NZ623145B2 - Humanized il-6 and il-6 receptor - Google Patents

Abstract

Disclosed is a genetically modified mouse comprising a replacement at an endogenous mouse IL-6 locus of a mouse gene encoding IL-6 with a human gene encoding human IL-6, wherein the human gene encoding human IL-6 is under control of endogenous mouse regulatory elements at the endogenous mouse IL-6 locus. Also disclosed is a genetically modified mouse, comprising a humanisation of an endogenous mouse IL-6R? gene, wherein the humanisation comprises a replacement of mouse IL-6R? ectodomain-encoding sequence with human IL-6R? ectodomain-encoding sequence at the endogenous mouse IL-6R? locus, and wherein the humanised IL-6R? gene is under control of endogenous mouse regulatory elements. ocus. Also disclosed is a genetically modified mouse, comprising a humanisation of an endogenous mouse IL-6R? gene, wherein the humanisation comprises a replacement of mouse IL-6R? ectodomain-encoding sequence with human IL-6R? ectodomain-encoding sequence at the endogenous mouse IL-6R? locus, and wherein the humanised IL-6R? gene is under control of endogenous mouse regulatory elements.

Description

HUMANIZED IL-6 AND IL-6 RECEPTOR FIELD OF INVENTION Non-human animals having a replacement of the endogenous non-human animal IL-6 and/or IL-6 receptor genes are provided. IL-6 and/or IL-6 receptor genes of the nonhuman animal are replaced, at the endogenous man loci, with human IL-6 and/or humanized IL-6 receptor genes comprising human sequence. Non-human animals that have human IL-6 and/or humanized IL-6 receptor genes, wherein the non-human animals do not exhibit one or more pathologies that are characteristic of non-human animals transgenic for human IL-6.

BACKGROUND Mice transgenic for a human IL-6 gene are known in the art. However, random insertion of a human IL-6 transgene into the mouse genome results in poorly regulated expression of the human IL-6 protein, which manifests itself in a variety of pathologies in such transgenic mice, including, but not limited to, plasmacytosis and glomerulonephritis. As a result, these mice have limited usefulness.

There is a need for non-human s, e.g., mice and rats, the express human or humanized IL-6 and/or human or humanized IL-6 receptor. There is a need for such humanized mice that do not exhibit one or more pathologies exhibited by transgenic hIL-6 mice. [0003a] In one aspect, the present invention es a genetically modified mouse comprising a replacement at an endogenous mouse IL-6 locus of a mouse gene encoding IL-6 with a human gene encoding human IL-6, wherein the human gene encoding human IL- 6 is under control of endogenous mouse tory elements at the endogenous mouse IL-6 locus. [0003b] In another aspect, the present invention provides a genetically modified mouse, comprising a humanization of an endogenous mouse IL-6Rα gene, wherein the humanization comprises a replacement of a mouse IL-6Rα main-encoding sequence with a human IL-6Rα main-encoding sequence at the endogenous mouse IL-6Rα locus, and wherein the zed IL-6Rα gene is under l of endogenous mouse regulatory elements. [0003c] In another aspect, the present invention provides a method for making a humanized mouse, comprising replacing a mouse gene sequence encoding mouse IL-6 with a human gene encoding human IL-6 so that the human IL-6 gene is under control of endogenous mouse regulatory elements. [0003d] In another aspect, the present invention provides a method for making a humanized mouse, comprising replacing all mouse exons encoding ectodomain ces of mouse IL-6Rα with a human genomic fragment encoding human IL-6Rα main to form a humanized IL-6Rα gene, wherein the humanized IL-6Rα gene is under control of endogenous mouse regulatory elements. [0003e] In another aspect, the present invention provides a genetically modified mouse comprising a zed IL-6Rα gene comprising a replacement of mouse ectodomain-encoding sequence with human ectodomain sequence, wherein the zed IL-6Rα gene comprises a mouse transmembrane sequence and a mouse cytoplasmic sequence; wherein the mouse further comprises a gene encoding a human IL-6, wherein the genes encoding human IL-6 and humanized IL-6Rα are under control of endogenous mouse regulatory elements.

In another aspect, genetically ed man animals are provided that comprise a replacement at an endogenous IL-6 and/or IL-6 or locus of a gene encoding an endogenous IL-6 and/or IL-6 receptor with a gene encoding a human or humanized IL-6 and/or IL-6 receptor. Murine animals are provided that comprise a ement of an endogenous IL-6 gene, at an endogenous murine IL-6 locus, with a human IL-6 gene; and/or that comprise a ement of an endogenous IL-6 receptor gene (or nucleotide sequence encoding an ectodomain thereof) with a human IL-6 receptor gene (or nucleotide sequence encoding an ectodomain thereof).

In another , genetically modified murine animals are provided that express a human IL-6 gene under the control of endogenous murine promoter and/or endogenous murine regulatory elements, from an endogenous murine IL-6 locus.

In another , genetically ed murine animals are provided that express a human IL-6 receptor gene (or a gene encoding a human ectodomain and murine embrane and intracellular domains) under the control of endogenous murine promoter and/or endogenous murine tory elements, from an endogenous murine lL-6 receptor locus.

In one aspect, a genetically modified animal (e.g., a murine , e.g., a mouse or rat) is provided that expresses a human lL-6 protein, wherein the non- human animal does not exhibit a pathology selected from plasmacytosis, glomerulonephritis, glomerulosclerosis, mesangio-proliferative glomerulonephritis, intestinal lymphoma, kidney lymphoma, megaly, lymph node enlargement, liver enlargement, megakaryocytes in bone marrow, compacted abnormal plasma cells, infiltration of plasma cells into lung or liver or kidney, mesangial cell proliferation in kidney, cerebral overexpression of lL-6, ramified microglial cells in white matter, reactive astrocytosis in brain, kidney failure, elevated megakaryocytes in spleen, muscle wasting (e.g., gastrocnemius muscle wasting), elevated muscle cathepsins B and B+L (e.g., around 20-fold and 6-fold), and a combination f.

In one embodiment, the man animal ses a normal B cell population. In one embodiment, the normal B cell population is approximately the same in number and immunophenotype as a wild-type animal, e.g., a wild-type mouse.

In one embodiment, the non-human animal is murine (e.g., a mouse or rat) and expresses human lL-6 (hlL-6) in serum at a level below about 800 pg/mL, below about 700, 600, 500, 400, 300, or 200 pg/mL. In a specific embodiment, the murine animal expresses hlL-6 in serum at a level of about 50 to about no more than 200 pg/mL, in another embodiment about 75-125 pg/mL, in another embodiment at about 100 pg/mL.

In one aspect, a non-human animal is ed that expresses hlL-6 and/or hlL-6R, wherein the non-human animal expresses hlL-6 and/or hlL-6R from an endogenous non-human lL-6 locus and/or an endogenous non-human hlL-6R locus.

In a specific ment, the non-human animal is murine (e.g., mouse or rat).

In one , a genetically modified mouse is provided that expresses hlL-6 from an endogenous mouse lL-6 locus, n the endogenous mouse lL-6 gene has been replaced with a hlL-6 gene.

In one embodiment, the mouse comprises a cell that expresses an lL-6 receptor (IL-6R) that comprises a human ectodomain on the surface of the cell. In one embodiment, the cell is a lymphocyte. In one ment, the lymphocyte is a B cell.

In one embodiment, about 6.8 kb at the endogenous mouse lL-6 locus, including exons 1 through 5 and a 3’ untranslated sequence, is deleted and replaced with about 4.8 kb of human lL-6 gene sequence comprising exons 1 through 5 of the human lL-6 gene. In a specific embodiment, the human lL-6 gene ses exons 1 through 5 of the human lL-6 gene of human BAC CTD-2369M23.

In one aspect, a genetically modified mouse is provided that expresses lL-6 from a human lL-6 gene, wherein the mouse expresses human lL-6 in its serum.

In one embodiment, the mouse serum exhibits a serum concentration of human lL-6 of about 25 to about 300 pg/mL, 50 to about 250 pg/mL, 75 to about 200 pg/mL, or 100 to about 150 pg/mL. In a specific embodiment, the level of human lL-6 in the serum of the mouse is about 100 pg/mL.

In one ment, the level of a pan B cell-specific marker in bone marrow of the mouse is about the same as that of a ype mouse. In one embodiment, the level of a pan B cell-specific marker in spleen is about the same as that of a wild- type mouse. In one embodiment, the pan B cell-specific marker is selected from B220, CD19, CD20, CD22, CD79a, CD79b, L26, and Pax-5 (BSAP).

In one , a genetically modified mouse is provided that expresses hlL6, wherein the mouse does not exhibit a feature selected from plasmacytosis, splenomegaly, lymph node enlargement, compacted abnormal plasma cells, and a combination thereof.

In one embodiment, the mouse comprises a spleen that is about the same weight (per body weight) as a wild-type mouse. In one embodiment, the lymph nodes of the mouse are about the same weight (per body ) as a wild-type mouse. In one embodiment, plasma cells of the mouse do not exhibit plasmocytosis characteristic of mice that overexpress human lL-6.

In one embodiment, the mouse does not t glomerulonephritis.

In one embodiment, the mouse exhibits a mesangial cell level comparable to a wild-type mouse.

In one aspect, a genetically modified mouse is provided that expresses hlL6 from an endogenous mouse lL-6 locus, wherein the endogenous mouse lL-6 gene has been replaced with a hlL-6 gene, wherein the mouse does not exhibit a feature selected from a logically detectable neuropathology, a reactive astrocytosis, and a combination thereof. In one ment, the mouse ses a brain that is morphologically indistinct from a ype mouse brain. In one embodiment, the mouse comprises brain tissue that exhibits a level of reactive astrocytosis that is no higher than that of a wild-type mouse.

In one embodiment, the mouse does not express human lL-6 in neurons. In one embodiment, the mouse comprises activated astrocyte levels that are comparable to activated astrocyte levels in a wild-type mouse.

In one embodiment, the mouse comprises ramified microglial cells in its white matter, wherein the ramified lial cells are present in an amount equivalent to an amount of ramified microglial cells in a wild-type mouse.

In one embodiment, the mouse does not t a reactive atrocytosis. In one ment, the white matter of the mouse is morphologically indistinct from the white matter of a wild-type mouse. In one embodiment, the white matter of the mouse is histologically indistinct from a wild-type mouse white matter with t to hemical staining of reactive astrocytes.

In one embodiment, the mouse comprises a brain that is morphologically inct from a wild-type mouse brain. In one embodiment, the mouse comprises brain tissue that exhibits a level of reactive astrocytosis that is no higher than that of a wild-type mouse.

In one aspect, a genetically modified mouse is provided that expresses hlL6 from an endogenous mouse lL-6 locus, wherein the endogenous mouse lL-6 gene has been ed with a hlL-6 gene, wherein the mouse does not exhibit a feature selected from a life span shortened by about 50% or more, kidney failure, hypergammaglobulinemia, elevated megakaryocytes in spleen, elevated megakaryocytes in bone marrow, plasmacytosis of , plasmacytosis of thymus, plasmacytosis of lymph nodes, glomerulonephritis, glomerulosclerosis, and a combination thereof.

In one embodiment, the mice have a life span that exceeds 20 weeks. In one embodiment, the mice have a life span that exceeds 30 weeks, 40 weeks, or 50 weeks. In one embodiment, the mice exhibit a life span about equal to that of a wild- type mouse of the same strain.

In one ment, the mice exhibit a level of megakaryocytes in spleen that is no more than about the splenic megakaryocyte level of a wild-type mouse In one embodiment, the mice comprise lymphoid organs that are ially devoid of abnormal and compactly arranged plasmacytoid cells.

In one embodiment, the mice exhibit gamma in serum levels equivalent to gamma globulin serum levels in wild-type mice. In one embodiment, the levels of (11- and B-globulin in serum of the mice are equivalent to (11- and B-globulin serum levels of wild-type mice of the same strain.

In one aspect, a genetically modified mouse is provided that expresses human lL-6 from an endogenous mouse lL-6 locus, wherein the endogenous mouse lL-6 gene has been replaced with a hlL-6 gene, wherein the mouse does not exhibit a feature selected from muscle wasting, an elevated cathepsin B level as compared with a wild-type mouse of the same strain, an elevated cathepsin A+B level as compared with a wild-type mouse of the same strain, an increased liver weight as compared with a wild-type mouse of the same strain, and a combination f.

In one embodiment, the weight of the liver of the mouse is about 800-900 mg at 12 weeks.

In one embodiment, the mouse exhibits a cathepsin B level throughout its life span that is no more than about the level observed in a wild-type mouse. In one embodiment, the mouse exhibits a cathepsin A+B level throughout its life span that is no more than about the level observed in a wild-type mouse.

In one ment, the mouse as an adult exhibits a gastrocnemus muscle weight that is within about 10% of the weight of a wild-type mouse of the same strain.

In one embodiment, the mouse as an adult ts a gastrocnemus muscle weight that is about the same as that of a wild-type mouse.

In one aspect a mouse is provided that ses a nucleotide sequence encoding a human lL-6 protein, wherein the nucleotide sequence encoding the human lL-6 protein replaces in whole or in part an endogenous nucleotide sequence encoding and endogenous mouse lL-6 protein.

In one aspect, a mouse is provided that comprises a replacement at an endogenous mouse lL-6 receptor locus of mouse lL-6Ra ectodomain with an ectodomain sequence of a human lL-6Rq to form a chimeric human/mouse lL-6Ra gene.

In one ment, the chimeric lL-6Ra gene is under the control of a mouse promoter and/or mouse regulatory elements at the endogenous mouse lL-6Rq locus.

In one embodiment, about 35.4 kb of mouse lL-6Rq ectodomain-encoding sequence is replaced with about 45.5 kb of human lL-6R ectodomain-encoding In one embodiment, the human lL-6R ectodomain-encoding sequence encompasses the first (ATG) codon in exon 1 through exon 8.

In one embodiment, the mouse lL-6Rq ce that is replaced includes a uous sequence that encompasses exons 1 h 8. In a ic embodiment, exons 1 through 8 and a n of intron 8 is deleted.

In one , a genetically ed mouse is provided, comprising a replacement at an endogenous mouse lL-6 locus of a mouse gene encoding lL-6 with a human gene encoding human lL-6, wherein the human gene encoding human lL-6 is under control of endogenous mouse tory elements at the endogenous mouse lL-6 locus.

In one embodiment, the human gene encoding human lL-6 is a human lL-6 gene of BAC ID CTD-2369M23.

In one embodiment, the mouse expresses a mouse lL-6Rq. In one embodiment, the mouse expresses a human lL-6Rq. In one embodiment, the humanized lL-6Ror comprises a human main. In one embodiment, the humanized lL-6Rq comprises a mouse transmembrane domain and a mouse cytoplasmic domain. In one embodiment, the mouse expresses a humanized lL-6Rq that comprises a humanization of ectodomain but not transmembrane and/or cytosolic domain.

In one embodiment, the mouse does not exhibit a feature selected from plasmocytosis, glomerulosclerosis, glomerulonephritis, kidney e, hypergammaglobulinemia, ed ryocytes in spleen, elevated megakaryocytes in bone marrow, splenomegaly, lymph node enlargement, compacted abnormal plasma cells, and a combination thereof.

In one aspect, a genetically modified mouse is provided, comprising a humanization of an endogenous mouse lL-6Ror gene, wherein the zation comprises a replacement of mouse lL-6Rq ectodomain-encoding sequence with human lL-6Rq ectodomain-encoding sequence at the endogenous mouse lL-6Rq locus.

In one ment, a contiguous mouse sequence comprising mouse exons 1 h 8 is replaced with a contiguous genomic fragment of human lL-6Ror sequence encoding a human lL-6Rq ectodomain. In one embodiment, the contiguous genomic fragment of human lL-6Rq sequence encoding the ectodomain is from BAC CTD-2192J23.

In one embodiment, the mouse further comprises a humanized lL-6 gene. In one embodiment, the mouse comprises a replacement at an endogenous mouse lL-6 locus of a mouse lL-6 gene with a human lL-6 gene. In one embodiment, the humanized lL-6 gene is under control of endogenous mouse regulatory elements.

In one aspect, a method is provided for making a humanized mouse, sing replacing a mouse gene sequence encoding mouse lL-6 with a human gene encoding human lL-6.

In one embodiment, the replacement is at an nous mouse lL-6 locus, and the human gene encoding human lL-6 is operably linked to endogenous mouse tory sequences.

In one aspect, a method for making a humanized mouse is provided, comprising replacing mouse exons ng ectodomain sequences of mouse |L- 6Rq with a human genomic fragment encoding ectodomain sequences of human |L- 6Rq to form a humanized lL-6Rq gene.

In one embodiment, the replacement is at an endogenous mouse lL-6Rq locus, and the humanized lL-6Rq gene is ly linked to nous mouse tory sequences.

In one aspect, a genetically ed mouse is provided, comprising a humanized lL-6Rq gene comprising a ement of mouse ectodomain-encoding sequence with human ectodomain sequence, wherein the humanized lL-6Rq gene ses a mouse transmembrane sequence and a mouse cytoplasmic sequence; wherein the mouse further comprises a gene encoding a human lL-6, wherein the gene encoding a human lL-6 is under control of endogenous mouse lL-6 regulatory elements.

In one embodiment, the mouse is incapable of expressing a fully mouse lL- 6Rq and incapable of expressing a mouse lL-6.

In various s, the genetically modified mice described herein comprise the genetic modifications in their germline.

In one aspect, a tissue, cell, or membrane fragment from a mouse as described herein is provided.

In one embodiment, the tissue or cell is from a mouse that expresses a human lL-6 protein, but that does not express a mouse lL-6 protein. In one embodiment, the tissue or cell is from a mouse that expresses a humanized lL-6Rq n, but not a mouse lL-6Rq protein. In one embodiment, the humanized lL-6Rq protein comprises a human ectodomain and a mouse transmembrane domain and a mouse cytosolic domain. In one embodiment, the tissue or cell is from a mouse that expresses a human lL-6, a humanized lL-6Rq, and that does not express a mouse lL-6 and does not express an lL-6Rq that comprises a mouse ectodomain.

In one aspect, an ex vivo complex of a mouse cell bearing a humanized lL- 6Rq (human ectodomain and mouse transmembrane and mouse cytoplasmic domain) and a human lL-6 is provided.

In one , a mouse embryo comprising a genetic modification as described herein is provided.

In one aspect, a mouse host embryo is provided that comprises a donor cell that comprises a genetic modification as described herein.

In one , a pluripotent or totipotent man animal cell comprising a genetic modification as described herein is provided. In one embodiment, the cell is a murine cell. In one embodiment, the cell is an E8 cell.

In one aspect, a mouse egg is ed, wherein the mouse egg comprises an ectopic mouse chromosome, n the c mouse chromosome comprises a genetic cation as described herein.

In one aspect, the mouse, embryo, egg, or cell that is genetically modified to comprise a human lL-6 gene or human or humanized lL-6Ra gene is of a mouse that is of a C57BL strain selected from C57BL/A, C57BL/An, C57BL/GrFa, C57BL/KaLwN, C57BL/6, C57BL/6J, C57BL/6ByJ, C57BL/6NJ, C57BL/10, C57BL/10808n, C57BL/10Cr, and C57BL/Ola. In another embodiment, the mouse is a 129 strain selected from the group consisting of a strain that is 129P1, 129P2, 129P3, 129X1, 12981 (e.g., 8V, 12981/8vlm), 12982, 12984, 12985, 12989/8vaH, 12986 (129/8vaTac), 12987, 12988, 129T1, 129T2 (see, e.g., Festing et al. (1999) d nomenclature for strain 129 mice, Mammalian Genome :836, see also, Auerbach et al (2000) Establishment and Chimera Analysis of 129/8va- and C57BL/6-Derived Mouse Embryonic Stem Cell Lines). In a specific embodiment, the genetically modified mouse is a mix of an aforementioned 129 strain and an aforementioned C57BL/6 strain. In another specific embodiment, the mouse is a mix of aforementioned 129 strains, or a mix of aforementioned BL/6 strains. In a specific embodiment, the 129 strain of the mix is a 12986 (129/8vaTac) . In another embodiment, the mouse is a BALB strain, e.g., BALB/c strain. In yet another embodiment, the mouse is a mix of a BALB strain and another aforementioned strain. In one embodiment, the mouse is 8wiss or 8wiss Webster mouse.

Each of the aspects and embodiments described herein are capable of being used together, unless excluded either explicitly or y from the context of the embodiment or aspect.

BRIEF PTION OF THE GS provides an illustration, not to scale, of the human (top) and mouse (bottom) lL-6 genomic loci. Exons I, II, III, IV, and V (in both human and mouse) are indicated by closed boxes to the right in the figure. 8elected ve regulatory regions are indicated by open boxes to the left in the figure. shows acute phase response (mSAA level) in the presence or absence of turpentine in wild-type mice, humanized ectodomain IL-6R mice, and mice with humanized IL-6 and IL-6R genes. shows turpentine-dependent acute phase response (8AA) in wild-type mice the e or presence of anti-mouse IL-6R antibody (left); and turpentine- dependent acute phase response in humanized L-6R mice in the absence or present of anti-human IL-6R antibody (right). shows FACS analysis for splenic B cells of wild-type and humanized IL-6 mice; pan B cell marker. shows FACS analysis for splenic T cells of wild-type an humanized IL- 6 mice; T helper cells and cytotoxic T cells. shows FACS analysis for c cells of wild-type and humanized IL-6 mice; Ly6G/C(Gr1). shows FACS analysis for splenic cells of wild-type and humanized IL-6 mice; NK cells and granulocytes (Ly6Ghi+/CD11bhi+). shows FACS analysis for blood B cells of wild-type and humanized IL- 6 mice; pan B cell marker. shows FACS analysis for blood T cells of wild-type and humanized IL- 6 mice; T helper cells and cytotoxic T cells. shows FACS analysis for blood myeloid cells of wild-type and humanized IL-6 mice; Gr1+ cells. shows FACS analysis for blood myeloid cells of wild-type and humanized IL-6 mice; CD11b vs. Ly6G/C(Gr1). shows FACS analysis for blood d cells of wild-type and humanized IL-6 mice; DX5 vs CD11b cells. shows FACS is of bone marrow IgM/CD24/B220 for wild-type and zed IL-6 mice. Top: normal progression in bone marrow. Bottom: FACS analysis for wild-type, hlL-6 heterozygotes, and hlL-6 homozygotes (IgM staining). shows FACS analysis of bone marrow IgM/CD24/B220 for ype and humanized IL-6 mice. Top: normal progression in bone marrow. Bottom: FACS analysis for ype, hlL-6 heterozygotes, and hlL-6 homozygotes (CD24 staining). shows FACS analysis of bone marrow CD43 and B220 for wild-type and zed IL-6 mice. Top: normal progression in bone marrow. Bottom: FACS analysis for wild-type, hlL-6 heterozygotes, and hlL-6 homozygotes (CD43 staining).

DETAILED DESCRIPTION lL-6 and lL-6R The |L-6 or (IL-6R) was long terized as a receptor for a B cell atory factor (BSF-2, or B cell Stimulatory Factor 2; also, BCDF, or B Cell Differentiation Factor) responsible for inducing B cells to synthesize immunoglobulin (Yamasaki et al. (1988) Cloning and Expression of the Human Interleukin-6(BSF- 2/lFNB 2) Receptor, Science 241 :825-828). lL-6 was first described as interferon-[52 as the result of its discovery during a search for a virally-induced protein termed interfereon-B, by treating human fibroblasts with dsRNA poly(|)poly(C) to induce an anti-viral response (Weissenbach et al. (1980) Two interferon mRNAs in human fibroblasts: In vitro translation and Escherichia coli cloning studies, Proc. Natl Acad.

Sci. USA 77(12):?152-7156; Keller et al. (1996) Molecular and Cellular Biology of Interleukin-6 and Its Receptor, Frontiers in ence 1:d340-357).

The human cDNA encodes a 468 amino acid protein having a 19-mer signal sequence and a cytoplasmic domain of about 82 amino acids that lacks a tyrosine kinase domain (see, Id.). The N-terminal (ectodomain) of the protein has an lg superfamily domain of about 90 amino acids, a 250-amino acid domain between the Ig superfamily domain and the membrane, a transmembrane span of about 28 amino acids (see, Id.). The ectodomain of the receptor binds its ligand lL-6, which triggers association with gp130 in the membrane and it is this complex that conducts signal transduction; the cytoplasmic domain reportedly does not transduce signal (Taga et al. (1989) Interleukin-6 Triggers the Association of Its Receptor with a Possible Signal Transducer, gp130, Cell 58:573-581). Indeed, a soluble form of lL-6R lacking a cytoplasmic domain can associate with |L-6 and bind gp130 on the e of a cell and effectively transduce signal (Id.).

The homology of hlL-6R and mlL-6R at the protein level is only about 54%; the transmembrane domain has a homology of about 79%, whereas the cytoplasmic domain has a homology of about 54% (Sugito et al. (1990)).

The natural ligand for the |L-6R, |L-6, was first isolated from es of HTLV- 1-transformed T cells (see, Hirano et al. (1985) cation to homogeneity and characterization of human B cell differentiation factor (BCDF or BSFp-2), Proc. Natl.

Acad. Sci. USA 82:5490-5494). A human cDNA for the |L-6 gene was cloned at least twice, once as BSF-2 (see, Hirano et al. (1086) mentary DNA fro a novel human interleukin (BSF-2) that induces B lymphocytes to produce immunoglobulin, Nature 324:73-76) and once as lFNB 2 (see, stein et al. (1986) ure and expression of cDNA and genes for human eron-beta-2, a distinct species inducible by growth-stimulatory cytokines, EMBO 5:2529-2537), although it has since been demonstrated that recombinant human lL-6 ts no detectable IFN activity.

Human lL-6 is a 184-amino acid protein that exhibits only about 42% homology with mouse lL-6, although the genomic organization of the human and mouse genes are basically the same, and the promoter s of the human and mouse genes share a 400-bp stretch that is highly conserved (see, Tanabe et al. (1988) Genomic Structure of the Murine lL-6 Gene: High Degree Conservation of Potential tory Sequences between Mouse and Human, J. lmmunol. 141(11):3875-3881).

The human lL-6 gene is about 5 kb (Yasukawa et al. (1987) Structure and expression of human B cell stimulatory factor-2 (BSC-2/lL-6) gene, EMBO J. 2939-2945), whereas the mouse lL-6 gene is about 7 kb (Tanabe et al. (1988) Genomic Structure of the Murine lL-6 Gene: High Degree Conservation of Potential Regulatory Sequences between Mouse and Human, J. lmmunol. 141(11):3875- 3881). The mouse and human lL-6 genes reportedly share highly conserved 5'- flanking sequence important to regulation. A schematic diagram of the human and mouse lL-6 genomic loci is shown in (not to scale). Exons I, II, III, IV, and V (in both human and mouse) are indicated by closed boxes to the right in the figure.

Selected putative regulatory regions are indicated by open boxes to the left in the figure. The putative regulatory s for humans are, from left to right, a glucocorticoid t from -557 to -552; an IFN enhancer core sequence from -472 to -468; a glucocorticoid element from -466 to -461; an AT-rich region from -395 to - 334, a consensus AP-1 binding site from -383 to -277; an IFN enhancer core ce from -253 to -248; a containing motif from -205 to -192; a c-fos SRE homology sequence from -169 to -82 containing an IFN enhancer core sequence, a cAMP-response t, a GGAAA motif, a CCAAT box, and a GC-rich ; and AP-1 binding site from -61 to -55; and a CCAAT box from -34 to -30. The putative regulatory regions for mouse are, from left to right, a GC rich region from - 553 to -536, a orticoid t from -521 to -516 and from -500 to -495; a Z- DNA stretch from -447 to -396; an AP-1 binding site overlapping an IFN enhancer core sequence from -277 to -288, a GGAAA motif overlapping an IFN enhancer core sequence from -210 to -195; a c-fos SRE homology region from -171 to -82 containing a cAMP response element, a GGAAA motif overlapping an IFN enhancer core sequence, and a GC-rich region; and, an AP-1 binding site from -61 to -55.

Mouse codons l-V have lengths 19, 185, 114, 150, and 165, respectively. Mouse intron lengths are: l-ll, 162 bp; ll-lll, 1253 bp; lll-IV, 2981 bp; |V-V, 1281 bp. Human codons l-V have lengths 19, 191, 114, 147, and 165. Human intron lengths are l-ll, 154; ll-lll, 1047; lll-lV, 706; lV-V, 1737. Genomic organization data are from Tanabe et al. (1988), and Yasukawa et al. (1987) Structure and expression of human B cell stimulatory factor-2 (BSF-2/lL-6) gene, EMBO J. 9(10):2939-2945.

It might be reasonable to presume that the mouse and human |L-6 genes appear to be similarly regulated based on the similarity of their 5'-flanking sequence.

A variety cell types exhibit enhanced |L-6 expression in response to |L-1, TNF, PDGF, IFNB, serum, poly(l)poly(C), and cycloheximide (see, Tanabe etal. (1988). lL-6 in humans mediates the acute phase response, hematopoiesis, B cell differentiation, T cell activation, growth and/or differentiation and/or activation of a variety of cell types (e.g., hepatocytes, lasts, elial cells, neurons, pituitary cells, lymphomas, myelomas, breast carcinomas, NK cells, macrophages, osteoclasts, etc.) (reviewed in, e.g., Heinrich et al. (1990), Kishimoto et al. (1989), and Keller et al. (1996); Sugita et al. (1990) Functional Murine Interleukin Receptor with lntracisternal A Particle Gene Product at its Cytoplasmic Domain, J. Exp. Med. 171:2001-2009).

In practice, however, mice transgenic for human lL-6 exhibit a panoply of ntial and debilitating ogies, reflecting a significant pleiotropy of the |L-6 gene. enic mice comprising a 6.6-kb fragment containing the human |L-6 gene and a p er (Eu) produce high concentrations of hlL-6 and extremely high lgG1 levels (120- to 400-fold over wild-type mice), reflecting an |L-6 deregulation that is accompanied by plasmacytosis, mesangio-proliferative glomerulonephritis, and high bone marrow megakaryocyte levels (Suematsu et al. (1989) lgG1 plasmacytosis in interleukin 6 transgenic mice, Proc. Natl Acad. Sci.

USA 86:7547-7551). Aberrant regulation of |L-6 and/or |L-6R is associated with myelomas, plastocytomas, toid tis, Castleman's disease, mesangial proliferative ulonephritis, cardiac myxoma, plams cell neoplasias, psoriasis, and other disorders (see, Kishimoto, T. (1989) The Biology of eukin-6, Blood 74(1):1-10; Sugita et al. (1990); also, Hirano et al. (1990) Biological and al aspects of interleukin 6, Immunology Today 11(12):443-449)). |L-6 is also implicated in sustaining levels of intra-prostatic androgens during androgen deprivation ent of prostate cancer patients by a paracrine and/or autocrine ism, potentially providing castration-resistant prostate tumor growth (Chun et al. (2009) eukin-6 Regulates Androgen Synthesis in Prostate Cancer Cells, Clin. Cancer Res. 15:4815-4822).

The human protein is encoded as a 212 amino acid protein, in mature form a 184 amino acid protein following cleavage of a 28 amino acid signal sequence. It contains two osylation and two osylation sites, and human |L-6 is phosphorylated in some cells. The mouse protein is encoded as a 211 amino acid protein, in mature form a 187 amino acid protein following cleavage of a 23 amino acid signal sequence. O-glycosylation sites are present, but not N-glycosylation sites. (See reviews on lL-6, e.g., Heinrich et al. (1990) Interleukin-6 and the acute phase response, Biochem. J. 1-636.) lL-6 on is pleiotropic. The lL-6 receptor is found on activated B cells but reportedly not on resting B cells. In contrast, lL-6R is found on resting T cells and can reportedly e T cell differentiation, activation, and proliferation, including the differentiation of T cells into cytotoxic T lymphocytes in the presence of |L-2.

Humanized lL-6/lL-6R Ectodomain Mice and lLMediated Acute Phase Response In humans, lL-6 induces the acute phase response. Early studies with human hepatocytes established that lL-6 induces acute phase proteins such as, e.g., C- reactive protein (CRP) and serum amyloid A (SAA) in a dose-dependent and time- dependent manner (reviewed in Heinrich et al. (1990) Interleukin-6 and the acute phase response, Biochem. J. 265:621-636). Non-human animals, e.g., mice or rats, sing humanized lL-6 and lL-6R genes are therefore useful systems for measuring the acute phase response mediated by human lL-6. Such animals are also useful for ining whether a nce induces an lLmediated acute phase response, by ng a humanized lL-6/lL-6R animal as described herein to the substance, and measuring a level of one or more acute phase response proteins (or RNAs). In one embodiment, the humanized animal is exposed to the substance in the presence of an antagonist of a human lL-6R, and a level of one or more acute phase response proteins (or RNAs) is measured, wherein a reduction in a level of an acute phase response protein (or RNA) in the presence of the human lL-6R antagonist indicates a human lL-6R-mediated acute phase response.

Human lL-6 can bind both human lL-6R and mouse lL-6R; mouse lL-6 binds mouse lL-6R but not human lL-6R (no binding of mlL-6 to hlL-6R detectable, whereas hlL-6 can compete with mlL-6 for binding mlL-6R; Coulie et al. (1989) High- and finity ors for murine interleukin 6. Distinct bution on B and T cells, Eur. J. l. 19:2107-211); see also, e.g., Peters etal. (1996) The Function of the Soluble Interleukin 6 (IL-6) Receptor In Vivo: Sensitization of Human Soluble IL-6 Receptor Transgenic Mice Towards IL-6 and Prolongation of the Plasma Half-life of IL-6, J. Exp. Med. 183:1399-1406). Thus, human cells that bear hlL-6R in a mouse (e.g., in a xenogenic transplant) cannot rely on endogenous mlL-6 to carry out ILmediated functions, including but not limited to the role of IL-6 blood cell or lymphocyte development (e.g., hematopoiesis, B cell activation, T cell activation, etc.) In a mixed in vivo system sing a wild-type mouse IL-6 gene and a human IL-6R gene (but no mouse IL-6R gene), an acute phase response r is not expected to induce detectable levels of acute phase proteins that would indicate an acute phase response. However, a humanized mouse as described herein, comprising a humanized IL-6 gene and an IL-6R gene comprising a humanized ectodomain sequence will respond to an acute phase response inducer and exhibit acute phase response proteins in serum. Mice wild-type for L-6R tested for acute phase proteins in the presence or absence of the acute phase inducer turpentine showed a turpentine-dependent increase in acute phase proteins. Mice with humanized IL-6 gene, but not IL-6R, showed no acute phase se in the presence of turpentine. But mice bearing both a human IL-6 gene and an IL-6R gene with a humanized ectodomain exhibited a strong acute phase response (. The ediated acute phase response was IL-6 dependent in both wild-type mice ( top) and in humanized IL-6/IL-6R ectodomain mice ( bottom), as evidenced by the ability of the appropriate L-6R antibody to te the acute phase response at a sufficiently high antibody dose. Thus, a double humanization of IL-6 and IL-6R recapitulates the wild-type ILmediated acute phase response with t to serum acute phase proteins. cally Modified Mice Genetically modified mice are provided that express a human IL-6 and/or a humanized IL-6 receptorfrom endogenous mouse loci, wherein the endogenous mouse IL-6 gene and/or the endogenous mouse IL-6 receptor gene have been replaced with a human IL-6 gene and/or a human sequence comprising a sequence that encodes an ectodomain of a human IL-6 receptor. The genetically modified mice express the human IL-6 and/or humanized IL-6 receptor from zed endogenous loci that are under control of mouse promoters and/or mouse regulatory elements. The replacement(s) at the endogenous mouse loci provide non-human animals that express human IL-6 and a zed IL-6 or in a manner that does not result in the panoply of ntial pathologies observed in IL-6 transgenic mice known in the art.

Transgenic mice that express human lL-6 are known in the art. However, they generally suffer from significant pathologies that ly limit their usefulness.

Humanized mice as described herein express a human lL-6 and/or humanized lL-6 receptor under the control of endogenous mouse regulatory elements at endogenous mouse lL-6 and lL-6Rq loci. These mice, in contrast, exhibit expression patterns with respect to these genes that are different from transgenic mice known in the art.

Replacement of man genes in a non-human animal with homologous or orthologous human genes or human sequences, at the endogenous non-human locus and under l of nous promoters and/or regulatory elements, can result in a non-human animal with qualities and characteristics that may be substantially different from a typical knockout-plus—transgene animal. In the typical knockout-plus-transgene animal, an endogenous locus is removed or d and a fully human transgene is inserted into the animal's genome and presumably integrates at random into the genome. Typically, the location of the ated transgene is unknown; expression of the human protein is measured by transcription of the human gene and/or protein assay and/or functional assay. Inclusion in the human transgene of upstream and/or downstream human sequences are apparently presumed to be sufficient to provide suitable support for expression and/or tion of the transgene no matter where in the 's genome the transgene winds up.

But in many cases the transgene with human regulatory elements expresses in a manner that is unphysiological or othenNise unsatisfactory, and can be actually detrimental to the animal. In contrast, the inventors demonstrate that a replacement with human sequence at an endogenous locus under control of endogenous regulatory elements provides a physiologically riate expression pattern and level that s in a useful humanized animal whose logy with respect to the replaced gene are meaningful and appropriate and context of the humanized animal's physiology. ized mouse eggs injected with a construct having the MHC class I promoter H2 and a B-globin intron driving expression of a 695-bp mouse lL-6 gene reportedly produce mice that constitutively express mouse lL-6 at relatively high levels (as compared with wild-type mice) (see, Woodrofe et al. (1992) Long-Term Consequences of Interleukin-6 Overexpression in enic Mice, DNA and Cell Biology 11(8):587-592). But these mice are prone to develop lymphomas ated with the intestines, lymph nodes, and kidney, as well as splenic amyloid deposits.

They also exhibit abnormal B cell maturation (see, Woodrofe et al., ld.), so that studies of B cell on are compromised. In contrast, mice as described herein that comprise a replacement of the mouse lL-6 gene with a human lL-6 gene at the mouse lL-6 locus are not prone to develop these lymphomas, and the mice exhibit apparently normal B cell populations.

Mice (C57BL/6) transgenic for hlL-6 due to a random insertion of a 6.6-kb (BamHl-Pvu ll fragment) length of human DNA containing the hlL-6 gene coupled with an lgM enhancer have been reported (see, Suematsu et al. (1989) lgG1 cytosis in interleukin 6 transgenic mice, Proc. Natl. Acad. Sci. USA 86:7547- 7551). The mice express hlL-6 at between 800 pg/mL and 20,000 pg/mL in serum, where wild-type mice typically express only about 100 pg/mL lL-6. The mice exhibit an increase in serum lg (120 to 400-fold over wild-type mice) and a decrease in albumin as they age. The mice suffer from a massive plasmacytosis, exhibit megaly and lymph node enlargement, as well as exhibiting plasma cells and sed megakaryocytes in bone marrow. Upon inspection, what appear to be enlarged lymph nodes are instead massed of compacted al plasma cells.

Both spleen and thymus exhibit massive proliferation of plasma cells, which also infiltrate portions of the lung, liver, and kidney. Kidney in these mice also exhibits |L- ulated mesangial cell proliferation typical of mesangio-proliferative glomerulonephritis. Similarly, mice (BALB/c) transgenic for a trimmed hlL-6 cDNA driven by a mouse H-2Ld promoter randomly inserted into the genome display severe cytosis (see, Suematsu et al. (1992) Generation of cytomas with the chromosomal translocation t(12;15) in interleukin 6 transgenic mice, Proc. Natl.

Acad. Sci. USA -235). Although C57BL/6 mice that overexpress hlL-6 do not develop transplantable plasmacytomas (they do exhibit plasmacytosis), transgenic BL/6 mice back-crossed into BALB/c mice reportedly do.

Random transgenesis of a hlL-6 cDNA driven by a glial fibrillary acidic protein (GFAP) gene er reportedly results in hlL-6 overexpression in the mouse central nervous , which also leads to icant pathologies (see, Campbell et al. (1993) Neurologic disease induced in transgenic mice by al overexpression of interleukin 6, Proc. Natl. Acad. Sci. USA 90:10061-10065). These mice exhibit ive neuropathology and reactive astrocytosis resulting from lL-6 expression in the CNS due to loss of control as the result of random integration of an lL-6 transgene at an apparently CNS-permissive transcriptional locus. Although expression of hlL-6 cDNA linked to a B-globin 3’-UTR and driven by a neuron-specific e promoter microinjected into fertilized mouse eggs (F1 C57BL/6 x BALB/c) produced mice with a normal lifespan and without apparent neurological defects that expressed hlL-6 in neurons but not elsewhere (see, Fattor et al. (1994) lL-6 Expression in Neurons of Transgenic Mice Causes Reactive Astrocytosis and Increase in Ramified Microglial Cells But No al Damage, Eur. J. Neuroscience 7:2441-2449), the mice exhibited high levels (20- to 30-fold higher than wild-type) of activated and enlarged astrocytes with increased processes throughout the brain, as well as a 10- to 15-fold increase in ramified microglial cells in white matter. Thus, brain expression of IL-6 reportedly leads to conditions that range from ve astrocytosis to frank and profound neuropathology.

Microinjection into fertilized eggs of an F1 cross of C57BL/6x”DBAII” mice of a 639-bp hlL-6 cDNA linked to a B-globin 3’-UTR and a mouse MT-1 promoter reportedly ed a transgenic mouse in which the hlL-6 gene was ly integrated produced a weakened and diseased mouse that dies young of kidney failure (see Fattori et al. (1994) Blood, pment of ssive Kidney Damage and Myeloma Kidney in Interleukin-6 Transgenic Mice, Blood 63(9):2570-2579).

Transgenic mice expired at 12-20 weeks and exhibited elevated levels of a1 and [5- globulins in plasma, hypergammaglobulinemia, elevated megakaryocytes in spleen (3-fold higher than wild-type) and bone , plasmacytosis of lymphoid organs (spleen, thymus, and lymph nodes) characterized by abnormal and compactly arranged plasmocytoid cells, and glomerulonephritis leading to glomerulosclerosis similar to multiple myeloma.

Microinjection into fertilized eggs of a C57BL/6J mouse of a H-2Ld-driven hlL- 6 cDNA caused ILdependent muscle wasting in mice, characterized in part by a icantly lower gastrocnemius muscle weight in enic mice as compared to weight-matched controls, a difference that was ameliorated by treatment with an IL-6 antagonist (see, Tsujinaka et al. (1996) Interleukin 6 or Antibody Inhibits Muscle Atrophy and Modulates Proteolytic Systems in Interleukin 6 Transgenic Mice, J. Clin. Invest. 244-249). At 12 weeks these mice displayed serum hlL-6 levels of more than 600,000 pg/mL. The transgenic mice also had livers that weighed about 1,242 mg, as compared to control livers that were about 862 mg. Transgenic mice treated with IL-6 antagonist had livers that weighed about 888 mg. Muscle cathepsins B and B+L were significantly higher (20-fold and 62-fold) in transgenic mice than in controls, a phenomenon that was eliminated in transgenic mice treated with an IL-6 antagonist. cathepsin B and L mRNAs were estimated to be about 277% and 257%, respectively, as compared with ype mice; the difference was icantly d with IL-6 nist treatment.

Mice comprising a hlL-6 minigene driven by a mouse MHC class I H- 2Ld promoter and a hlL-6R minigene driven by a chicken B-actin promoter, and a gp130 gene, ted pathologies typical of hlL-6 transgenic mice (e.g., hepergammaglobulinemia, splenomegaly, mesangial proliferative glomerulonephritis, lung lymphoid infiltration) as well as ventricular hypertrophy (Hirota et al. (1995) Continuous activation of gp130, a signal-transducing or component for interleukin 6—related nes, causes myocardial hypertrophy in mice, Proc. Natl Acad. Sci. USA 92:4862—4866). The ventricular hypertrophy is believed to be mediated by a continuous activation of gp130 (Id.). The role of lL-6 is reportedly to help strengthen the ne or complex and induce dimerization of gp130, which is the signal transducing component responsible for transducing the lL-6 signal (Paonessa et al. (1995) Two distinct and ndent sites on lL-6 trigger gp130 dimer formation and signalling, EMBO J. 14(9):1942—1951). The activated complex is believed to be a r composed of two lL-6, each lL-6 bound to one lL-6Ra and two gp130 (each lL-6 contains two independent gp130-binding sites) exhibiting a 2:2:2 stoichiometry, wherein the dimerization of gp130 causes activation of JAK—Tyk ne kinases, phosphorylation of gp130 and STAT family transcription s and other intracellular substrates (Id.; Stahl, N. (1994) Association and Activation of Jak- Tyk Kinases by CNTF-LlF-OSM-lL-6 [5 Receptor Components, Science 263:92—95), tent with a general model of cytokine receptor complex formation (see, Stahl, N. and oulos, G. (1993) The Alphas, Betas, and Kinases of Cytokine Receptor Complexes, Cell 74:587-590; Davis et al. (1993) LIFRB and gp130 as Heterodimerizing Signal ucers of the Tripartite CNTF Receptor, Science 260:1805-1808; Murakami et al. (1993) lLlnduced Homodimerization ofgp130 and Associated tion of a Tyrosine Kinase, Science 2601808—1810).

Mice transgenic for human le-6R driven by a rat PEP carboxykinase promoter and human lL-6 driven by a mouse metallothionein-1 promoter are reportedly markedly smaller that mice transgenic for human lL-6 alone or human le- 6R alone (Peters et al. (1997) Extramedullary Expansion of Hematopoietic Progenitor Cells in |nterleukin(|L-)—6—le-6R Double Transgenic Mice, J. Exp. Med. 185(4):?55- 766), reflected in d body fat and d weight (20-25 g vs. 40 g). Double transgenic mice reportedly also exhibit spleen (5-fold) and liver (2—fold) enlargement as compared with reportedly normal organ weights for single transgenic mice, apparently due to extramedullary proliferation of hematopoeitic cells of spleena and liver but not bone marrow, as well as elevated megakaryocytes in spleen and plasmacellular infiltrates in all parenchymal organs (Id.). Double transgenics also exhibit livers with an increase of about 200- to about 300-fold in granulocytes, macrophages, progenitor cells, and B cells as compared with single transgenics; in contrast, lL-6 single enic mice exhibited lesser increases in macrophages (15- fold) and B cells (45-fold) (Id.). The extraordinary findings are presumably due to stimulation of growth and differentiation of hematopoietic progenitor cells by ting gp130 signal transduction (Id.).

Further, double transgenic (mouse metallothionine promoter-driven rat PEP ykinase promoter-driven hlL-6R) mice exhibit a hepatocellular hyperplasia that is reportedly identical to human nodular regenerative lasia with sustained hepatocyte proliferation that strongly suggests that |L-6 is responsible for both hepatocyte proliferation and pathogenic hepatocellular transformation (Maione et al. (1998) Coexrpession of |L-6 and soluble lL-6R causes nodular regenerative hyperplasia and adenomas of the liver, EMBO J. 17(19):5588—5597).

Because hepatocellular hyperplasia is reportedly not observed in single transgenic hlL-6 mice and hlL-6 can bind mlL-6R, the finding may appear paradoxical until it is considered that the double transgenic may result in higher levels of hlL-6 complexed to soluble |L-6R (here, soluble hlL-6R), which complex is a more potent inhibitor that |L-6 alone (Id.).

] In contrast to mice that are transgenic for human |L-6, humanized |L-6 mice that comprise a replacement at an endogenous mouse |L-6 locus, which retain mouse regulatory elements but comprise a humanization of |Lencoding sequence, do not exhibit the severe ogies of prior art mice. Genetically modified mice that were heterozygous or gous for hlL-6 were grossly normal.

Mice with a humanized lL-6 gene (MAID 760) as described in the Examples were immunophenotyped and found to have normal B cell numbers in FACS analyses (lymphocyte-gated) of spleen B cells using a pan B cell marker R(B220)) (. For spleen, wild-type mice ted 63% B cells; hlL-6 heterozygote mice exhibited 63% B cells; and mice homozygous for hlL-6 at the endogeous mouse locus ted 63% B cells. B cell numbers for homozygous hlL- 6 mice immunized with TNP-KLH were also normal (65% for wild-type, and 61% for hlL-6 homozygotes).

Splenic T cells were also about the same as wild-type (. tages of splenic T cells for Thelper/Tcytoxic were, for wild-type 20%/40% (ratio of 1.4:1); for hlL-6 heterozygotes 23%/14% (ratio of 16:1); for hlL-6 homozygotes 21%/15% (ratio of 1.4:1) (markers were CD8a-APC; CD4-FITC). gous hlL-6 mice immunized with TNP-KLH exhibited similar splenic T cell numbers to wild-type mice, i.e., Thelper/Tcytotoxic were 22%/20% (ratio of 1.1:1) as compared with 21 %/19% for wild-type (also a ratio of 1.1 :1 ).

Humanized |L-6 mice also ted about normal levels of splenic NK cells on FACS analysis (CD11b and DX5) (. hlL-6 heterozygotes exhibited 2.2% NK cells, and hlL-6 homozygotes exhibited 1.8% NK cells, whereas wild-type mice exhibited 2.4% NK cells. Following immunization with TN P-KLH, homozygotes exhibited 1.6% splenic NK cells, s wild-type mice exhibited 2.1% splenic NK cells. zed |L-6 mice also exhibited normal levels of splenic (Gr1) cells (. hlL-6 heterozygotes exhibited 7.0% GR1+ cells (1.3% Gr1hi); homozygotes exhibited 6.8% Gr1+ cells (0.9% Gr1hi), whereas wild-type mice ted 8.0% Gr1+ cells (1 .8%Gr1hi). zed |L-6 homozygotes (immunized with TNP-KLH) exhibited 11% Gr1+ cells (4.0% Gr1hi), whereas wild-type mice exhibited 10% Gr1+ cells (30% GM“).

Humanized |L-6 mice also exhibited normal blood B and T cell numbers in FACS analysis (and . FACs with a pan B cell marker R(B220)) revealed that homozygous hlL-6 mice ed 52% B cell as compared with wild-type 53%; zygotes exhibited 38% (an average of two ent stainings of 29% and 47%). Homozygous hlL-6 mice immunized with TNP- KLH gave similar B cell numbers (43%, as compared with 45% for wild-type mice).

Humanized |L-6 mice exhibited normal blood T cell numbers in FACS analysis as measured by CD8a and CD4 staining. Heterozygous hlL-6 mice exhibited Thelper/Tcytotoxic numbers of 39%/26% (ratio of 1.5:1); homozygous hlL- 6 mice exhibited Th/Tc numbers of 24%/20% (ratio of 1.2:1), whereas wild-type mice exhibited Th/Tc numbers of % (ratio of 1.3:1). Homozygous hlL-6 mice immunized with TNP-KLH had Th/Tc numbers of 29%/21% (ratio of 1.4:1), whereas wild-type immunized mice had Th/Tc numbers of 28%/23% (1 .2:1).

Humanized lL-6 mice also exhibited myeloid cell numbers in blood that were similar to wild-type mice as measured by FACS analysis of naive and immunized mouse blood stained with Ly6G/C(Gr1) and CD11b, as well as CD11b and DX5 (, , and FIG 12). Heterozygous hlL-6 mice exhibited %Gr+ cells of 10.8%, homozygotes 6.9%, whereas wild-type mice exhibited 9.7%.

Immunized hlL-6 homozygotes exhibited M1(Ly6G/C(Gr) of 101-104)/ G/C(Gr) staining of about 102-103) numbers of 43%/34%, whereas ype mice had numbers of 45%/38%. FACS plots of CD11b (vertical axis) vs. Ly6G/C (horizontal axis) for immunized homozygous hlL-6 mice showed cell percentage in quadrants (upper left/upper right/lower right) of 16%/8%/3%, which were identical to immunized ype quadrant numbers.

Homozygous TNP-KLH-immunized humanized IL-6 mice exhibited CD11b vs. DX5(NK) staining FACS plots that were similar to immunized wild-type mice. Quadrant analysis blood FACS plots (CD11b vertical axis, DX5(NK) horizontal axis) revealed upper left/upper lower right numbers of 9.5%/17%/10% for hlL-6 homozygotes and 6.5%/17.3%/14% for wild-type mice.

Humanized IL-6 mice exhibited an isotype response that was essentially the same as observed in wild-type mice. Early and final lgG1, IgG2a, IgG2b, IgG3, lgA, IgE, and IgM levels were about the same as observed in wild-type mice. In one experiment, final IgM was slightly higher in humanized mice; final lgG3 was also elevated in humanized mice.

B cell development in naive hlL-6 mice was essentially indistinguishable from development in wild-type mice based on FACS analysis of bone marrow lgM/CD24/B220 staining (). lmmunophenotyping of immune mice ed that marker populations for various cell types in the B cell development ssion were ially normal in hlL-6 mice. Progression of cells from hematopoietic stem cells, common lymphoid itors, ProB cells, PreB cells, and immature and mature B cells is normal in hlL-6 mice ( and ) EXAMPLES Example 1: Replacement of Endogenous Mouse lL-6 Gene with hlL-6 Gene The 4.8-kb human IL-6 gene containing exons 1 through 4 of the human IL-6 gene ed 6.8 kb of the murine IL-6 gene locus.

A targeting construct for replacing the mouse with the human IL-6 gene in a single targeting step was constructed using VELOCIGENE® genetic engineering logy (see, Valenzuela et al. (2003) High-throughput engineering of the mouse genome d with high-resolution expression analysis, Nature Biotech, 21(6):652-659). Mouse and human IL-6 DNA were obtained from bacterial artificial some (BAC) RPCI-23 clone 368C3, and from BAC CTD clone 2369M23, respectively. Briefly, a Notl linearized targeting construct generated by gap repair cloning containing mouse IL-6 upstream and downstream homology arms flanking a 4.8 kb human IL-6 ce extending from ATG in exon 1 to exon 5 with 16 nucleotides of 3’ downstream sequence (genomic nates: NCBIh37.1: ,766,882 to 22,771,637) and a floxed neo selection cassette, was electroporated into F1 H4 mouse embryonic stem (ES) cells (C57BL/6 x 129 F1 hybrid). Correctly targeted ES cells (MAID 790) were further electroporated with a transient Cre-expressing vector to remove the drug ion cassette. Targeted ES 2012/062379 cell clones without drug cassette (MAID 1428) were introduced into an 8-cell stage mouse embryo by the VELOCIMOUSE® method (see, US Pat. No. 7,294,754, 7,576,259, 7,659,442, and Poueymirou et al. (2007) F0 generation mice that are essentially fully derived from the donor gene-targeted ES cells allowing immediate phenotypic analyses Nature Biotech. 25(1):91-99). VELOCIMICE® (F0 mice fully derived from the donor ES cell) bearing the humanized IL-6 gene were identified by ping for loss of mouse allele and gain of human allele using a cation of allele assay (see, Valenzuela et al. ).

] Correctly targeted ES cell clones were identified by a Ioss-of—native- allele (LONA) assay (Valenzuela et al. 2003) in which the number of copies of the native, unmodified [[6 gene were determined by two TaqManTM quantitative polymerase chain reactions (qPCRs) specific for sequences in the mouse ”6 gene that were targeted for deletion. The qPCR assays comprised the following primer- probe sets (written 5’ to 3’): upstream fonNard primer, TTGCCGGTTT TCCCTTTTCT C (SEQ ID NO:1); upstream reverse primer, GGCC GTGGTTGTC (SEQ ID NO:2); upstream probe, FAM-CCAGCATCAG TCCCAAGAAG GCAACT-BHQ (SEQ ID NO:3); ream forward primer, TCAGAGTGTG GGCGAACAAA G (SEQ ID NO:4); downstream reverse primer, GTGGCAAAAG CAGCCTTAGC (SEQ ID NO:5); ream probe, FAM-TCATTCCAGG CCCTTCTTAT TGCATCTG-BHQ (SEQ ID NO:6); in which FAM refers to the 5-carboxyfluorescein fluorescent probe and BHQ refers to the fluorescence quencher of the black hole quencher type (Biosearch Technologies). DNA purified from ES cell clones that that have taken up the targeting vector and incorporated in their genomes was combined with TaqManTM Gene Expression Master Mix (Life Technologies) according to the manufacturer’s suggestions in a ll PCR plate (MicroAmpTM Optical 384-Well Reaction Plate, Life Technologies) and cycled in an d Biosystems Prism 7900HT, which collects fluorescence data during the course of the PCRs and determines a threshold cycle (Ct), the fractional PCR cycle at which the accumulated fluorescence reaches a pre-set threshold. The upstream and downstream l/6-specific qPCRs and two qPCRs for non-targeted nce genes were run for each DNA . The differences in the Ct values (ACt) between each l/6-specific qPCR and each reference gene qPCR were calculated, and then the difference between each ACt and the median ACt for all samples assayed was calculated to obtain AACt values for each sample. The copy number of the ”6 gene in each sample was calculated from the following formula: copy number = 2 . 2-AACt. A correctly targeted clone, having lost one of its native , will has an ”6 gene copy number equal to one. Confirmation that the human IL6 gene ce replaced the deleted mouse II6 gene sequence in the humanized allele was confirmed by a TaqManTM qPCR assay that comprises the following primer-probe sets (written 5’ to 3’): the human fonNard primer, TCCACTGGAATTTG (SEQ ID NO:7); the human reverse primer, GTTCAACCACAGCCAGGAAAG (SEQ ID NO:8); and the human probe, FAM- AGCTACAACTCATTGGCATCCTGGCAA-BHQ (SEQ ID NO:9).

] The same LONA assay was used to assay DNA purified from tail biopsies for mice derived from the targeted ES cells to determine their II6 genotypes and m that the humanized II6 allele had transmitted through the ne. Two pups heterozygous for the replacement are bred to generate a mouse that is homozygous for the replacement of the endogenous mouse lL-6 gene by the human IL-6 gene. Pups that are homozygous for the replacement are used for phenotyping.

The am junction of the murine locus and the sequence containing the hlL-6 gene is designed to be within 5’-AATTAGAGAG CCTA ATAAATATGA GACTGGGGAT GTCTGTAGCT CATTCTGCTC TGGAGCCCAC CAAGAACGAT AGTCAATTCC AGAAACCGCT ATGAACTCCT CAAG TAAGTGCAGG AAATCCTTAG CCCTGGAACT GCCAGCGGCG GTCGAGCCCT GTGTGAGGGA GGGGTGTGTG GCCCAGG (SEQ ID NO:10), wherein the final mouse nucleotide priorto the first nucleotide of the human gene is the “T” in CCGCT, and the first nucleotide of the human sequence is the first “A” in ATGAA. The downstream junction of the sequence containing the hlL-6 gene and the murine locus is designed to be within 5’-TTTTAAAGAA ATATTTATAT TGTATTTATA TAATGTATAA ATGGTTTTTA TACCAATAAA TGGCATTTTA AAAAATTCAG CAACTTTGAG ACGC TCCCGGGCTC GATAACTATA ACGGTCCTAA GGTAGCGACT CGAGATAACT T-3’ (SEQ ID NO:11), wherein the final nucleotide of the human sequence is with the final “G” in TCACG and the first nucleotide of the mouse sequence is the first “C” in CTCCC; the downstream junction region also contained a onP site at the 3’ end (the beginning of which is shown) for removal of a floxed ubiquitin promoter-driven neo cassette. The junction of the neo cassette with the mouse lL-6 locus is designed to be within 5’-TATACGAAGT TATCCTAGGT TGGAGCTCCT AAGTTACATC CAAACATCCT CCCCCAAATC TAAG CACTTTTTAT GACATGTAAA GTTAAATAAG AAGTGAAAGC TGCAGATGGT GAGTGAGA (SEQ ID , where the final “C” of AGCTC is the final nucleotide of the neo te; the first nucleotide of the mouse genome ing the cassette is the initial “C” of CTAAG.

Example 2: phenotyping of Naive and lmmunized hlL-6 Mice: B Cells Mice homozygous for the h|L-6 gene replacement were analyzed for B cells (DC445R(B220). Lymphocyte-gated fractions from spleen cell preparations of naive and immunized LH) h|L-6 mice were stained and immunophenotyped using flow cytometry. FACS analysis showed that the percentage of B cells of the spleen cell preparation as measured by CD45R(8220)-FITC staining were about the same (63% of cells) for preparations from naive ype, hlL-6 heterozygotes, and hlL-6 homozygotes. For immunized mice, B cells accounted for about 65% of total cells of the spleen cell preparation in wild-type mice, and about 61% of total cells in h|L-6 gotes. Spleens of h|L-6 mice (both naive and immunized) contain a population of B cells that is about the same size as the splenic B cell population in wild-type mice.

Bone marrow of wild-type, h|L-6 heterozygotes, and h|L-6 homozygotes was stained with B cell markers (CD45R(B220)-APC, CD24(HSA)-PE, or CD43 conjugated to a dye and/or lgM (lgM-FITC). B cell development in bone marrow of normal mice will be ted in e markers as cells progress from stem cells to early pro-B cells to late pro-B cells, to large pre-B cells to small pre-B cells to re B cells and finally, to mature B cells. Common lymphocyte progenitor pro-B cells will express CD45R, and in later stages will s lgM as immature and later as mature B cells. Thus, CD45R-stained and anti-lgM-stained B cells should reveal a pattern characteristic of B cell development. Bone marrow of h|L-6 heterozygotes and homozygotes displayed a pattern of CD45R(8220)-APC and anit-lgM-FITC staining that was ially indistinguishable from wild-type bone marrow, showing populations of B cells that stained positive for CD45R(B220) and lgM, or CD45R(B220) alone. B cell sub-populations in bone marrow of h|L-6 mice revealed by FACS staining were similar to those in wild-type mice (Table 1; see also, ).

Table 1. B Cells in Bone Marrow of Naive Mice hlL-6 Mouse Wild-type Mouse (%) Heterozygote Homozygote (%) (%) CLP-ProB PreB-lmmatureB ] ng for CD24 (see ) revealed the (normal) pattern shown in Table 2, indicating normal development in bone marrow.

Table 2. B Cells in Bone Marrow of Naive Mice hlL-6 Mouse Wild-type Mouse_ (0/0) Heterozygote Homozygote (%) (%) Developing HSC- 46.6 46 43 Mature CLP/early -2 9-0 10-1 ProB Late ProB, PreB, 7-2 11.6 10-7 Immature B Mature B 14.1 14.9 17 ng for CD43 (see ) revealed the (normal) pattern shown in Table 3, indicating normal development in bone marrow.

Table 3. B Cells in Bone Marrow of Naive Mice hlL-6 Mouse Wild-type Mouse (°/o) Heterozygote Homozygote (%) (%) PreBll-lmmature 28.4 21.4 21.2 B cells Mature B cells 8.1 Thus, immunophenotyping of nai've hlL-6 mice revealed that B cell development in such mice is essentially .

Example 3: Replacement of nous Mouse lL-6Ra Ectodomain Gene Sequence with hlL-6Rq Ectodomain Gene Sequence The 45 kb human lL-6Ra gene containing exons 1 through 8 of the human lL-6Ra gene replaced 35.4 kb of the murine lL-6Rq gene locus. Mouse exons 9 and 10 were retained; only exons 1-8 were humanized. In total, 35,384 nt of mouse sequence was replaced by 45,047 nt of human sequence.

A targeting construct for replacing the mouse with the human lL-6Rd gene in a single targeting step was constructed using VELOCIGENE® c engineering technology (see, Valenzuela et al. (2003) High-throughput engineering of the mouse genome coupled with high-resolution expression analysis, Nature Biotech, 21(6):652-659). Mouse and human lL-6Rq DNA were ed from bacterial artificial chromosome (BAC) RPCl-23 clone 125J8, and from BAC CTD clone 2192J23, respectively. Briefly, a Notl linearized targeting construct generated by gap repair cloning containing mouse lL-6Rq upstream and downstream homology arms flanking a 45 kb human lL-6Rq sequence extending from ATG in exon 1 to exon 8 with 69 nucleotides of 3’ downstream sequence and a floxed neo selection cassette, was electroporated into F1H4 mouse embryonic stem (ES) cells (C57BL/6 x 129 F1 hybrid). Correctly targeted ES cells (MAID 794) were further electroporated with a transient Cre-expressing vector to remove the drug selection te.

Targeted ES cell clones t drug cassette (MAID 1442) were introduced into an 8-cell stage mouse embryo by the VELOCIMOUSE® method (see, US Pat. No. 7,294,754, 259, 7,659,442, and Poueymirou et al. (2007) F0 generation mice that are essentially fully derived from the donor gene-targeted ES cells allowing immediate phenotypic analyses Nature h. 25(1):91-99). VELOCIMICE® (F0 mice fully derived from the donor ES cell) bearing the humanized lL-6Rq gene were identified by genotyping for loss of mouse allele and gain of human allele using a modification of allele assay (see, Valenzuela et al. (2003)).

Correctly targeted ES cell clones were identified by a Ioss-of—native- allele (LONA) assay (Valenzuela et al. 2003) in which the number of copies of the native, unmodified ”6 gene were determined by two TaqManTM quantitative polymerase chain reactions (qPCRs) specific for sequences in the mouse ”6 gene that were targeted for deletion. The qPCR assays comprised the following - probe sets (written 5’ to 3’): upstream forward , GCCCTAGCAT GCAGAATGC (SEQ ID NO:13); upstream reverse primer, TCCC ACATCCTTTG C (SEQ ID NO:14); upstream probe, TCCA TCCT GTGAG (SEQ ID NO:15); downstream forward primer, GAGCTTGCCC CCAGAAAGG (SEQ ID NO:16); downstream reverse primer, CGGCCACATC AGAC (SEQ ID NO:17); downstream probe, CATGCACTGC CCCAAGTCTG GTTTCAGT (SEQ ID NO:18). DNA ed from ES cell clones that that have taken up the targeting vector and incorporated in their genomes was combined with TaqManTM Gene Expression Master Mix (Life Technologies) according to the cturer’s suggestions in a 384-well PCR plate (MicroAmpTM Optical 384-Well Reaction Plate, Life Technologies) and cycled in an Applied Biosystems Prism 7900HT, which collects fluorescence data during the course of the PCRs and ines a old cycle (Ct), the fractional PCR cycle at which the accumulated fluorescence reaches a pre- set threshold. The upstream and downstream lL-6Rq-specific qPCRs and two qPCRs for non-targeted reference genes were run for each DNA sample. The differences in the Ct values (ACt) between each IL-6Rq-specific qPCR and each reference gene qPCR were ated, and then the difference between each ACt and the median ACt for all samples assayed was calculated to obtain AACt values for each sample. The copy number of the ”6 gene in each sample was ated from the following formula: copy number = 2 - 2-AACt. A correctly ed clone, having lost one of its native copies, will have an IL-6Rq gene copy number equal to one.

Confirmation that the human IL-6Ra gene sequence replaced the deleted mouse IL- 6Rq gene sequence in the humanized allele was med by a TaqManTM qPCR assay that comprises the following primer-probe sets (written 5’ to 3’): the human fonNard primer, GGAGAGGGCA GAGGCACTTA C (SEQ ID NO:19); the human reverse , GGCCAGAGCC CAAGAAAAG (SEQ ID N020); and the human probe, CCCGTTGACT GTAATCTGCC CCTGG (SEQ ID NO:21).

The same LONA assay was used to assay DNA purified from tail biopsies for mice derived from the targeted ES cells to determine their IL-6Ra genotypes and confirm that the humanized IL-6Rq allele had itted through the germline. Pups heterozygous for the replacement are bred to generate a mouse that is gous for the replacement of the endogenous mouse IL-6Rq gene by the human IL-6Ra (ectodomain) gene. Pups that are homozygous for the replacement are used for phenotyping.

The upstream junction of the murine locus and the sequence containing the hlL-6Rq gene is designed to be within 5’-CGAGGGCGAC TGCTCTCGCT GCCCCAGTCT GCCGGCCGCC CGGCCCCGGC TGCGGAGCCG CTCTGCCGCC CGCCGTCCCG CGTAGAAGGA AGCATGCTGG CCGTCGGCTG CGCGCTGCTG GCTGCCCTGC TGGCCGCGCC GGGAGCGGCG CTGGCCCCAA GGCGCTGCCC TGCGCAGGGT AAGGGCTTCG G (SEQ ID NO:22), wherein the final mouse nucleotide prior to the first nucleotide of the human gene is the “C” in GAAGC, and the first nucleotide of the human sequence is the first “A” in ATGCT.

The downstream junction of the sequence containing the hlL-6 gene and the murine locus is designed to be within 5’-CAAGATTATT GGAGTCTGAA ATGGAATACC TGTTGAGGGA AATCTTTATT GCCC TCAA TGCTTTTGAT TCCCTATCCC TGCAAGACCC GGGCTCGATA ACTATAACGG TCCTAAGGTA CGAG ATAACTTC-3’ (SEQ ID NO:23), wherein the final nucleotide of the human sequence is with the final “A” in CAAGA and the first nucleotide of the mouse ce is the first “C” in CCCGG; the downstream junction region also contained a onP site at the 3’ end for removal of a floxed ubiquitin promoter-driven neo cassette.

The first nucleotide of the loxp site is the first "A" in ATAAC. The junction of the neo cassette with the mouse IL-6Ra locus is designed to be within 5'-TATACGAAGT TATCCTAGGT TCTA CTCCATATGC TCACTTGCCG GCTA CGATACGGTG AGGCCCGTGC GAAGAGTGGC ACAGATCAGG AGGCTTATGT GGTCAGTCCA CAGTATGGC (SEQ ID NO:24), where the final "C" of AGCTC is the final nucleotide of the neo cassette; the first nucleotide of the mouse genome following the cassette is the initial “T” of TACTC.

Where the terms “comprise”, “comprises”, “comprised” or “comprising” are used in this ication (including the claims) they are to be interpreted as ying the presence of the stated features, integers, steps or components, but not precluding the presence of one or more other features, integers, steps or components, or group thereof.

The discussion of documents, acts, materials, devices, articles and the like is ed in this specification solely for the purpose of providing a context for the present invention. It is not suggested or represented that any or all of these matters formed part of the prior art base or were common general knowledge in the field relevant to the present invention as it d before the priority date of each claim of this application.

Claims (18)

The claims defining the invention are as follows:

1. A genetically modified mouse comprising a replacement at an endogenous mouse IL- 6 locus of a mouse gene encoding IL-6 with a human gene encoding human IL-6, wherein the human gene encoding human IL-6 is under control of endogenous mouse regulatory ts at the endogenous mouse IL-6 locus.

2. The genetically modified mouse of claim 1, wherein the human gene encoding human IL-6 ses exons 1 h 5 of the human IL-6 gene of the CTD-2369M23 bacterial artificial chromosome.

3. The genetically modified mouse of claim 1, wherein the mouse expresses a mouse IL-6Rα.

4. The genetically modified mouse of claim 1, wherein the mouse expresses a humanized IL-6Rα wherein the endogenous mouse IL-6Rα gene has been replaced with a human sequence comprising a sequence that encodes an main of a human .

5. The genetically modified mouse of claim 1, n the mouse does not exhibit a feature selected from plasmocytosis, glomerulosclerosis, glomerulonephritis, kidney failure, hypergammaglobulinemia, elevated megakaryocytes in spleen, elevated megakaryocytes in bone marrow, splenomegaly, lymph node ement, compacted abnormal plasma cells, and a combination thereof.

6. The genetically modified mouse of claim 4, wherein the humanized IL-6Rα comprises a human ectodomain.

7. The genetically modified mouse of claim 6, wherein the humanized IL-6Rα comprises a mouse transmembrane domain and a mouse cytoplasmic domain.

8. The genetically modified mouse of claim 1, wherein the mouse ses a cell that expresses an IL-6Rα that comprises a human ectodomain on the surface of the cell.

9. A genetically modified mouse, comprising a humanization of an endogenous mouse IL-6Rα gene, wherein the zation comprises a replacement of a mouse IL-6Rα ectodomain-encoding sequence with a human IL-6Rα ectodomain-encoding sequence at the nous mouse IL-6Rα locus, and wherein the humanized IL-6Rα gene is under control of endogenous mouse regulatory elements.

10. The genetically modified mouse of claim 9, wherein a contiguous mouse sequence comprising mouse exons 1 through 8 is replaced with a contiguous genomic fragment of human IL-6Rα ce encoding a human IL-6Rα ectodomain.

11. The genetically modified mouse of claim 10, wherein the contiguous genomic fragment of human IL-6Rα sequence encoding the ectodomain is found in the CTD-2192J23 bacterial artificial chromosome.

12. The genetically modified mouse of claim 9, r sing a humanized IL-6 gene comprising a replacement at an endogenous mouse IL-6 locus of a mouse gene encoding IL-6 with a human gene encoding human IL-6.

13. The genetically modified mouse of claim 12, wherein the humanized IL-6 gene is under control of nous mouse regulatory elements.

14. A method for making a humanized mouse, comprising replacing a mouse gene sequence ng mouse IL-6 with a human gene encoding human IL-6 so that the human IL-6 gene is under control of endogenous mouse tory elements.

15. A method for making a humanized mouse, comprising replacing all mouse exons encoding ectodomain sequences of mouse IL-6Rα with a human genomic fragment encoding human IL-6Rα ectodomain to form a humanized IL-6Rα gene, wherein the humanized IL-6Rα gene is under control of endogenous mouse regulatory elements.

16. A cally modified mouse comprising a humanized IL-6Rα gene comprising a replacement of mouse ectodomain-encoding sequence with human ectodomain ce, n the humanized IL-6Rα gene comprises a mouse transmembrane sequence and a mouse cytoplasmic sequence; wherein the mouse further comprises a gene encoding a human IL-6, n the genes encoding human IL-6 and humanized IL-6Rα are under control of endogenous mouse regulatory elements.

17. The genetically modified mouse of claim 16, wherein the mouse does not express a mouse IL-6Rα and does not express a mouse IL-6.

18. The genetically modified mouse according to claim 1, 9 or 16, or the method according to claim 14 or 15, substantially as herein described with nce to the Examples and/or