAC addition. See below for detailed description of this performance test.

[0870] Nine additional BAC insertions will be required for the IgH locus and five for IgK to provide the full complement of human V-regions covering all 0.96 Mb and 0.473 Mb of the IgH and IgK loci, respectively.

[0871] Not all BACs retain their wild-type configuration when inserted into the ES cell genome. Thus, high density genomic arrays were deployed to screen ES cells to identify those with intact BAC insertions (Barrett, M. T., Scheffer, A., Ben-Dor, A., Sampas, N., Lipson, D., Kincaid, R., Tsang, P., Curry, B., Baird, K., Meltzer. P. S., et al. (2004). Comparative genomic hybridization using oligonucleotide microarrays and total genomic DNA. Proceedings of the National Academy of Sciences of the United States of America 101, 17765-17770.). This screen also enables one to identify and select against ES clones in which the ES cell genome is compromised and thus not able to populate the germ line of chimeric animals. Other suitable genomic tools to facilitate this assessment include sequencing and PCR verification.

[0872] Thus in one aspect the correct BAC structure is confirmed before moving to the next step.

[0873] It is implicit from the description above that in order to completely engineer the loci with 90 kb BACs, it is necessary to perform a minimum of 10 targeting steps for IgH and 5 steps for the IgK. Mice with an IgL locus can be generated in a similar manner to the IgK locus. Additional steps are required to remove the selection markers required to support gene targeting. Since these manipulations are being performed in ES cells in a step-wise manner, in one aspect germ line transmission capacity is retained throughout this process.

[0874] Maintaining the performance of the ES cell clones through multiple rounds of manipulation without the need to test the germ line potential of the ES cell line at every step may be important in the present invention. The cell lines currently in use for the KOMP and EUCOMM global knockout projects have been modified twice prior to their use for this project and their germ line transmission rates are unchanged from the parental cells (these lines are publicly available, see World Wide Web (www) komp.org and World Wide Web (www) eucomm.org). This cell line, called JM8, can generate 100% ES cell-derived mice under published culture conditions (Pettitt, S. J., Liang, Q., Rairdan, X. Y., Moran, J. L., Prosser, H. M., Beier, D. R., Lloyd, K. C., Bradley, A., and Skarnes, W. C. (2009). Agouti C57BL/6N embryonic stem cells for mouse genetic resources. Nature Methods.). These cells have demonstrated ability to reproducibly contribute to somatic and germ line tissue of chimaeric animals using standard mouse ES cell culture conditions. This capability can be found with cells cultured on a standard feeder cell line (SNL) and even feeder-free, grown only on gelatine-coated tissue culture plates. One particular sub-line, JM8A3, maintained the ability to populate the germ line of chimeras after several serial rounds of sub-cloning. Extensive genetic manipulation via, for example, homologous recombination—as would be the case in the present invention—cannot compromise the pluripotency of the cells. The ability to generate chimeras with such high percentage of ES cell-derived tissue has other advantages. First, high levels of chimerism correlates with germ line transmission potential and provide a surrogate assay for germ line transmission while only taking 5 to 6 weeks. Second, since these mice are 100% ES cell derived the engineered loci can be directly tested, removing the delay caused by breeding. Testing the integrity of the new Ig loci is possible in the chimera since the host embryo will be derived from animals that are mutant for the RAG-1 gene as described in the next section.

[0875] Another cell line that may be used is an HPRT-ve cell line, such as AB2.1, as disclosed in Ramirez-Solis R, Liu P and Bradley A, “Chromosome engineering in mice,” Nature, 1995; 378; 6558:720-4.

[0876] RAG-1 complementation: While many clones will generate 100% ES derived mice some will not. Thus, at every step mice are generated in a RAG-1-deficient background. This provides mice with 100% ES-derived B- and T-cells which can be used directly for immunization and antibody production. Cells having a RAG-2 deficient background, or a combined RAG-1/RAG-2 deficient background may be used, or equivalent mutations in which mice produce only ES cell-derived B cells and/or T cells.

[0877] In order that only the human-mouse IgH or IgK loci are active in these mice, the human-mouse IgH and IgK loci can be engineered in a cell line in which one allele of the IgH or IgK locus has already been inactivated. Alternatively the inactivation of the host Ig locus, such as the IgH or IgK locus, can be carried out after insertion.

[0878] Mouse strains that have the RAG-1 gene mutated are immunodeficient as they have no mature B- or T-lymphocytes (U.S. Pat. No. 5,859,307). T- and B-lymphocytes only differentiate if proper V(D)J recombination occurs. Since RAG-1 is an enzyme that is crucial for this recombination, mice lacking RAG-1 are immunodeficient. If host embryos are genetically RAG-1 homozygous mutant, a chimera produced by injecting such an embryo will not be able to produce antibodies if the animal's lymphoid tissues are derived from the host embryo. However, JM8 cells and AB2.1 cells, for example, generally contribute in excess of 80% of the somatic tissues of the chimeric animal and would therefore usually populate the lymphoid tissue. JM8 cells have wild-type RAG-1 activity and therefore antibodies produced in the chimeric animal would be encoded by the engineered JM8 ES cell genome only. Therefore, the chimeric animal can be challenged with an antigen by immunization and subsequently produce antibodies to that antigen. This allows one skilled in the art to test the performance of the engineered human/mouse IgH and IgK loci as described in the present invention. See FIGS. 19 and 20.

[0879] One skilled in the art would use the chimeric animal as described to determine the extent of antibody diversity (see e.g. Harlow, E. & Lane, D. 1998, 5.sup.th edition, Antibodies: A Laboratory Manual, Cold Spring Harbor Lab. Press, Plainview, N.Y.). For example, the existence in the chimeric animal's serum of certain antibody epitopes could be ascertained by binding to specific anti-idiotype antiserum, for example, in an ELISA assay. One skilled in the art could also sequence the genomes of B-cell clones derived from the chimeric animal and compare said sequence to wild-type sequence to ascertain the level of hypermutation, such hypermutation indicative of normal antibody maturation.

[0880] One skilled in the art would also use said chimeric animal to examine antibody function wherein said antibodies are encoded from the engineered Ig loci (see e.g. Harlow, E. & Lane, D. 1998, 5.sup.th edition, Antibodies: A Laboratory Manual, Cold Spring Harbor Lab. Press, Plainview, N.Y.). For example, antisera could be tested for binding an antigen, said antigen used to immunize the chimeric animal. Such a measurement could be made by an ELISA assay. Alternatively, one skilled in the art could test for neutralization of the antigen by addition of the antisera collected from the appropriately immunized chimeric animal.

[0881] It is well known to those skilled in the art that positive outcomes for any of these tests demonstrate the ability of the engineered Ig loci, the subject of the instant invention, to encode antibodies with human variable regions and mouse constant regions, said antibodies capable of functioning in the manner of wild-type antibodies.

[0882] Experimental Techniques: Recombineering for the production of vectors for use in homologous recombination in ES cells is disclosed in, for example, WO9929837 and WO0104288, and the techniques are well known in the art. In one aspect the recombineering of the human DNA takes place using BACs as a source of said human DNA. Human BAC DNA will be isolated using QIAGEN®, BAC purification kit. The backbone of each human BAC will be modified using recombineering to the exact same or similar configuration as the BAC already inserted into the mouse IgH region. The genomic insert of each human BAC will be trimmed using recombineering so that once the BACs are inserted, a seamless contiguous part of the human V(D)J genomic region will form at the mouse IgH or IgK locus, BAC DNA transfection by electroporation and genotyping will be performed accordingly to standard protocols (Prosser, H. M., Rzadzinska, A. K., Steel, K. P., and Bradley, A. (2008). “Mosaic complementation demonstrates a regulatory role for myosin Vila in actin dynamics of stereocilia.” Molecular and Cellular Biology 28, 1702-1712; Ramirez-Solis, R.; Davis, A. C., and Bradley, A. (1993). “Gene targeting in embryonic stem cells.” Methods in Enzymology 225, 855-878.). Recombineering will be performed using the procedures and reagents developed by Pentao Liu and Don Court's laboratories (Chan, W., Costantino, N., Li, R., Lee, S. C., Su, Q., Melvin, D., Court. D. L., and Liu, P. (2007). “A recombineering based approach for high-throughput conditional knockout targeting vector construction.” Nucleic Acids Research 36, e64).

[0883] These and other techniques for gene targeting and recombination of BAC-derived chromosomal fragments into a non-human mammal genome, such as a mouse are well-known in the art and are disclosed in, for example, in World Wide Web (www) eucomm.org/information/targeting and World Wide Web (www) eucomm.org/information/publications.

[0884] Cell culture of C57BL/6N-derived cell lines, such as the JM8 male ES cells will follow standard techniques. The JM8 ES cells have been shown to be competent in extensively contributing to somatic tissues and to the germline, and are being used for large mouse mutagenesis programs at the Sanger Institute such as EUCOMM and KOMP (Pettitt, S. J., Liang, Q., Rairdan, X. Y., Moran, J. L., Prosser, H. M., Beier, D. R., Lloyd, K. C., Bradley, A., and Skarnes, W. C. (2009). “Agouti C57BL/6N embryonic stem cells for mouse genetic resources.” Nature Methods.). JM8 ES cells (1.0×10.sup.7) will be electroporated (500 ΞF, 230V; Bio-Rad®) with 10 μg I-Scel linearized human BAC DNA. The transfectants will be selected with either Puromycin (3 μg/ml) or G418 (150 μg/ml). The selection will begin either 24 hours (with G418) or 48 hours (with Puromycin) post electroporation and proceed for 5 days. 10 μg linearized human BAC DNA can yield up to 500 Puromycin or G418 resistant ES cell colonies. The antibiotic resistant ES cell colonies will be picked into 96-well cell culture plates for genotyping to identify the targeted clones.

[0885] Once targeted mouse ES cell clones are identified, they will be analyzed by array Comparative Genomic Hybridization (CGH) for total genome integrity (Chung, Y. J., Jonkers, J., Kitson, H., Fiegler, H., Humphray, S., Scott, C., Hunt, S., Yu, Y., Nishijima, I., Velds, A., et al. (2004). “A whole-genome mouse BAC microarray with 1-Mb resolution for analysis of DNA copy number changes by array comparative genomic hybridization.” Genome research 14, 188-196, and Liang, Q., Conte, N., Skarnes, W. C., and Bradley, A. (2008). “Extensive genomic copy number variation in embryonic stem cells.” Proceedings of the National Academy of Sciences of the United States of America 105, 17453-17456.). ES cells that have abnormal genomes do not contribute to the germline of the chimeric mice efficiently. BAC integrity will be examined by PCR-amplifying each known functional V gene in the BAC. For example, in one approach the first human BAC chosen for the IgH locus has 6 functional V genes. To confirm the integrity of this BAC for the presence of these 6 IGH V genes, at least 14 pairs of PCR primers will be designed and used to PCR-amplify genomic DNA from the targeted ES cells. The human wild-type size and sequence of these fragments will ensure that the inserted BAC has not been rearranged.

[0886] More detailed CGH will also confirm the integrity of the inserted BACs. For example, one skilled in the art could use an oligo aCGH platform, which is developed by Agilent Technologies, Inc. This platform not only enables one to study genome-wide DNA copy number variation at high resolution (Barrett, M. T., Scheffer, A., Ben-Dor, A., Sampas, N., Lipson, D., Kincaid, R., Tsang, P., Curry, B., Baird, K., Meltzer, P. S., et al. (2004). “Comparative genomic hybridization using oligonucleotide microarrays and total genomic DNA.” Proceedings of the National Academy of Sciences of the United States of America 101, 17765-17770.), but permit examination of a specific genome region using custom designed arrays. Comparing the traditional aCGH techniques which rely on cDNA probes or whole BAC probes, the 60-mer oligonucleotides probes can ensure specific hybridization and high sensitivity and precision that is needed in order to detect the engineered chromosome alterations that were made. For example, oligos designed to hybridize at regular intervals along the entire length of the inserted BAC would detect even quite short deletions, insertions or other rearrangements. Also, this platform provides the greatest flexibility for customized microarray designs. The targeted ES cell genomic DNA and normal human individual genomic DNA will be labelled separately with dyes and hybridized to the array. Arrays slides will be scanned using an Aglient Technologies DNA microarray scanner. Reciprocal fluorescence intensities of dye Cy5 and dye Cy3 on each array image and the log 2 ratio values will be extracted by using Bluefuse software (Bluegnome). Spots with inconsistent fluorescence patterns (“confidence”<0.29 or “quality”=0) will be excluded before normalizing all log 2 ratio values. Within an experiment, Log 2 ratio between −0.29 and +0.29 for the signal from any oligo probe are regarded as no copy number change. The log 2 ratio threshold for “Duplication” is usually >0.29999, and for deletion is <0.29999.

[0887] Once the first human SAC is inserted into the mouse IgH locus and confirmed to be in its intact, native configuration, the FRT-flanked BAC backbone will be excised by using Flp site-specific recombinase. If regular Flp-catalyzed FRT recombination is not high enough, one can use Flo, an improved version of Flpo recombinase which in certain tests is 3-4 times more efficient than the original Flp in ES cells. After the BAC backbone is excised, ES cells will become sensitive to Puromycin (or G418) and resistant to FIAU (for loss of the TK cassette). The excision events will be further characterized by PCR amplification of the junction fragment using human genomic DNA primers. These FRT-flanked SAC backbone-free ES cells will be used for the next round of human BAC insertion and for blastocyst injection.

[0888] Targeting of the genome of an ES cell to produce a transgenic mouse may be carried out using a protocol as explained by reference to the attached FIGS. 1-18.

[0889] FIG. 1 illustrates three basic backbone vectors; an initiating cassette and 2 large insert vectors 1 and 2 respectively. The initiating cassette comprises sequences homologous to the desired site of insertion into the mouse genome, those sites flanking a selectable marker and stuffer primer sequence for PCR based genotyping to confirm correct insertion of BACs. The Stuffer-primer sequence provides the basis for genotyping each BAC addition step.

[0890] This sequence is considered to provide a robust well validated sequence template for PCR primer and may be located at the IScel site, ideally ˜1 kb from the BAC insert.

[0891] The large insert vectors comprise human DNA on plasmids with selectable markers and a unique restriction site for linearisation of the plasmid to aid in homologous recombination into the genome of the ES cell.

[0892] FIG. 2 illustrates insertion of an initiating cassette into the mouse genome by Homologous recombination between the mouse J4 and C alpha exons. Puromycin selection allows identification of ES cells with insertion of the cassette, pu(Delta)tk is a bifunctional fusion protein between puromycin N-acetyltransferase (Puro) and a truncated version of herpes simplex virus type 1 thymidine kinase (DeltaTk). Murine embryonic stem (ES) cells transfected with pu(Delta)tk become resistant to puromycin and sensitive to 1-(-2-deoxy-2-fluoro-1-beta-D-arabino-furanosyl)-5-iodouracil (FIAU). Unlike other HSV1 tk transgenes, puDeltatk is readily transmitted through the male germ line. Thus pu(Delta)tk is a convenient positive/negative selectable marker that can be widely used in many ES cell applications.

[0893] FIG. 3 illustrates targeting of the large insert vector 1 to the mouse ES cell genome. Linearisation of the vector is made at the same position as the stuffer primer sequence which allows for a gap repair genotyping strategy, well known in the art—see Zheng et al NAR 1999, Vol 27, 11, 2354-2360. In essence, random insertion of the targeting vector into the genome will not ‘repair’ the gap whereas a homologous recombination event will repair the gap. Juxtaposition of appropriate PCR primer sequences allows colonies to be screened individually for a positive PCR fragment indicating proper insertion. Positive selection using G418 allows for identification of mouse ES cells containing the neo selection marker. PCR verification can be made of all critical V, D and J regions. Array comparative genomic hybridization can be used to validate the BAC structure.

[0894] FIG. 4 illustrates the puro-delta-tk cassette and the BAC plasmid backbone is deleted using Flpe and select in FIAU. Since Flpe works inefficiently in mouse ES cells (5% deletion with transient Flpe expression), it is expected that in most cases, the recombination occurs between the two FRT sites flanking the BAC backbone. Flpo can also be tested to find out the recombination efficiency between two FRT sites that are 10 kb away.

[0895] Given that the FRT deletion step is selectable it is possible to pool FIAU resistant clones and proceed immediately to the next step in parallel with clonal analysis. Alternatively it may be desirable to show by short range PCR that the human sequences are now adjacent to those of the mouse as shown (Hu-primer 1 and Mo-primer)

[0896] At this stage a 200 kb human locus will have been inserted.

[0897] FIG. 5 illustrates a second large insert vector is targeted into the ES cell chromosome. The human BAC is targeted to the mouse IgH locus using the same initiation cassette insertion followed by /Sce1BAC linearization, BAC targeting to the initiation cassette and gap-repair genotyping strategy. Verification of the BAC insertion is carried out as before.

[0898] FIG. 6 illustrates the FRTY flanked BAC backbone of large insert vector 2 and the neo marker are deleted via Flpo. Note that this is not selectable, thus it will be necessary for clonal analysis at this point. This will enable confirmation of the juxtaposition of the human 2 insert with human 1 and other validation efforts.

[0899] At this stage a ˜200 kb human locus will have been inserted.

[0900] FIG. 7 illustrates the next large insert vector targeted to the mouse IgH locus. The pu-delta TK cassette is then removed, as for FIG. 4. The process can be repeated to incorporate other BACs.

[0901] FIG. 8 illustrates the final predicted ES cell construct.

[0902] FIGS. 9-18 provide a further level of detail of this process.

Example 2

Site-Specific Recombination

[0903] In a further method of the invention site specific recombination can also be employed. Site-specific recombination (SSR) has been widely used in the last 20-years for the integration of transgenes into defined chromosomal loci. SSR involves recombination between homologous DNA sequences.

[0904] The first generation of SSR-based chromosomal targeting involved recombination between

(i) a single recombination target site (RT) such as loxP or FRT in a transfected plasmid with
(ii) a chromosomal RT site provided by a previous integration. A major problem with this approach is that insertion events are rare since excision is always more efficient than insertion. A second generation of SSR called RMCE (recombinase-mediated cassette exchange) was introduced by Schlake and Bode in 1994 (Schlake, T.; J. Bode (1994). “Use of mutated FLP-recognition-target-(FRT-)sites for the exchange of expression cassettes at defined chromosomal loci”. Biochemistry 33: 12746-12751). Their method is based on using two heterospecific and incompatible RTs in the transfected plasmid which can recombine with compatible RT sites on the chromosome resulting in the swap of one piece of DNA for another—or a cassette exchange. This approach has been successfully exploited in a variety of efficient chromosomal targeting, including integration of BAC inserts of greater than 50 kb (Wallace, H. A. C. et al. (2007). “Manipulating the mouse genome to engineering precise functional syntenic replacements with human sequence”. Cell 128: 197-209; Prosser, H. M. et al. (2008). “Mosaic complementation demonstrates a regulatory role for myosin Vila in actin dynamics of Stereocilia”. Mol. Cell. Biol. 28: 1702-12).

[0905] The largest insert size of a BAG is about 300-kb and therefore this places an upper limit on cassette size for RMCE.

[0906] In the present invention a new SSR-based technique called sequential RMCE (SRMCE) was used, which allows continuous insertion of BAC inserts into the same locus.

[0907] The method comprises the steps of [0908] 1 insertion of DNA forming an initiation cassette (also called a landing pad herein) into the genome of a cell; [0909] 2 insertion of a first DNA fragment into the insertion site, the first DNA fragment comprising a first portion of a human DNA and a first vector portion containing a first selectable marker or generating a selectable marker upon insertion; [0910] 3 removal of part of the vector DNA; [0911] 4 insertion of a second DNA fragment into the vector portion of the first DNA fragment, the second DNA fragment containing a second portion of human DNA and a second vector portion, the second vector portion containing a second selectable marker, or generating a second selectable marker upon insertion; [0912] 5 removal of any vector DNA to allow the first and second human DNA fragments to form a contiguous sequence; and [0913] 6 iteration of the steps of insertion of a part of the human V(D)J DNA and vector DNA removal, as necessary, to produce a cell with all or part of the human VDJ or VJ region sufficient to be capable of generating a chimaeric antibody in conjunction with a host constant region,
wherein the insertion of at least one DNA fragment uses site specific recombination.

[0914] In one specific aspect the approach utilizes three heterospecific and incompatible loxP sites. The method is comprised of the steps as follows, and illustrated in FIGS. 22-26: [0915] 1. Targeting a landing pad into the defined locus. An entry vector containing an HPRT mini-gene flanked by inverted piggyBac (PB) ITRs is targeted into defined region (for example: a region between IGHJ and Eμ or IGKJ and Eκ or IGLC1 and Eλ3-1) to serve as a landing pad for BAC targeting. The HPRT mini-gene is comprised of two synthetic exons and associated intron. The 5′ HPRT exon is flanked by two heterospecific and incompatible loxP sites (one wild-type and the other a mutated site, lox5171) in inverted orientation to each other (FIG. 22). These two loxP sites provide recombination sites for the BAC insertion through RMCE. [0916] 2. Insertion of the 1V modified BAC into the targeted landing pad. The 1.sup.st BAC has a length of DNA to be inserted into the genome flanked by engineered modifications. The 5′ modification (loxP-neogene-lox2272-PGKpromoter-PB 5′LTR) and 3′ modification (PB3′LTR-puroΔTK gene-lox5171) is depicted in FIG. 23 along with the relative orientations of the lox sites and PB LTRs. With transient CRE expression from a co-electroporated vector, the DNA sequence would be inserted into the defined locus through RMCE. The cells in which a correct insertion has occurred can be selected as follows: (i) Puromycin-resistance (the puroΔTK gene has acquired a promoter—“PGK”—from the landing pad), (ii) 6TG-resistance (the HPRT mini-gene has been disrupted), and (iii) G418-resistance (selects for any insertion via the 5′ region PGK-neo arrangement). Any combination of these selection regimes can be used. G418- and 6TG-resistance select for correct events on the 5′ end while puro-resistance selects for correct events on the 3′ end. [0917] 3. Curing (removing) the 3′ modification of the 1.sup.st insertion. A properly inserted 1.sup.st BAC results the 3′ end having a puroΔTK gene flanked by inverted PB LTRs (FIG. 24)—essentially a proper transposon structure. This transposon can then be removed by the transient expression of the piggyBac transposase (from an electroporated vector). Cells with the correct excision event can be selected by FIAU resistance—ie, no thymidine kinase activity from the puroΔTK gene. This completely removes the 3′ modification leaving no trace nucleotides. [0918] 4. Insertion of a 2.sup.nd modified BAC into the 5′ end of 1.sup.st insertion. The 2.sup.nd BAC has a length of DNA to be inserted into the genome (usually intended to be contiguous with the DNA inserted with the 1.sup.st BAC) flanked by engineered modifications. The 5′ modification (loxP-HPRT mini gene 5′ portion-lox5171-PGK promoter-PB5′LTR) and 3′ modification (PB3′LTR-puroΔTK-lox2272) is depicted in FIG. 25 along with the relative orientations of the lox sites and PB LTRs. With transient CRE expression from a co-electroporated vector, the DNA sequence would be inserted into the defined locus through RMCE. The cells in which a correct insertion has occurred can be selected as follows: (i) HAT-resistance (the HPRT mini-gene is reconstituted by a correct insertion event, ie: the 5′ and 3′ exon structures are brought together), and (ii) puromycin-resistance (puroΔTK gene has acquired a promoter—“PGK”—from the landing pad). [0919] 5. Curing (removing) the 3′ modification of the 2.sup.nd insertion. A properly inserted 2.sup.nd BAC results the 3′ end having a puroΔTK gene flanked by inverted PB LTRs (FIG. 26)—essentially a proper transposon structure, exactly analogous to the consequence of a successful 1.sup.st BAC insertion. And therefore this transposon can likewise be removed by the transient expression of the piggyBac transposase (from an electroporated vector). Cells with the correct excision event can be selected by FIAU resistance—ie, no thymidine kinase activity from the puroΔTK gene. This completely removes the 3′ modification leaving no trace nucleotides. [0920] 6. After curing of the 3′ modification of the 2.sup.nd BAC insertion, the landing pad becomes identical to the original. This entire process, steps 2 through 5, can be repeated multiple times to build up a large insertion into the genome. When complete, there are no residual nucleotides remaining other than the desired insertion.

[0921] With the insertion of an odd number of BACs into the Ig loci, the endogenous VDJ or VJ sequences can be inactivated through an inversion via chromosomal engineering as follows (see FIGS. 27-29): [0922] 1. Targeting a “flip-over” cassette into a 5′ region 10 to 40 megabases away from the endogenous VDJ or VJ. The flip-over vector (PB3′LTR-PGK promoter-HPRT mini gene 5′ portion-loxP-puroΔTK-CAGGS promoter-PB3′LTR) is depicted in FIG. 27 along with the relative orientations of the lox sites and PB LTRs. [0923] 2. Transient CRE expression will result in recombination between the loxP site in the “flip-over” cassette and the loxP site in the 5′ modification. This 5′ modification is as described in Steps 2 and 3 above—essentially the modification resulting from insertion of an odd number of BACs, after the 3′ modification has been cured. The loxP sites are inverted relative to one another and therefore the described recombination event results in an inversion as depicted in FIG. 28. Cells with the correct inversion will be HAT-resistance since the HPRT mini-gene is reconstituted by a correct inversion. [0924] 3. A correct inversion also leaves two transposon structures flanking the “flip-over” cassette and the 5′ modification. Both can be excised with transient piggyBAC transposase expression, leaving no remnant of either modification (FIG. 29). Cells with the correct excisions can be selected as follows: (i) 6TG-resistance (the HPRT mini-gene is deleted) and (ii) FIAU-resistance (the puroΔTK gene is deleted). An inversion as described in the Ig loci would move the endogenous IGH-VDJ or IGK-VJ region away from the Eμ or Eκ enhancer region, respectively, and lead to inactivation of the endogenous IGH-VDJ or IGK-VJ regions.

[0925] The methods of insertion of the invention suitably provide one or more of: [0926] Selection at both 5′ and 3′ ends of the inserted DNA fragment; [0927] Efficient curing of the 3′ modification, preferably by transposase mediated DNA excision; [0928] Inactivation of endogenous IGH or IGK activity through an inversion; and [0929] Excision of modifications, leaving no nucleotide traces remaining in the chromosome.

Example 3

[0930] Insertion of a Test Vector into the Genome at a Defined Location

[0931] Proof of concept of the approach is disclosed in FIG. 30. In FIG. 30 a landing pad as shown in FIG. 22 was inserted into the genome of a mouse by homologous recombination, followed by insertion of the R21 plasmid into that landing pad via ore-mediated site specific recombination. The insertion event generated a number of general insertion events, 360 G418 resistant colonies, of which ˜220 were inserted into the desired locus, as demonstrated by disruption of the HRPT minilocus.

[0932] The R21 vector mimicks the 1.sup.st BAC insertion vector at the 5′ and 3′ ends, including all selection elements and recombinase target sites. In place of BAC sequences, there is a small ‘stuffer’ sequence. This vector will both test all the principals designed in the invention and allow easy testing of the results in that PCR across the stuffer is feasible and therefore allows both ends of the insertion to be easily tested. R21 was co-electroporated with a cre-expressing vector into the ES cells harbouring the landing pad in the IGH locus. Four sets of transformed cells were transfected in parallel and then placed under different selection regimes as indicated in FIG. 30. G418 selection (neo gene expression) resulted in the largest number of colonies due to there being no requirement for specific landing-pad integration. Any integration of R21 into the genome will provide neo expression leading to G418-resistance. Puro selection resulted in a similar colony number to Puro+6TG or G418+6TG, suggesting that the stringency of Puro selection is due to the PuroΔTK lacking a promoter in the vector. Puro expression is only acquired when an integration occurs near a promoter element—in this design most likely specifically in the landing pad. These conclusions are supported by the results from junction PCR which is shown in FIG. 31.

[0933] The next step in the invention is to ‘cure’ the 3′ end of the integrated BAC vector, leaving a seamless transition between the insertion and the flanking genome. This curing was demonstrated by expanding an individual clone from above (R21 inserted into the landing pad) and expressing piggyBac recombinase in this clone via transfection of an expressing plasmid. FIAU was used to select colonies in which the 3′ modification was excised—ie, through loss of the ‘PGK-puroΔTK’ element between the piggyBac terminal repeats. Fifty such clones resulted from a transfection of 10 cells; of these six were tested for the expected genomic structure. Successful curing resulted in positive PCR between the primer set labelled “3” in FIG. 32. Of the 6 clones, 4 had correct excisions, 1 clone remained in the original configuration and 1 other had a deletion.

[0934] These data demonstrate iterative insertion of DNA into a landing pad at a defined genomic locus using the approaches outlined above.

Example 4

[0935] Insertion of Large Parts of the Human IG Loci into Defined Positions in the Mouse Genome

[0936] Example 3 demonstrated that the design of the claimed invention was capable of providing for the insertion of a test vector into the genome at a defined location, in this case the R21 vector into the mouse IGH locus. The use of the appropriate selection media and the expression of cre-recombinase resulted in a genomic alteration with the predicted structure.

[0937] The same design elements described in this invention were built into the 5′ and 3′ ends of a BAC insert. Said insert comprised human sequences from the IGH locus and was approximately 166-kb. This engineered BAC was electroporated along with a cre-expressing plasmid DNA into mouse ES cells harbouring the landing pad at the mouse IGH locus. The transfected cell population was grown in puro-containing media to select for appropriate insertion events.

[0938] Seven resulting clones were isolated and further analysed. The expected recombination event and resulting structure are depicted in FIG. 33. Based upon data from the R21 experiment outlined in Example 3, a stringent selection for correct clones was expected when the transfected population was selected in puro-containing media. This is because the puro-coding region requires a promoter element and this is preferentially supplied by the landing pad after recombination. Accordingly, the majority of the 7 isolated clones had inserted correctly into the genome at the landing pad as determined by the diagnostic PCR. The primers for diagnosing a correct insertion are depicted in FIG. 33. Correct junctions are present in the genome if a 610-bp fragment is amplified between primers ‘A’ and ‘X’ and a 478-bp fragment is amplified between primers ‘Y’ and ‘B’ (FIGS. 33 and 34). Note that there are amplified fragments between ‘A’ and ‘1’ primers and ‘2’ and ‘B’ primers indicating the presence of parental genome (that is, the landing pad alone). These result from parental cells present internally in the cell colonies under puro-selection that escape the selection due to the geometry of a colony. After passaging the colony through puro-containing media, these parental junction fragments disappear indicating that the parental cells are removed from the population. In addition, all the clones were shown to be resistant to 6-TG as expected if the HPRT gene is inactivated by the correct insertion event.

[0939] These data indicate that the disclosed strategy for inserting large parts of the human IG loci into defined positions in the mouse genome will enable the construction of a mouse with a plurality of the variable regions of human IG regions upstream of the mouse constant regions as described.

Example 5

Inserted Loci are Functional in Terms of Gene Rearrangement, Junctional Diversity as Well as Expression

[0940] Bacterial artificial chromosomes (BACs) were created, wherein the BACs had inserts of human Ig gene segments (human V, D and/or J gene segments). Using methods described herein, landing pads were used in a method to construct chimaeric Ig loci in mouse embryonic stem cells (ES cells), such that chimaeric IgH and IgK loci were provided in which human gene segments are functionally inserted upstream of endogenous constant regions. To test if the human IgH-VDJ or IgK-VJ gene segments in the chimaera mice derived from human BAC-inserted ES cell clones appropriately rearrange and express, RT-PCR was performed for the RNA samples of white blood cells from those mice with the primer pairs of human variable (V) region and mouse constant (C) region. The sequences of oligos are shown as follows (Table 1). Each V oligo is paired with C oligo (HV with Cμ; KV with Cκ) for PCR reaction.

TABLE-US-00001 TABLE 1 Oligo Sequence HV2-5 AGATCACCTTGAAGGAGTCTGGTCC (SEQ ID NO 7) HV4-4 TGGTGAAGCCTTCGGAGACCCTGTC (SEQ ID NO 8) HV1-3 CACTAGCTATGCTATGCATTGGGTG (SEQ ID NO 9) HV1-2 ATGGATCAACCCTAACAGTGGTGGC (SEQ ID NO 10) HV6-1 GGAAGGACATACTACAGGTCCAAGT (SEQ ID NO 11) Cμ TAGGTACTTGCCCCCTGTCCTCAGT (SEQ ID NO 12) KV1-9 AGCCCAGTGTGTTCCGTACAGCCTG (SEQ ID NO 13) KV1-8 ATCCTCATTCTCTGCATCTACAGGA (SEQ ID NO 14) KV1-6 GGTAAGGATGGAGAACACTGGCAGT (SEQ ID NO 15) KV1-5 TTAGTAGCTGGTTGGCCTGGTATCA (SEQ ID NO 16) C.sub.K CTTTGCTGTCCTGATCAGTCCAACT (SEQ ID NO 17)

[0941] Using the one-step formulation of SuperScript™ III One-Step RT-PCR System with Platinum® Taq High Fidelity (Invitrogen™; World Wide Web (www) invitrogen.com/site/us/en/home/References/protocols/nucleic-acid-amplification-and-expression-profiling/pcr-protocol/superscript-3-one-step-rt-pcr-system-with-platinum-taq-high-fidelity.html#prot3), both cDNA synthesis and PCR amplification were achieved in a single tube using gene-specific primers and target RNAs.

[0942] The RT-PCR results showed most of the human IGH-VDJ or IGK-VJ gene segments appropriately rearrange and express in the chimaera mice. To investigate the details about the diversity generated from VDJ/VJ rearrangement, those specific RT-PCR fragments were cloned into a common vector for sequencing.

[0943] Sequencing results indicate that JH, DH, and Jκ usages (FIG. 35 and FIG. 36) are similar to human results. In addition, the results from the IGH-VDJCμ transcripts show that the range and mean of CDR-H3 length (FIG. 37) are similar to that observed in human. The junctional diversity generated from exonuclease and nucleotide addition activities (FIG. 38) was also observed. The IGH rearrangement possessed a higher frequency of these activities compared to the IGK one. These data suggest that the inserted loci are functional in terms of gene rearrangement, junctional diversity as well as expression,

Example 6

Productive VJ Rearrangement and Somatic Hypermutation can be Obtained

[0944] FIG. 41 shows an analysis of kappa mRNA from mice B-cells bearing rearranged VJ, the VJ having been rearranged from human germline kappa V1-8 and J1, and demonstrates that both that productive VJ rearrangement and somatic hypermutation can be obtained, the latter as seen from the changes in antibodies encoded by mRNA with respect to the germline sequences. The same is displayed for V1-6 and J1 in FIG. 42. Importantly, the recombination eliminates stop codons that are encoded by the combination of (unmutated) human germline gene segments, thereby allowing for antibody-encoding mRNA sequences. FIG. 43 demonstrates that inserted human kappa V1-5 J1 and V1-5 J4 can produce functional coding sequences in vivo and junctional diversity.

Example 7

Inactivation of Use of Endogenous IGHV Gene Segments for Expressed Rearranged Heavy Chain by Inversion

Introduction

[0945] A 5′-RACE Cμ-specific library was generated from the splenic B lymphocytes of transgenic mice, denoted S1 mice. These mice comprise transgenic heavy chain loci, each locus containing the six most 3′ functional human V.sub.H gene segments (V.sub.H2-5, 7-4-1, 4-4, 1-3, 1-2, 6-1), and all the human D and J.sub.H gene segments (comprising functional human D gene segments D1-1, 2-2, 3-3, 4-4, 5-5, 6-6, 1-7, 2-8, 3-9, 5-12, 6-13, 2-15, 3-16, 4-17, 6-19, 1-20, 2-21, 3-22, 6-25, 1-26 and 7-27; and functional human J gene segments J1, J2, J3, J4, J5 and J6) inserted into the endogenous heavy chain locus between endogenous IGHJ4 and Eμ (mouse chromosome 12: between coordinates 114666435 and 114666436). The human DNA was obtained from a bacterial artificial chromosome (BAC) containing the sequence of human chromosome 14 from coordinate 106328951 to coordinate 106494908. Further details on the construction of transgenic antibody loci using sRMCE is given elsewhere herein and in WO2011004192 (which is incorporated herein by reference). 4×96-well plates of clones were randomly picked for sequencing to determine the usage of the gene segments. All detected immunoglobulin heavy chains were rearranged from mouse V.sub.H or human V.sub.H with human D-J.sub.H. No mouse D and J.sub.H segments were detected in rearranged products (FIG. 44).

[0946] This result indicates that insertion of human V.sub.H-D-J.sub.H gene segments into an endogenous locus between the last endogenous J region (in this case, J.sub.H4) and the Eμ enhancer effectively inactivates the use of endogenous D and J.sub.H gene segments for expressed rearranged immunoglobulin heavy chains.

[0947] The ratio of mouse V.sub.H to human V.sub.H usage was around 3 to 1 (FIG. 45). To completely eliminate mouse V.sub.H use for antibody generation, the endogenous mouse V.sub.H-D-J.sub.H was inverted and moved to a distant region of the same chromosome. The rearrangement of mouse V.sub.Hs to human D-J.sub.H segments was totally blocked by effects of inversion and distance from the heavy chain locus.

[0948] The inversion strategy included three steps: (a) targeting of an inversion cassette, (b) inversion of endogenous VDJ and (c) excision of markers (FIG. 46).

[0949] (a) Targeting of the Inversion Cassette:

[0950] The inversion cassette consists of four components: a CAGGS promoter-driven puromycin-resistant-delta-thymidine kinase (puroΔtk) gene, a 5′ HPRT gene segment under the PGK promoter control, a loxP site between them and inversely oriented to another loxP site already in the heavy chain locus, and two flanking piggyback LTRs (PB3′LTRs). The inversion targeting cassette was inserted to a region that is 5′ and distant to the endogenous IGH locus at chromosome 12 as shown in FIG. 46. The targeted ES clones were identified and confirmed by PCR.

[0951] (b) Inversion:

[0952] Following the insertion, transient expression of cre from a transfected plasmid resulted in inversion of a section of chromosome 12 fragment including the endogenous V.sub.H-D-J.sub.H locus and intervening sequences through recombination of two inverted loxP sites, ie, those in the inversion cassette and the landing pad for the BAC insertion respectively. The invertants were selected by HAT and confirmed by junction PCRs cross the two recombined loxP sites.

[0953] (c) Excision of Markers:

[0954] The inversion rearranged the relative orientation of the PB3′LTRs from the inversion cassette and PB5′LTR from the landing pad to generate two piggyBac transposon structures flanking the inverted region. With transient expression of piggyBac transposase (PBase), these two transposons were excised from the chromosome (and thus the mouse cell genome). The cured ES clones were selected by 1-(-2-deoxy-2-fluoro-1-b-D-arabinofuranosyl)-5-iodouracil (FIAU) and 6TG, and confirmed by junction PCRs cross the excised regions.

Methods

[0955] Tissue culture: The procedures for ES cell culture, electroporation and drug selection have been described previously (Ramirez-Solis, R., A. C. Davis, and A. Bradley. 1993. Gene targeting in mouse embryonic stem cells. Methods Enzymol. 225:855-878).

[0956] Targeting of the locus for inversion: Briefly, S1 cell line (S1.11.1) was cultured in M15 medium (Knockout™ DMEM supplemented with 15% fetal bovine serum, 2 mM glutamine, antibiotics, and 0.1 mM 2-mercaptoethonal). Targeting construct R57 (FIG. 47) was linearized outside the region of homology by Notl. A total of 20 μg of the linearized construct was electroporated into S1 cell lines (AB2.1-derived) with a Bio-Rad® Gene Pulser™, and 107 cells were plated onto three 90-mm-diameter SNL76/7 feeder plates containing M15 medium. At 24 h after electroporation, M15 containing puromycin (3 μg of the active ingredient per ml) was added to each 90-mm-diameter plate, and the cells were maintained under selection for 9 days. 96 puromycin-resistant clones were then picked and expanded in 96-well plates. The targeting events were identified by long-range PCR.

[0957] Cre-loxP mediated inversion: 12 positive clones were pooled together and cultured in a 6-well tissue culture plate with M15 medium. The cells were transfected with 10 μg of pCAGGS-Cre plasmid for the inversion of mouse endogenous locus and then plated onto three 90-mm-diameter SNL76/7 feeder plates containing M15 medium. At 24 h after electroporation, M15 containing 1×HAT (hypoxanthine-aminoptern-thymidine) was added to each 90-mm-diameter plate, and the cells were maintained under selection for 7 days and then treated with 1×HT (hypoxanthine-thymidine) for 2 days. 48 HAT resistant colonies were picked and genotyped by PCR amplification of the junctions after Cre-loxP mediated inversion.

[0958] HyPBase-mediated marker excision: 12 positive clones were pooled together and cultured in 6-well tissue culture plate using M15 medium. The cells were transfected with 5 μg of HyPBase plasmid to activate the PB transposon LTRs flanking two selection markers (Hprt-mini gene and PGK-puro-tk gene) and plated onto one 90-mm-diameter SNL76/7 feeder plates containing M15 medium. At 72 h after electroporation, a serial dilution of the cells was then plated onto three 90-mm-diameter SNL76/7 feeder plates containing M15 supplemented with 1-(-2-deoxy-2-fluoro-1-b-D-arabinofuranosyl)-5-iodouracil (FIAU). Cells were maintained under selection for 10 days, and FIAU-resistant colonies were counted, picked, and expanded in 96-well plates. Positive clones were identified by PCR amplification of the junctions after excision of the selection markers. Positive clones were then expanded for blastocyst microinjection.

[0959] Generation of chimera and breeding: Mouse chimaeras were generated by microinjection of ES cells into C57/BL6 blastocysts and transferred into pseudopregnant recipients. Male chimaeras were test-crossed with C57/BL6 mice. Agouti F1 offspring were genotyped by S1 3′ junction PCR. Test-cross positive heterozygotes were further intercrossed to generate homozygotes.

[0960] Determination of VH-D-JH usage by rapid amplification of 5′-cDNA ends (5′ RACE) PCR: Total RNA was extracted from the spleen of S1inv1 mouse (KMSF30.1d) with TRIzol® Reagent (Invitrogen™, Life Technologies Ltd™) and treated with DNase I. Rapid amplification of 5′-cDNA ends (5′ RACE) PCR was performed using 5′/3′ RACE kit (2nd Generation, Roche) following the protocol supplied by the manufacturer. The first-strand cDNA was synthesised using primer E1554 (5′-ATGACTTCAGTGTTGTTCTGGTAG-3′; SEQ ID No 25) which is located at the mouse endogenous Cμ region. The synthesised first cDNA strand was purified using High Pure PCR Product Purification Kit (Roche). Poly(A) tail was added following the protocol supplied with the 5′/3′ RACE kit (2nd Generation, Roche). The 5′ end of the V.sub.M-D-J.sub.H rearranged transcript was amplified by nested PCR with forward primers Oligo dT, which is included in the kit, and nested Cμ-specific reverse primers E1555 (5′-CACCAGATTCTTATCAGAC-3′; SEQ ID No 26). Following reaction, the 5′ RACE PCR product was checked on a 1% agarose gel and purified using QIAquick® Gel Extraction Kit (QIAGEN) as the protocol supplied with the kit, then cloned into pDrive vector using QIAGEN PCR Cloning Kit (QIAGEN) for sequencing analysis.

Results

[0961] The sequence analysis from a CV-specific 5′-RACE library of splenic B lymphocytes of S1.sup.inv1 (one human IGH BAC (ie, multiple human VH, all functional human D and JH) with an inverted endogenous IGH locus version 1) mouse shows that practically all the transcripts came from rearranged human V.sub.H-D-J.sub.H gene segments (FIG. 48). Mouse V.sub.H usage was rarely detected (0.4%), and no mouse D and J.sub.1 usage was detected. Human V.sub.H usage was 99.6% and only human D and J.sub.H were used; it was hypothesized that the rare mouse V.sub.H, usage was due to trans-switching with another chromosome and not due to use of moue V.sub.H from the inverted sequences. The inversion resulted in complete inactivation of the endogenous V.sub.H use.

[0962] This result indicates that inversion is an effective way to inactivate the rearrangement of endogenous V.sub.H gene segments. The S1.sup.inv1 mouse also shows a similar usage of both D and J, gene segments to human (FIG. 49) (Link, J M et al. Mol. Immunol. 2005. 42, 943-955). Thus, a mouse was produced that comprises a transgenic heavy chain locus that expresses heavy chains comprising human variable regions, but no mouse variable regions, and furthermore the human variable regions demonstrated a normal, human sequence distribution corresponding to human D and J usage observed in humans.

Example 8

Inactivation of Use of Endogenous IGHV Gene Segments for Expressed Rearranged Heavy Chain by Insertion of Human IgH Genomic DNA

Introduction

[0963] Insertion of human BACs with V.sub.H-D-J.sub.H gene segments into an endogenous mouse heavy chain locus between J.sub.H4 and Eμ in chromosome 12 allows human V.sub.H-D-J.sub.H gene segments to effectively use mouse Eμ and 3′ enhancers and rearrange to generate chimeric antibody with human variable region and mouse constant region. Meanwhile, the endogenous V.sub.H-D-J.sub.H gene segments are pushed away from endogenous enhancers and constant regions. This distance effect results in inactivation of mouse D and J.sub.H use for expressed rearranged antibody products. As the distance increases by stepwise BAC insertion, it is expected that the mouse VH usage would be significantly reduced.

Results

[0964] Insertion of human DNA from a 1.sup.st human BAC (BAC comprising a the sequence of mouse Chromosome 14 from coordinate 106328951 to coordinate 106494908; containing six most 3′ functional V.sub.H gene segments (V.sub.H2-5, 7-4-1, 4-4, 1-3, 1-2, 6-1), and all the human D and J.sub.H gene segments) into the heavy chain endogenous locus of a AB2.1 ES cell genome between endogenous IGHJ4 and Eμ (at mouse chromosome 12: between coordinates 114666435 and 114666436) effectively inactivates the use of endogenous D and J.sub.H gene segments for expressed rearranged immunoglobulin heavy chain (FIG. 44). The rearranged transcripts with mouse V.sub.H gene segments are reduced in the resulting S1 mouse. The proportion of transcripts using mouse V.sub.H is around 75% of all observed sequences (FIG. 45).

[0965] Following the 1′ BAC DNA insertion, human DNA from a 2.sup.nd human BAC (Chr14: 106494909-106601551) (BAG comprising a the sequence of mouse Chromosome 14 from coordinate 106494909 to coordinate 106601551; containing 5 more functional VH gene segments (V.sub.H3-13, 3-11, 3-9, 1-8, 3-7)) was inserted into the landing pad left behind after curing following the 1V BAC insertion (see, eg, FIG. 24). The mouse V.sub.H usage is further significantly reduced following this insertion of the 2.sup.nd BAG into the locus. The proportion of transcripts using mouse VH was further reduced to 35% of all observed sequences (FIG. 50).

[0966] This result indicate that the endogenous V.sub.H-D-J.sub.H gene segments could be inactivated (ie, not used for expressed rearranged heavy chains) through insertion of human VDJ sequences from one or more BACs. As the distance increases by stepwise BAC insertion, it is expected that the mouse VH usage would be significantly reduced.

Example 9

Normal Class Switch and Hypermutation in Transgenic Mice of the Invention

Introduction

[0967] The B cell arm of the immune system has evolved to produce high affinity, antigen-specific antibodies in response to antigenic challenge. Antibodies are generated in B lymphocytes by a process of gene rearrangement in which variable (V), diversity (D; for the IGH locus) and joining (J) gene segments are recombined, transcribed and spliced to a Cμ (for IGH) or a Cκ or Cλ (for IGL) constant region gene segment to form an IgM antibody. Depending on the stage of B cell development, IgM is either located on the cell surface or secreted. The recombination process generates a primary antibody repertoire with sufficient germ line diversity to bind a wide range of antigens. However, it is usually not large enough to provide the high affinity antibodies that are required for an effective immune response to an antigen such as an infectious agent. Therefore, the immune system adopts a two-stage diversification process to increase diversity further. When challenged with antigens, B cells undergo selection and maturation by a process called somatic mutation. B cells expressing antibodies which bind to antigen undergo multiple rounds of diversification, clonal expansion and antigen selection in the germinal centres (GCs) of the secondary lymphoid organs. During this process, the rearranged variable regions of the immunoglobulin genes acquire somatic hypermutation through nucleotide substitution, addition or deletion. This stepwise process creates a secondary repertoire from the weak binders selected originally from the primary repertoire and combines rapid proliferation of antigen-reactive B cells with intense selection for quality of binding, eventually giving rise to high affinity antibodies with broad epitope coverage. During this process, antibodies undergo class switching in which the Cμ constant region is replaced by Cy, Ca or Cc to produce respectively IgG, A or E classes of antibody with different effector functions.

[0968] Insertion of 1.sup.st human BAC (Chr14: 106328951-106494908) containing six most 3′ functional V.sub.H gene segments (V.sub.H2-5, 7-4-1, 4-4, 1-3, 1-2, 6-1), and all the D and J.sub.H gene segments into the locus between endogenous IGHJ4 and Eμ (Chr12: 114666435 and 114666436) produces transgenic mice that generate chimeric immunoglobulin heavy chains containing human variable and mouse constant regions. This result demonstrates that human immunoglobulin gene segments are able to be rearranged and expressed in mice. Here, RT-PCR experiments and sequence analysis were performed to further demonstrate that immunized transgenic mice have proper class switch and hypermutation for generated antibodies.

Methods

[0969] RT-PCR and sequence analysis: Wild type or S1 chimera mice at 6-8 weeks of age were primed by intraperitoneal injection of 10 sheep RBCs suspended in phosphate buffer saline (PBS). The immunized mice were boosted twice with the same amount of sheep RBCs two and four weeks after priming. Four days after the last boost, peripheral blood cells were collected from the immunized mice. Total RNA was isolated from peripheral blood cells with TRIzol® reagent (Invitrogen™) and treated with DNase I. Reverse transcription polymerase chain reaction (RT-PCR) was performed using SuperScript® III First-Strand Synthesis System (Invitrogen™) following the protocol supplied by the manufacturer. The 1st strand cDNA was synthesized with the specific Cγ primers (Cγ1, Cγ2a, Cγ2b), following by PCR with specific human V primers (VH1-2,3, VH4-4, VH6-1) and Cy primers (Table 2). Following reaction, the RT-PCR product was checked on a 1% agarose gel and purified using QIAquick® Gel Extraction Kit (QIAGEN) as the protocol supplied with the kit, then cloned into pDrive vector using QIAGEN PCR Cloning Kit (QIAGEN) for sequencing analysis.

TABLE-US-00002 TABLE 2 ELP1352_Cy1 5′-AGAGCGGCCGCTGGGCAAC SEQ ID GTTGCAGGTGACGGTC-3′ No 327 ELP1353_Cy2b 5′-AGAGCGGCCGCTTTGTCCA SEQ ID CCGTGGTGCTGCTGG-3′ No 328 ELP1354_CY2a 5′-AGAGCGGCCGCACATTGCA SEQ ID GGTGATGGACTGGC-3′ No 329 ELP1356_VH4-4 5′-AGGACGCGTGAAACACCTG SEQ ID TGGTTCTTCCTCCTGC-3′ No 330 ELP1357_VH1-2,3 5-AGGACGCGTCACCATGGACT SEQ ID GGACCTGGAGGAT-3′ No 331 ELP1358_VH6-1 5′-AGGACGCGTATGTCTGTCT SEQ ID CCTTCCTCATCTTCC-3′ No 32

Results

[0970] The rearranged transcripts were detected using RT-PCR with human VH-specific and mouse Cy-specific primers for amplification from peripheral blood cells of immunized transgenic mice (FIG. 51). Further sequence analysis of these amplified fragments demonstrated hypermutation happened within the human variable regions of these IGγ chains (FIG. 52). These results indicate that loci of the invention comprising insertion of human IGH BAC containing V.sub.H, D and J.sub.H gene segments into the locus between endogenous IGHJ4 and Eμ regions has normal class switching and hypermutation functionality (IgM to IgG) following antigen challenge.

Example 10

Normal B Cell Compartments in Transgenic Mice of the Invention

Introduction

[0971] In mice, about 2×10.sup.7 bone marrow immature B cells are produced daily. Among them, only 10-20% of these cells survive to exit the bone marrow and enter the spleen. The immature splenic B cell population is divided into two distinct subsets: transitional 1 (T1) and transitional 2 (T2) B cells. In vivo experiments indicate that T1 cells give rise to T2 cells, whereas T2 cells can further differentiate into mature (M) B cells. In contrast to immature B cells (3-4 days old), mature B cells are long-lived (15-20 weeks old) and are ready to respond to antigens (Pillai S et al; Immunol. Reviews. 2004. 197: 206-218). Thus, the component of mature B cell population is directly linked to the efficiency of humoral immune response.

[0972] The T1, T2 and M cell populations can be categorized by their cell surface IgM and IgD levels. A normal phenotype of splenic B cell compartment is required to mount a robust immune response.

Methods

[0973] Flow cytometric analysis of mature B lymphocytes: To obtain a single cell suspension from spleen, the spleens of mice listed below were gently passaged through a 30 μm cell strainer. Single cells were resuspended in PBS supplemented with 3% heat inactivated foetal calf serum (FCS; Gibco®). The following antibodies were used for staining:

[0974] Antibody against B220/CD45R conjugated with allophycocyanin (APC) (eBioscience, clone RA3-6B2), antibody against IgD receptor conjugated with phycoerythrin (PE) (eBioscience, clone 11-26) and IgM receptor conjugated with fluorescein isothiocyanate (FITC) (eBioscience, clone 11/41).

[0975] 5×10.sup.6 cells were used for each staining. To each vial containing splenocytes a cocktail of antibodies was added consisting of; IgD (PE) (eBioscience, clone 11-26), IgM (FITC) and B220/CD45R (APC). Cells were incubated at 6° C. for 15 minutes, washed to remove excess of unbound antibodies and analysed using a fluorescence-activated cell sorting (FACS) analyser from Miltenyi Biotech. B-cells were gated as B220.sup.+IgM.sup.+IgD.sup.− for T1 population, B220.sup.+IgM.sup.+IgD.sup.+ for T2 population and B220.sup.+IgM.sup.+IgD.sup.− for M population. Percentage of cells was calculated using gating system.

Results

[0976] Four different genotypes of mice were generated:— [0977] Wild type (WT): [0978] A transgenic mouse homozygous for a heavy chain transgene comprising insertion of the 1.sup.st BAC human DNA noted above in which there are 6 human VH, all functional human D and JH gene segments (S1/S1); [0979] A transgenic mouse homozygous for a heavy chain transgene comprising insertion of a human VH, all functional human D and JH gene segments (H1/H1); and [0980] A transgenic mouse homozygous for a kappa chain transgene comprising insertion of 6 functional human Vκ and 5 functional Jκ gene segments (K1/K1).

[0981] Spleens from these naïve mice were collected and analysed for their B cell compartments. The number and percentages of T1, T2 and M cells among those mice are similar (FIG. 53), indicating that genetic manipulation of endogenous IG loci in transgenic mice according to the invention do not compromise their B cell development. These data help to establish that animals according to the invention provide a robust platform for antibody discovery.

[0982] As explained in Example 16 below, further analysis was performed on S1 mice in which endogenous heavy chain expression has been inactivated (S1F mice in which there is inactivation by inversion as herein described). As explained, normal splenic and bone marrow compartments are seen in such mice of the invention (ie, equivalent to the compartments of mice expressing only mouse antibody chains).

Example 11

Normal IgH Isotypes & Serum Levels in Transgenic Animals of the Invention

[0983] Transgenic mice (H1) carrying all human JH, all human DH and human Vh2-5 under control of a rat switch region or mice (S1) carrying all human JH, all human OH and human Vh2-5, Vh7-41, Vh4˜4, Vh1-3, Vh1-2 and Vh6-1 under control of a mouse switch region were immunised with 100 μg Cholera Toxin B subunit (CTB; Sigma-Aldrich® C9903) emulsified in Complete Freund's Adjuvant CFA; Sigma-Aldrich® F 5881). At least three animals were injected sc or ip and then boosted with 25 μg antigen in Incomplete Freund's Adjuvant (IFA; Sigma-Aldrich® F 5506) at (i) 14 days and 21 days or (ii) 28 days after priming. Blood was taken before priming at day “−1” (pre-bleeds) and on the day the spleens were taken (usually 4d after last boost). Serum was analysed by ELISA using an antigen independent assessment of Ig isotypes. This assay detects total serum antibodies of all species. Specific detection for mouse IgG1, IgG2a, IgG2b and IgM was used ((Anti-mouse IgG1 HRP AbD Serotec STAR132P, Anti-mouse IgG2a HRP AbD Serotec STAR133P, Anti-mouse IgG2b HRP AbD Serotec STAR134P, Anti-mouse IgM HRP Abcam® ab97230) and concentrations were read off a standard curve produced for each isotype using polyclonal isotype controls (IgG1, Kappa murine myeloma Sigma-Aldrich® M9269, IgG2a, Kappa murine myeloma Sigma-Aldrich® M9144, IgG2b, Kappa from murine myeloma Sigma-Aldrich® M8894, IgM, Kappa from murine myeloma Sigma-Aldrich® M3795). Results (FIGS. 54 & 55 for H1 homozygous and S1 homozygous and heterozygous mice) showed that even with these relatively short immunisation regimes mice showed an increase in overall IgG levels after immunisation over pre-bleeds. In cases where control mice (+/+) not carrying any human immunoglobulin genes were included and immunised, these mice showed comparable changes in total observed Ig levels (FIG. 54). Individual isotype levels were more variable between animals possibly showing various stages of class switching. IgM levels never exceeded 800 μg/ml whereas IgG levels reached more than 6 mg/ml in some animals. Non-immunised controls showed no such increases in switched isotype Ig levels.

[0984] These results demonstrate that mice comprising multiple human VDJ gene segments under the control of a rat Sp rat or mouse switch are able to undergo productive recombination and class switching in response to antigen challenge and that the mice produce antibody levels that are broadly comparable to unmodified mice The transgenic mice are able to produce antibodies of each of the IgG1, IgG2a, IgG2b and IgM isotypes after immunisation. Titers for CTB-specific Ig in pre-bleeds and terminal bleeds were determined and all immunised animals showed at CTB-specific titres of at least 1/100 000.

Example 12

[0985] Generation of Anti-Ovalbumin Antibodies with Sub-50 nm Affinities from Animals of the Invention

[0986] Transgenic mice carrying all human JH, all human DH and human Vh2-5 under control of a rat Sp switch region were immunised with 25 μg ovalbumin (OVA; Sigma-Aldrich® A7641) in Sigma-Aldrich® adjuvant (Sigma Adjuvant System® S6322) ip and then boosted with the same amount of OVA in adjuvant at day 14 and day 21. Spleenocytes were taken 4 days later and fused using 1 ml polyethyleneglycol (PEG Average MW1450; Sigma-Aldrich® P7306) with a myeloma line. Fused hybridoma cells were plated on 5 96-well plates and after selection with hypoxanthine-aminopterin-thymidine (HAT) wells tested for expression of OVA-specific antibodies by ELISA. Clones positive by ELISA were re-tested by surface plasmon resonance (SPR) and binding kinetics determined using the ProteOn™ XPR36 (Bio-Rad®). Briefly, anti-mouse IgG (GE Biacore™ BR-1008-38) was coupled to a GLM biosensor chip by primary amine coupling, this was used to capture the antibodies to be tested directly from tissue culture supernatants. Ovalbumin was used as the analyte and passed over the captured antibody surface at 1024 nM, 256 nM, 64 nM, 16 nM, 4 nM with a 0 nM (i.e. buffer alone) used to double reference the binding data. Regeneration of the anti-mouse IgG capture surface was by 10 mM glycine pH1.7, this removed the captured antibody and allowed the surface to be used for another interaction. The binding data was fitted to 1:1 model inherent to the ProteOn™ XPR36 analysis software. The run was carried out 1×HBS-EP (10 mM Hepes, 150 mM NaCl, 3 mM EDTA, 0.05% polysorbate, pH7.6 (Teknova H8022)) used as running buffer and carried out at 25° C.

[0987] For 8 positive clones, heavy chain V-regions were recovered by RT-PCR (Access RT-PCR System, A1250, Promega) using forward primers specific for Ig signal sequences (Wardemann et al Science 301, 1374 (2003)) and the following reverse primers for the constant regions of mouse IgG (Table 3):

TABLE-US-00003 TABLE 3 Primer Name Sequence bp   migG1_2 rev GGGGCCAGTGGATAGACA 21 SEQ ID No 33 GAT migG2b rev CAGTGGATAGACTGATGG 18 SEQ ID No 34 migG2a_2 rev CAGTGGATAGACCGATGG 21 SEQ ID No 35 mCH1 unirev KCAGGGGCCAGTGGATAG 20 SEQ ID No 36 AC mCH1 unirev_2 TARCCYTTGACMAGGCAT 20 SEQ ID No 37 CC

[0988] RT-PCR products were either directly sequenced using the same primer pairs or cloned in to TA plasmids (TOPO® TA Cloning® Kit for Sequencing, K4595-40, Invitrogen™) and submitted for plasmid sequencing. Results (Table 4, below) show that CDRH3 sequences had variable CDRs except for two identical clones (16C9 and 20B5) that also had near identical KD kinetic values. The determined equilibrium binding constant KD ranged from 0.38 nM to 40.60 nM, as determined by SPR at 25° C.

[0989] These results demonstrate that mice comprising multiple human VDJ gene segments under the control of a rat Cμ switch are able to undergo productive recombination and produce high affinity antigen-specific antibodies whose CDR3 regions have sequences encoded by human gene segments (human JH was separately identified by V-Quest, IMGT).

TABLE-US-00004 TABLE 4 CDR3 and FR4 (underlined) KD clone according to [nM] code Kabat definition  0.38 16C9 QEVINYYYYGMDVWGQGTTVTVSS SEQ ID No 38  0.52 20B5 QEVINYYYYGMDVWGQGTTVTVSS SEQ ID No 39  5.89 19F4 LEMATINYYYYGMDVWGQGTMVTV SEQ ID No 40 SS 39.70 19E1 QEFGNYYYYGMDVWGQGTTVTVSS SEQ ID No 41  3.10 19G8 QEDGNPYYFGMDFWGQGTTVTVSS SEQ ID No 42  8.95 20H10 GSSYYYDGMDVWGQGTTVTVSS SEQ ID No 43  4.46 18010 LENDYGYYYYGMDVWGQGTTVTV SEQ ID No 44 SS 40.60 16F2 RGGLSPLYGMDVWGQGTTVTVSS SEQ ID No 45

Example 13

[0990] Generation of Anti-Cholera Toxin B Antibodies With Human Vh Regions from Animals of the Invention

[0991] Transgenic mice carrying all human JH, all human DH and human Vh2-5, Vh7-41, Vh4-4, Vh1-3, Vh1-2 and Vh6-1 under control of a mouse Sp switch region were immunised and fused as described in Example 11. Fused hybridoma cells were plated on 5 96-well plates and after selection with hypoxanthine-aminopterin-thymidine (HAT) or G418 (Gibco® Cat No 10131-027, Lot 503317) and wells tested for expression of CTB-specific antibodies by ELISA. Clones positive by ELISA were re-tested by surface plasmon resonance SPR and binding kinetics determined using the ProteOn XPR36™ (Bio-Rad®); Briefly, anti-mouse IgG (GE Biacore™ BR-1008-38) was coupled to a GLM biosensor chip by primary amine coupling, this was used to capture the antibodies to be tested directly from tissue culture supernatants. Cholera toxin B was used as analyte and passed over the captured antibody surface at 256 nM, 64 nM, 16 nM, 4 nM and 1 nM, with a 0 nM (i.e. buffer alone) used to double reference the binding data. Regeneration of the anti-mouse IgG capture surface was by 10 mM glycine pH1.7, this removed the captured antibody and allowed the surface to be used for another interaction. The binding data was fitted to 1:1 model inherent to the ProteOn XPR36 analysis software. The run was carried out 1×HBS-EP (10 mM Hepes, 150 mM NaCl, 3 mM EDTA, 0.05% polysorbate, pH7.6 (Teknova H8022)) used as running buffer and carried out at 37° C.

[0992] From the clones initially identified by ELISA, binding to CTB was confirmed by SPR. However, due to the pentameric nature of the cholera toxin B, the majority of fits to the 1:1 model were poor and the equilibrium binding constant KDs could not be accurately determined. Where fits were acceptable, equilibrium binding constant KDs determined ranged from 0.21 nM to 309 nM but due to the pentameric nature of cholera toxin B these are likely to be the result of multimeric interactions and therefore apparent affinities with possible avidity components.

[0993] Clones identified by SPR for binding to CTB were subjected to RT-PCR as described in Example 12 to recover the Vh regions. RT-PCR products were directly sequenced using the same primer pairs. Results were obtained for only 14 clones presumably because the human primers described in Wardemann et at were not designed to amplify mouse Vh regions and therefore may have failed to amplify certain mouse Vh classes. Results showed that 3 of the 14 CTB-specific recovered heavy chain V-region sequences were human V, D and J regions as identified by V-Quest, IMGT (Table 5).

TABLE-US-00005 TABLE 5 Alignment of heavy chain CDRs and J-region of 3 clones identified as binding to CTB and preferentially matching with human reference sequences from IMGT database; note that the KD values given here are apparent values due to the avidity of the CTB-antibody interaction indicates data missing or illegible when filed

Example 14

[0994] High Human Lambda Variable Region Expression in Transgenic Mice Comprising Human Lambda Gene Segments Inserted into Endogenous Kappa Locus

[0995] Insertion of human lambda gene segments from a 1.sup.st IGL BAC to the IGK locus of mouse AB2.1 ES cells (Baylor College of Medicine) was performed to create a chimaeric light chain allele denoted the P1 allele (FIG. 56). The inserted human sequence corresponds to the sequence of human chromosome 22 from position 23217291 to position 23327884 and comprises functional lambda gene segments Vλ3-1, Jλ1-Cλ1, Jλ2-Cλ2, Jλ3-Cλ3, Jλ6-Cλ6 and Jλ7-Cλ7. The insertion was made between positions 70674755 and 706747756 on mouse chromosome 6, which is upstream of the mouse Cκ region and 3′Eκ (ie, within 100 kb of the endogenous light chain enhancer) as shown in FIG. 56. The mouse Vκ and Jκ gene segments were retained in the chimaeric locus, immediately upstream of the inserted human lambda DNA. The mouse lambda loci were left intact. Mice homozygous for the chimaeric P1 locus were generated from the ES cells using standard procedures.

[0996] A second type of mice were produced (P2 mice) in which more human functional Vλ gene segments were inserted upstream (5′) of human Vλ3-1 by the sequential insertion of the BAC1 human DNA and then BAC2 DNA to create the P2 allele. The inserted human sequence from BAC2 corresponds to the sequence of human chromosome 22 from position 23064876 to position 23217287 and comprises functional lambda gene segments Vλ2-18, Vλ3-16, V2-14, Vλ3-12, Vλ2-11, Vλ3-10, Vλ3-9, Vλ2-8 and Vλ4-3. Mice homozygous for the chimaeric P2 locus were generated from the ES cells using standard procedures.

[0997] FACS analysis of splenic B cells from the P1 and P2 homozygotes was performed to assess lambda versus kappa expression and human lambda versus mouse lambda expression in the transgenic mice.

[0998] Standard 5′-RACE was carried out to analyse RNA transcripts from the light chain loci in P2 homozygotes.

Light Chain Expression & FACS Analysis

[0999] To obtain a single cell suspension from spleen, the spleen was gently passage through a 30 μm cell strainer. Single cells were resuspended in Phosphate-Buffered Saline (PBS) supplemented with 3% heat inactivated foetal calf serum (FCS).

[1000] The following antibodies were used for staining:

[1001] Rat anti-mouse lambda (mCλ) phycoerythrin (PE) antibody (Southern Biotech), rat anti-mouse kappa (mCκ) (BD Pharmingen, clone 187.1) fluorescein isothiocyanate (FITC), anti-human lambda (hCλ) (eBioscience, clone 1-155-2) phycoerythrin (PE), anti-B220/CD45R (eBioscience, clone RA3-6B2) allophycocyanin (APC). NB: light chains bearing human Cλ was expected to have variable regions derived from the rearrangement of inserted human Vλ and human Jλ. Light chains bearing mouse Cλ was expected to have variable regions derived from the rearrangement of mouse Vλ and Jλ from the endogenous lambda loci.

[1002] 5×10.sup.6 cells were added to individual tubes, spun down to remove excess of fluid, and resuspended in fresh 100 μl of PBS+3% FCS. To each individual tube the following antibodies were added:

[1003] For staining of mλ versus mκ 1 μl of each antibody was added in addition to 1 μl of B220/CD45R antibody. For detection of B cells expressing human lambda light chain, the mλ antibody was substituted with hλ antibody. Cells were incubated in the dark at 6° C. for 15 minutes followed by several washes with fresh PBS+3% FCS to remove unbound antibody. Cells were analysed using fluorescence-activated cell sorting (FACS) analyser from Miltenyi Biotech.

[1004] Alive spleenocytes were gated using side scatter (SSC) and forward scatter (FSC). Within the SSC and FSC gated population, a subpopulation of B220/CD45R (mouse B-cells) was detected using the APC fluorochrome. Single positive B220/CD45R population was further subdivided into a cell bearing either mλ or hλ PE fluorochrome in conjunction with mκ FITC fluorochrome. The percentage of each population was calculated using a gating system.

[1005] Surprisingly, FACS analysis of splenic B cells from the P1 homozygotes showed no detectable mouse Cλ expression (FIG. 57), indicating that insertion of the human lambda locus DNA from BAC1 interrupts expression of the endogenous IGK chain.

[1006] The strong expression of endogenous Cλ and weak expression of human Cλ in the splenic B cells grouped by FACS analysis (mouse Cλ: human Cλ=65; 32) in these mice suggest that inserted human IGL sequence, although interrupts the IGK activity, cannot totally compete with the endogenous IGL genes.

[1007] The FACS analysis again surprisingly showed no detectable mouse Cκ expression in the P2 homozygotes (FIGS. 58A & B). However, the human Cλ greatly predominates in expressed B cells grouped as mouse or human Cλ following FACS analysis (mouse Cλ:human Cλ 15:80 corresponding to a ratio of 15 mouse lambda variable regions:80 human lambda variable regions, ie, 84% human lambda variable regions with reference to the grouped B-cells—which corresponds to 80% of total B-cells) from the P2 homozygotes. While not wishing to be bound by any theory, we suggest that the inserted human lambda locus sequence from the 2.sup.nd BAC provides some advantages to compete with endogenous lambda gene segment rearrangement or expression.

[1008] We analysed human Vλ and Jλ usage in the P2 homozygotes. See FIG. 59 which shows the human Vλ usage in P2 homozygotes. The observed usage was similar to that seen in humans (as per J Mol Biol. 1997 Apr. 25; 268(1):69-77; “The creation of diversity in the human immunoglobulin V(lambda) repertoire”; Ignatovich O et a). Further, the human Jλ usage was similar to that seen in humans (FIG. 60). The Vλ versus Vκ usage analysis of human Cλ transcripts by sequencing of non-bias 5′-RACE (rapid amplification of cDNA ends) PCR clones showed that among 278 clone sequences, only one used Vκ for rearrangement to Jλ (human Jλ), and all others (277 clones) used human Vλ (FIGS. 61 & 62; Vλ2-5 was detected at the RNA transcript level, but this is a pseudogene which is usually not picked up by usage a the protein level). While not wishing to be bound by any theory, we suggest that the retained mouse Vκ gene segments essentially cannot efficiently rearrange with the inserted human Jλ gene segments because they have the same type of RSSs (recombination signal sequences; see explanation below) and are incompatible for rearrangement (FIG. 63). This result also indicates that the inactivation of the endogenous IGK activity and predominate expression of the inserted human lambda sequence can be achieved without further modification of the IGK locus, for example, deletion or inversion of endogenous kappa loci gene segments is not necessary, which greatly simplifies the generation of useful transgenic mice expressing light chains bearing human lambda variable regions (ie, variable regions produced by recombination of human Vλ and Jλ gene segments).

[1009] The arrangement of recombination signal sequences (RSSs) that mediate V(D)J recombination in vivo is discussed, eg, in Cell. 2002 April; 109 Suppl:S45-55; “The mechanism and regulation of chromosomal V(D)J recombination”: Bassing C H, Swat W, Alt F W (the disclosure of which is incorporated herein by reference). Two types of RSS element have been identified: a one-turn RSS (12-RSS) and a two-turn RSS (23-RSS). In natural VJ recombination in the lambda light chain locus, recombination is effected between a two-turn RSS that lies 3′ of a V lambda and a one-turn RSS that lies 5′ of a J lambda, the RSSs being in opposite orientation. In natural VJ recombination in the kappa light chain locus, recombination if effected between a one-turn RSS that lies 3′ of a V kappa and a two-turn RSS that lies 5′ of a J kappa, the RSSs being in opposite orientation. Thus, generally a two-turn RSS is compatible with a one-turn RSS in the opposite orientation.

[1010] Thus, the inventors have demonstrated how to (i) inactivate endogenous kappa chain expression by insertion of human lambda gene segments into the kappa locus; and (ii) how to achieve very high human lambda variable region expression (thus providing useful light chain repertoires for selection against target antigen)—even in the presence of endogenous lambda and kappa V gene segments. Thus, the inventors have shown how to significantly remove (lambda) or totally remove (kappa) V gene segment competition and thus endogenous light chain expression by the insertion of at least the functional human lambda gene segments comprised by BACs 1 and 2. In this example a very high level of human lambda variable region expression was surprisingly achieved (84% of total lambda chains and total light chains as explained above).

Example 15

[1011] High Human Lambda Variable Region Expression in Transgenic Mice Comprising Human Lambda Gene Segments Inserted into Endogenous Lambda Locus

[1012] Insertion of human lambda gene segments from the 1.sup.st and 2.sup.nd IGL BACs to the lambda locus of mouse AB2.1 ES cells (Baylor College of Medicine) was performed to create a lambda light chain allele denoted the L2 allele (FIG. 56). The inserted human sequence corresponds to the sequence of human chromosome 22 from position 23064876 to position 23327884 and comprises functional lambda gene segments Vλ2-18, Vλ3-16, V2-14, Vλ3-12, Vλ2-11, Vλ3-10, Vλ3-9, Vλ2-8, Vλ4-3, Vλ3-1, Jλ1-Cλ1, Jλ2-Cλ2, Jλ3-Cλ3, Jλ6-Cλ6 and Jλ7-Cλ7. The insertion was made between positions 19047551 and 19047556 on mouse chromosome 16, which is upstream of the mouse Cλ region and between Eλ4-10 and Eλ3-1 (ie, within 100 kb of the endogenous light chain enhancers) as shown in FIG. 56. The mouse Vλ and Jλ gene segments were retained in the locus, immediately upstream of the inserted human lambda DNA. The mouse kappa loci were inactivated to prevent kappa chain expression. Mice homozygous for the L2 locus were generated from the ES cells using standard procedures.

[1013] Using a similar method to that of Example 14, FACS analysis of splenic B cells from the L2 homozygotes was performed to assess lambda versus kappa expression and human lambda versus mouse lambda expression in the transgenic mice.

Light Chain Expression & FACS Analysis

[1014] The FACS analysis of splenic B-cells in L2 homozygotes under the IGK knockout background (in which Vκ and Jκ gene segments have been retained) surprisingly showed that expression of human Cλ greatly predominates in B-cells grouped as mouse or human Cλ following FACS analysis (mouse CA:human Cλ=5:93 corresponding to a ratio of 5 mouse lambda variable regions: 93 human lambda variable regions, ie, 95% human lambda variable regions with reference to the grouped B-cells—which corresponds to 93% of total B-cells) (FIG. 64A), demonstrating that inserted human IGA gene segments within the endogenous IGA locus can outcompete the endogenous IGA gene segment rearrangement or expression.

[1015] Thus, the inventors have demonstrated how to achieve very high human lambda variable region expression (thus providing useful light chain repertoires for selection against target antigen)—even in the presence of endogenous lambda and kappa V gene segments. Thus, the inventors have shown how to significantly remove endogenous lambda V gene segment competition and thus endogenous lambda light chain expression by the insertion of at least the functional human lambda gene segments comprised by BACs 1 and 2. In this example a very high level of human lambda variable region expression was surprisingly achieved (95% of total lambda chains and total light chains as explained above).

[1016] These data indicate that mice carrying either P (Example 14) or L (Example 15) alleles produced by targeted insertion of the functional gene segments provided by BAC1 and BAC2 can function in rearrangement and expression in mature B cells. These two types of alleles are very useful for providing transgenic mice that produce human Ig lambda chains for therapeutic antibody discovery and as research tools.

Transgenic Mice of the Invention Expressing Human Lambda Variable Regions Develop Normal Splenic Compartments

[1017] In spleen, B cells are characterized as immature (T1 and T2) and mature (M) based on the levels of cell surface markers, IgM and IgD. T1 cells have high IgM and low IgD. T2 cells have medium levels of both them. M cells have low IgM but high IgD (FIG. 65). See also J Exp Med. 1999 Jul. 5; 190(1):75-89; “B cell development in the spleen takes place in discrete steps and is determined by the quality of B cell receptor-derived signals”: Loder F et al.

[1018] Using methods similar to those described in Example 16 below, splenic B-cells from the animals were scored for IgD and IgM expression using FACS. We compared control mice KA/KA (in which endogenous kappa chain expression has been inactivated, but not endogenous lambda chain expression) with L2/L2; KA/KA mice (12 homozyotes). The L2 homozygotes surprisingly showed comparable splenic B-cell compartments to the control mice (FIG. 648).

Example 16

Assessment of B-Cell and Ig Development in Transgenic Mice of the Invention

[1019] We observed normal Ig subtype expression & B-cell development in transgenic mice of the invention expressing antibodies with human heavy chain variable regions substantially in the absence of endogenous heavy and kappa chain expression.

[1020] Using ES cells and the RMCE genomic manipulation methods described above, mice were constructed with combinations of the following Ig locus alleles:—

[1021] S1F/HA, +/KA=(i) S1F—first endogenous heavy chain allele has one human heavy chain locus DNA insertion, endogenous mouse VDJ region has been inactivated by inversion and movement upstream on the chromosome (see the description above, where this allele is referred to as S1.sup.inv1); (ii) HA—second endogenous heavy chain allele has been inactivated (by insertion of an endogenous interrupting sequence); (iii) +—first endogenous kappa allele is a wild-type kappa allele and (iv) KA—the second endogenous kappa allele has been inactivated (by insertion of an endogenous interrupting sequence). This arrangement encodes exclusively for heavy chains from the first endogenous heavy chain allele.

[1022] S1F/HA, K2/KA=(i) K2—the first endogenous kappa allele has two kappa chain locus DNA insertions between the most 3′ endogenous Jκ and the mouse Cκ, providing an insertion of 14 human Vκ and Jκ1-JκS; and (ii) KA—the second endogenous kappa allele has been inactivated (by insertion of an endogenous interrupting sequence). This arrangement encodes exclusively for heavy chains comprising human variable regions and substantially kappa light chains from the first endogenous kappa allele.

[1023] +/HA, K2/KA—this arrangement encodes for mouse heavy chains and human kappa chains.

[1024] +/HA, +/KA—this arrangement encodes for mouse heavy and kappa chains—the mice only produce mouse heavy and light chains.

[1025] In bone marrow. B progenitor populations are characterized based their surface markers, B220 and C043. PreProB cells carry germline IGH and IGK/L configuration and have low B220 and high CD43 on their cell surface. ProB cells start to initiate VDJ recombination in the IGH locus and carry medium levels of both B220 and CD43. PreB cells carry rearranged IGH VDJ locus and start to initiate light chain VJ rearrangement, and have high B220 but low CD43. In spleen, B cells are characterized as immature (T1 and T2) and mature (M) based on the levels of cell surface markers, IgM and IgD. T1 cells have high IgM and low IgD. T2 cells have medium levels of both them. M cells have low IgM but high IgD (FIG. 65). See also J Exp Med. 1991 May 1; 173(5):1213-25; “Resolution and characterization of pro-B and pre-pro-B cell stages in normal mouse bone marrow”; Hardy R R et al and J Exp Med. 1999 Jul. 5; 190(1):75-89; “B cell development in the spleen takes place in discrete steps and is determined by the quality of B cell receptor-derived signals”; Loder F et al.

[1026] Transgenic Mice of the Invention Develop Normal Splenic and BM Compartments

(a) Analysis of the Splenic Compartment

[1027] For each mouse, to obtain a single cell suspension from spleen, the spleen was gently passaged through a 30 μm cell strainer. Single cells were resuspended in Phosphate-Buffered Saline (PBS) supplemented with 3% heat inactivated foetal calf serum (FCS). 5×10.sup.6 cells were added to individual tubes, spun down to remove excess of fluid and resuspended in fresh 100 μl of PBS+3% FCS. To each individual tube the following antibodies were added: anti-B220/CD45R (eBioscience, clone RA3-6B2) allophycocyanin (APC), antibody against IgD receptor conjugated with phycoerythrin (PE) (eBioscience, clone 11-26) and antibody against IgM receptor conjugated with fluorescein isothiocyanate (FITC) (eBioscience, clone 11/41).

[1028] For staining of IgM vs IgD, 5×10.sup.5 cells were used for each staining. To each vial containing splenocytes a cocktail of antibodies was added consisting of: anti-IgD (PE), anti-IgM (FITC) and anti-B220/CD45R (APC). Cells were incubated at 6° C. for 15 minutes, washed to remove excess unbound antibodies and analysed using a fluorescence-activated cell sorting (FACS) analyser from Miltenyi Biotech. B-cells were gated as B220.sup.HIGH IgM.sup.HIGH IgD.sup.LOW (ie, B220.sup.+ IgM.sup.+ IgD.sup.−) for T1 population, B220.sup.HIGH IgD.sup.HIGH (B220.sup.+ IgM.sup.+ IgD.sup.+) for T2 population and B220.sup.HIGH IgM.sup.LOW IgD.sup.HIGH (B220.sup.+ IgM.sup.− IgD.sup.+) for M population. Percentage of cells was calculated using gating system. We used gates to identify and define subsets of cell populations on plots with logarithmic scale. Before gates are applied a single stain antibody for each fluorochrome is used to discriminate between a positive (high intensity fluorochrome) and negative (no detectable intensity fluorochrome) population. Gates are applied based on fluorochrome intensities in the same manner to all samples. The single stains were:

IgD-PE

IgM-FITC

B220-APC

[1029] Alive spleenocytes were gated using side scatter (SSC) and forward scatter (FSC). Within the SSC and FSC gated population, a subpopulation of B220/CD45R positive cells (mouse B-cells) was detected using the APC fluorochrome. The single positive B220/CD45R population was further subdivided into a cell bearing either IgM fluorescein isothiocyanate (FITC) or IgD fluorochrome in conjunction with mκ FITC fluorochrome. The percentage of each population was calculated using gating system. The splenic B-Cell compartments in the mice of the invention are normal (ie, equivalent to the compartments of mice expressing only mouse antibody chains).

(b) Bone Marrow B Progenitor Analysis

[1030] To obtain a single cell suspension from bone marrow for each mouse, the femur and tibia were flushed with Phosphate-Buffered Saline (PBS) supplemented with 3% heat inactivated foetal calf serum (FCS). Cells were further passage through a 30 μm cell strainer to remove bone pieces or cell clumps. Cells were resuspended in cold PBS supplemented with 3% serum. 2×10.sup.6 cells were added to individual tubes, spun down to remove excess of buffer, and resuspended in fresh 100 μl of PBS+3% FCS. To each individual tube the following antibodies were added: anti-Leukosialin (CD43) fluorescein isothiocyanate (FITC) (eBioscience, clone eBioR2/60) and anti-B220/CD45R (eBioscience, clone RA3-6B2) allophycocyanin (APC). Cells were incubated in the dark at 6° C. for 15 minutes followed by several washes with fresh PBS+3% FCS to remove unbound antibody. Cells were analysed using a fluorescence-activated cell sorting (FACS) analyser from Miltenyi Biotech. Alive bone marrow cells were gated using side scatter (SSC) and forward scatter (FSC). We used gates to identify and define subsets of cell populations on plots with logarithmic scale. Before gates are applied a single stain antibody for each fluorochrome is used to discriminate between a positive (high intensity fluorochrome) and negative (no detectable intensity fluorochrome) population. Gates are applied based on fluorochrome intensities in the same manner to all samples. The single stains were:

B220-APC

CD43-FITC

[1031] Within the alive population a double population of B220/CD45R and CD43 positive cells was identified as a pre-B, pro-B and pre-pro B cells. The splenic B-Cell compartments in the mice of the invention are normal (ie, equivalent to the compartments of mice expressing only mouse antibody chains).

Transgenic Mice of the Invention Develop Normal Ig Expression

Quantification of Serum IgM and IgG

[1032] 96-well NUNC plates were coated initially with a capture antibody (goat anti-mouse Fab antibody at 1 μg/ml) overnight at 4° C.). The IgG plates used anti-Fab, (M4155 Sigma) and the IgM plates used anti-Fab (OBT1527 AbD Serotec). Following three washes with phosphate buffer saline (PBS) containing 0.1% v/v Tween20, plates were blocked with 200 μl of PBS containing 1% w/v bovine serum albumin (BSA) for 1 hour at room temperature (RT). The plates were washed three times as above and then 50 μl of standards (control mouse isotype antibodies, IgG1 (M9269 Sigma), IgG2a (M9144 Sigma), IgG2b (M8894 sigma), IgM (M3795 Sigma) or serum samples diluted in PBS with 0.1% BSA were added to each well, and incubated for 1 hour at RT. After washing three times as above 100 μl of detection antibody (goat anti-mouse isotype specific antibody-horseradish peroxidase conjugated, 1/10000 in PBS with 0.1% Tween) (anti-mouse IgG1 (STAR132P AbD Serotec), anti-mouse IgG2a (STAR133P AdD Serotec), anti-mouse IgG2b (STAR134P AbD Serotec) and anti-mouse IgM (ab97230 Abcam) were added into each well and incubated for 1 hour at RT. The plates were washed three times as above and developed using tetramethylbenzidine substrate (TMB, Sigma) for 4-5 minutes in the dark at RT. Development was stopped by adding 50 μl/well of 1 M sulfuric acid. The plates were read with a Biotek Synergy HT plate reader at 450 nm.

Conclusion:

[1033] Inversion of endogenous V.sub.H-D-J.sub.H following the human IGH BAC insertion results in inactivation of rearrangement of endogenous V.sub.H to inserted human D-J.sub.H. The inventors observed, however, that surprisingly the inactivation of endogenous heavy chain expression does not change the ratio of B-cells in the splenic compartment (FIG. 66) or bone marrow B progenitor compartment (FIG. 67) and the immunoglobulin levels in serum are normal and the correct Ig subtypes are expressed (FIG. 68). This was shown in mice expressing human heavy chain variable regions with mouse light chains (FIGS. 66A and 67A) as well as in mice expressing both human heavy chain variable regions and human light chain variable regions (FIGS. 668 and 678). These data demonstrate that inserted human IGH gene segments (an insertion of at least human V.sub.H gene segments V.sub.H2-5, 7-4-1, 4-4, 1-3, 1-2, 6-1, and all the human D and J.sub.H gene segments D1-1, 2-2, 3-3, 4-4, 5-5, 6-6, 1-7, 2-8, 3-9, 5-12, 6-13, 2-15, 3-16, 4-17, 6-19, 1-20, 2-21, 3-22, 6-25, 1-26 and 7-27; and J1, J2, J3, J4, J5 and J6) are fully functional in the aspect of rearrangement, BCR signalling and B cell maturation. Functionality is retained also when human light chain VJ gene segments are inserted to provide transgenic light chains, as per the insertion used to create the K2 allele. This insertion is an insertion comprising human gene segments Vκ2-24, Vκ3-20, Vκ1-17, Vκ1-16, Vκ3-15, Vκ1-13, Vκ1-12, Vκ3-11, Vκ1-9, Vκ1-8, Vκ1-6, Vκ1-5, Vκ5-2, Vκ4-1, Jκ1, Jκ2, Jκ3, Jκ4 and Jκ5. Greater than 90% of the antibodies expressed by the S1F/HA; K2/KA mice comprised human heavy chain variable regions and human kappa light chain variable regions. These mice are, therefore, very useful for the selection of antibodies having human variable regions that specifically bind human antigen following immunisation of the mice with such antigen. Following isolation of such an antibody, the skilled person can replace the mouse constant regions with human constant regions using conventional techniques to arrive at totally human antibodies which are useful as drug candidates for administration to humans (optionally following mutation or adaptation to produce a further derivative, eg, with Fc enhancement or inactivation or following conjugation to a toxic payload or reporter or label or other active moiety).

[1034] A further experiment was carried out to assess the IgG and IgM levels and relative proportions in transgenic mice of the invention that express antibodies that have human heavy and light (kappa) variable regions (S1F/HA, K2/KA mice: n=15). These were compared against 12 mice expressing only mouse antibody chains (+/HA, +/KA (n=6) and wild-type mice (WT; n=6)). The results are tabulated below (Table 6) and shown in FIG. 69.

[1035] It can be seen that the mice of the invention, in which essentially all heavy chain variable regions are human heavy chain variable regions, expressed normal proportions of IgM and IgG subtypes, and also total IgG relative to IgM was normal.

TABLE-US-00006 TABLE 6 Total IgG1 IgG2a IgG2b IgM IgG + IgM (μg/mL) (μg/mL) (μg/mL) (μg/mL) (μg/mL) KMCB22.1a 30.5 38.3 49.9 224.4 343.1 S1F/HA, K2/KA KMCB 19.1d 103.6 181.2 85.6 351.7 722.1 S1F/HA, K2/KA KMCB 19.1h 191.4 456.6 383.3 643.2 1674.6 S1F/HA, K2/KA KMCB 20.1a 53.6 384.4 249.7 427.1 1114.7 S1F/HA, K2/KA KMCB 20.1c 87.3 167.0 125.7 422.1 802.1 S1F/HA, K2/KA KMCB 20.1f 55.4 177.2 95.6 295.7 623.9 S1F/HA, K2/KA KMCB22.1f 61.1 56.3 111.4 245.8 474.5 S1F/HA, K2/KA KMCB23.1c 71.4 70.7 80.5 585.4 808.0 S1F/HA, K2/KA KMCB23.1d 65.4 148.7 187.4 255.4 657.0 S1F/HA, K2/KA KMCB24.1f 60.0 56.6 150.5 294.8 561.9 S1F/HA, K2/KA KMCB13.1a 101.2 200.5 269.8 144.1 715.7 S1F/HA, K2/KA KMCB13.1d 124.5 117.5 246.6 183.2 671.9 S1F/HA, K2/KA KMCB17.1f 58.3 174.2 116.2 218.1 566.8 S1F/HA, K2/KA KMCB14.1a 51.9 46.5 27.9 222.2 348.6 S1F/HA, K2/KA KMCB14.1b 11.5 54.2 48.5 194.4 308.6 S1F/HA, K2/KA KMCB19.1e +/ 233.0 6.7 465.6 420.9 1126.3 HA, +/KA KMCB19.1f +/ 154.3 4.6 610.2 435.7 1204.8 HA, +/KA KMCB19.1l +/ 113.5 1.1 246.8 374.6 736.0 HA, +/KA KMCB20.1e +/ 561.0 4.3 614.3 482.1 1661.7 HA, +/KA KMCB13.1e +/ 439.3 17.1 584.1 196.9 1237.3 HA, +/KA KMCB14.1c +/ 93.4 1.3 112.0 106.8 313.6 HA, +/KA KMWT 1.3c WT 212.9 155.2 104.6 233.7 706.4 KMWT 1.3d WT 297.1 203.2 144.6 248.6 893.5 KMWT 1.3e WT 143.1 174.2 619.1 251.8 1188.2 KMWT 1.3f WT 218.8 86.8 256.1 294.8 856.4 KMWT 1.3b WT 150.2 114.2 114.7 225.6 604.7 KMWT 3.1e WT 125.9 335.5 174.6 248.9 884.9

Other Embodiments

[1036] From the foregoing description, it will be apparent that variations and modifications may be made to the invention described herein to adopt it to various usages and conditions. Such embodiments are also within the scope of the following claims.

[1037] The recitation of a listing of elements in any definition of a variable herein includes definitions of that variable as any single element or combination (or subcombination) of listed elements. The recitation of an embodiment herein includes that embodiment as any single embodiment or in combination with any other embodiments or portions thereof.

[1038] All publications and patent applications mentioned in the specification are indicative of the level of skill of those skilled in the art to which this invention pertains. All publications and patent applications are herein incorporated by reference to the same extent as if each individual publication or patent application was specifically and individually indicated to be incorporated by reference.