WO2009111086A1 - Transgenic non-human mammals with kappa light chain of xenogenous immunoglobulin - Google Patents

Abstract

Genetically modified non-human mammals (e.g., bovines, porcines, and other ungulates) and cells containing a nucleic acid encoding all or part of a xenogenous kappa immunoglobulin gene capable of undergoing rearrangement and expressing a xenogenous kappa immunoglobulin are described. Antibodies produced from these non-human mammals and cells are described. Methods of making these mammals and cells are presented. In addition, methods of producing xenogenous (e.g., human) antibodies from the genetically modified non-human mammals and cells are described.

Description

TRANSGENIC NON-HUMAN MAMMALS WITH KAPPA LIGHT CHAIN OF XENOGENOUS IMMUNOGLOBULIN

Field of the Invention

In general, the present invention relates to the field of genetic engineering.

Background of the Invention

In 1890, Shibasaburo Kitazato and Emil Behring reported an experiment with extraordinary results; particularly, they demonstrated that immunity can be transferred from one animal to another by taking serum from an immune animal and injecting it into a non-immune one. This landmark experiment laid the foundation for the introduction of passive immunization into clinical practice. Today, the preparation and use of human immunoglobulin for passive immunization is standard medical practice. In the United States alone, there is a $1.4B per annum market for human immunoglobulin, and each year more than 16 metric tons of human antibody is used for intravenous antibody therapy. Comparable levels of consumption exist in the economies of most highly industrialized countries, and the demand can be expected to grow rapidly in developing countries. Currently, human antibody for passive immunization is obtained from the pooled serum of human donors. This means that there is an inherent limitation in the amount of human antibody available for therapeutic and prophylactic usage. Already, the demand exceeds the supply and severe shortfalls in availability have been routine.

In an effort to overcome some of the problems associated with the inadequate supply of human immunoglobulin, various technologies have been developed. For example, the production of monoclonal antibody by recombinant methods in tissue culture is routine.

Particularly, the recombinant expression of monoclonal antibody against a specific antigen in CHO expression systems is well known, and is currently utilized for the production of several monoclonal antibodies now in therapeutic use.

Mice retaining an unrearranged human immunoglobulin gene have also been developed for the production of human antibodies (e.g., monoclonal antibodies) (see, for example, PCT Publication Nos. WO98/24893; WO96/33735; WO 97/13852; WO98/24884; WO97/07671 ; and U.S. Patent Nos. 5,877,397; 5,874,299; 5,814,318; 5,789,650; 5,770,429; 5,661 ,016; 5,633,425; 5,625,126; 5,569,825; and 5,545,806).

PCT Publication Nos. WOOO/10383 and WO02/092812 describe a transgenic mouse that has a human artificial chromosome fragment containing both unrearranged heavy chain genes and unrearranged light chain genes of human immunoglobulin. The transgenic mouse produced fully human rearranged immunoglobulin and fully human antibody against a target antigen. Usually the gene of an antibody that binds to a specific antigen is cloned in a mouse hybridoma and transfected into CHO cells in order to establish a production cell for a specific monoclonal antibody. An artificial chromosome that contained heavy chain genes and lambda chain genes of human immunoglobulin was germinally transmitted in mouse offspring. However, an artificial chromosome which contained heavy chain genes and kappa light chain genes of human immunoglobulin was not germinally transmitted.

U.S. Patent Nos. 5,849,992 and 5,827,690 describe the production of monoclonal antibodies in the milk of transgenic animals including mice, sheep, pigs, cows, and goats wherein the transgenic animals expressed human immunoglobulin genes under the control of promoters that provide for the expression of the antibodies in mammary epithelial cells. Essentially, this results in the expression of the monoclonal antibodies in the milk of such animals, for example, a cow.

Polyclonal antibody is useful for passive immunization for various patients who suffer infectious diseases. In order to produce human polyclonal antibody efficiently, non-human animals which produce a large amount of human polyclonal antibody in their serum or milk are preferable. For this purpose, an ungulate is one of the preferable candidates. Though ES cells are used for production of transgenic mice, there are no ES cells useful for transgenic ungulates. In order to generate a transgenic ungulate, an animal cloning method with chromosome transfer technology was developed (U.S. Patent No. 7,253,334). U.S. Patent No. 7,074,983 and U.S. Patent Application Publication Nos. 2004/0068760, 2005/0097627, and 2006/0041945 describe the expression of xenogenous human immunoglobulins in cloned, transgenic ungulates.

Notwithstanding the foregoing, further improved methods for producing ungulates that are amenable to being used as hosts for xenogenous antibody production are of great value to this industry. Summary of the Invention

In general, the invention features genetically modified non-human mammals (e.g., ungulates) that contain kappa light chain genes of xenogenous immunoglobulin and methods of producing the transgenic mammals. The invention also features methods of producing xenogenous (e.g., human) immunoglobulin using the transgenic ungulates.

In particular, we discovered that non-human mammals that contain both heavy chain and kappa light chain genes of xenogenous immunoglobulins generate antibody-producing cells and efficiently produce xenogenous (e.g., human) immunoglobulin.

Accordingly, the invention features a transgenic non-human mammal that produces xenogenous immunoglobulin efficiently, and that includes kappa genes as the source of immunoglobulin production. Desirably, the unrearranged kappa light chain is human. More desirably, the unrearranged kappa light chain is contained in a human artificial chromosome (e.g., a KHAC, such as κΔHAC-I or κΔHAC-II).

The invention is exemplified in a bovine and porcine, but is equally applicable to other mammals (e.g., mice, rats, and monkeys, or ungulates, such as bovines, ovines, and caprines). The non-human mammal may be an adult mammal (for example, an adult ungulate), a fetal mammal (for example, a fetal ungulate), or a mammalian embryo (for example, an ungulate embryo).

The invention also features a human artificial chromosome suitable for the production of xenogenous (e.g., human) immunoglobulin. Preferably, a germline transmissible artificial chromosome that contains both heavy chain genes and kappa light chain genes of xenogenous immunoglobulin is desired for this purpose (e.g., κΔHAC-1 or κΔHAC-11).

The invention also features a transgenic mammalian (e.g., ungulate) somatic cell that includes an unrearranged kappa light chain gene, as described above. Suitable somatic cells include, without limitation, fibroblasts, epithelial cells, endothelial cells, neural cells, keratinocytes, hematopoietic cells, melanocytes, chondrocytes, lymphocytes (B cells and T cells), macrophages, monocytes, mononuclear cells, cardiac muscle cells, other muscle cells, and epidermal cells.

The invention further features methods for producing non-human mammalian (e.g., ungulate) serum and/or milk that are suitable as a resource of xenogenous polyclonal antibodies. One such method for producing xenogenous antibodies includes the steps of: (a) administering one or more antigens of interest to a non-human mammal containing a nucleic acid encoding all or part of a xenogenous kappa immunoglobulin gene, wherein the nucleic acid segments in the xenogenous kappa gene locus undergo rearrangement and result in the production of antibodies specific for the antigen; and (b) recovering the milk and/or serum from the mammal. Another such method for producing xenogenous antibodies includes the step of recovering xenogenous antibodies from a non-human mammal containing a nucleic acid encoding all or part of a xenogenous kappa immunoglobulin gene, wherein the nucleic acid segments in the xenogenous kappa gene locus undergo rearrangement and result in the production of antibodies specific for the antigen. Xenogenous polyclonal antibody in non-human mammals may be highly purified using any of a variety of chromatography techniques, for example, with a method described in WO 05/1 13604, herein incorporated by reference.

Another method of antibody production includes the steps of: (a) providing a transgenic non-human mammal (e.g., ungulate) of the invention, the mammal having engrafted xenogenous hematopoietic stem cells; and (b) recovering the serum and/or milk that contain xenogenous polyclonal antibodies from this mammal.

The invention also features a method for maintaining a desired tissue or organ in vivo by: (a) providing a transgenic non-human mammal (e.g., ungulate) of the invention; (b) engrafting desired allogeneic or xenogeneic tissue or organ (e.g., skin, heart, lung, pancreatic, liver, or kidney tissue) into the mammal; and (c) maintaining the tissue or organ in the mammal in any of the foregoing aspects of the invention. The transgenic non-human mammal (e.g., ungulate) may optionally have one or more nucleic acids encoding all or part of a xenogenous kappa light chain gene that undergoes rearrangement and expresses one or more xenogenous kappa light chains. The transgenic non-human mammal (e.g., ungulate) may also have a mutation in an endogenous prion nucleic acid (see, methods in WO

04/044156, herein incorporated by reference). In a preferred embodiment, the nucleic acid encoding all or part of the xenogenous gene is substantially human. Preferably, the nucleic acid encodes a xenogenous antibody, such as a human antibody or a polyclonal antibody.

In other embodiments, the nucleic acid is contained within a chromosome fragment, such as a SC20 (FERM BP-7583, the International Patent Organism Depository, National Institute of Advanced Industrial Science and Technology, Tsukuba Central 6, 1 -1 , Higashi 1-Chome Tsukuba-shi, Ibaraki-ken, 305-8566, Japan), a ΔHAC (FERM BP-7582), a ΔΔHAC (FERM BP-7581), a κΔHAC-I, or a κΔHAC-II. In yet other embodiments, the nucleic acid is maintained in an ungulate cell independently from the host chromosome.

In still other embodiments of any of the aspects of the invention, the non-human mammal (e.g., ungulate) has a mutation in one or both alleles of an endogenous immunoglobulin gene (e.g., heavy chain immunoglobulin genes IgμAY or IgμU, lambda light chain immunoglobulin gene, and/or kappa light chain immunoglobulin gene), alpha-(l,3)- galactosyltransferase gene, PrP gene, and/or J chain gene. In other preferred embodiments, the non-human mammal (e.g., ungulate) has a nucleic acid encoding an exogenous J chain, such as a human J chain. Preferably, the mutation reduces or eliminates the expression of the endogenous immunoglobulin, alpha-(l,3)-galactosyltransferase enzyme, galactosyl(αl ,3)galactose epitope, prion protein, and/or J chain. In still other preferred embodiments, the non-human mammal (e.g., ungulate) contains a xenogenous J chain nucleic acid, such as a human J chain nucleic acid. Preferably, the non-human mammal produces human IgA or IgM molecules containing human J chain. In various embodiments of the invention, the nucleic acid used to mutate an endogenous ungulate nucleic acid (e.g., a knockout cassette which includes a promoter operably linked to a nucleic acid encoding a selectable marker and operably linked to a nucleic acid having substantial sequence identity to the gene to be mutated) is not contained in a viral vector, such as an adenoviral vector or an adeno-associated viral vector. For example, the nucleic acid may be contained in a plasmid or artificial chromosome that is inserted into a non-human mammalian cell (e.g., ungulate), using a standard method such as transfection or lipofection that does not involve viral infection of the cell. In yet another embodiment, the nucleic acid used to mutate an endogenous ungulate nucleic acid (e.g., a knockout cassette which includes a promoter operably linked to a nucleic acid encoding a selectable marker and operably linked to a nucleic acid having substantial sequence identity to the gene to be mutated) is contained in a viral vector, such as an adenoviral vector or an adeno-associated viral vector.

According to this embodiment, a virus containing the viral vector is used to infect a non-human mammalian (e.g., ungulate) cell, resulting in the insertion of a portion or the entire viral vector into the mammalian cell. In embodiments where the non-human mammal is an ungulate, the ungulate is preferably a bovine, ovine, porcine, or caprine. Preferably, the transgenic ungulate expresses an immunoglobulin chain or antibody from another genus, such as an antibody from any other mammal. Particularly preferred ungulates are cows, sheep, big-horn sheep, goats, buffalos, antelopes, oxen, horses, donkeys, mule, deer, elk, caribou, water buffalo, camels, llama, alpaca, pigs, and elephants.

The invention further features a method of producing a transgenic non-human mammal (for example, a transgenic bovine or porcine) that rearranges and expresses a heavy chain of xenogenous (e.g., human) immunoglobulin gene with a kappa light chain of xenogenous (e.g., human) immunoglobulin gene. This may be accomplished, for example, by stably introducing a human chromosome fragment containing both heavy chain and kappa light chain genes (e.g., a KHAC, such as κΔHAC-I or κΔHAC-II) into the non-human mammal, in order to produce a transgenic ungulate having B cells that produce fully xenogenous (e.g., human) immunoglobulins, in addition to or in lieu of endogenous immunoglobulins. This may also be accomplished by integrating a nucleic acid encoding a xenogenous immunoglobulin chain or xenogenous antibody into a chromosome of a non-human mammal (e.g., an ungulate). Preferably, the transgenic non-human mammal has a mutation in genes encoding IgM (e.g., IgμAY and/or IgμU), such that the expression of endogenous IgM has been reduced (e.g., by at least 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, or 90%) or eliminated. In one embodiment, the invention features a method of producing a transgenic non-human mammal (for example, a transgenic bovine) in which at least two mu constant regions have been disrupted (e.g., IgμAY and IgμU), and an artificial chromosome containing a gene locus encoding another species' immunoglobulin, preferably human, for example, a κΔHAC-I or a κΔHAC-II, has been stably incorporated. The invention also features a method of producing a non-human mammalian (e.g., ungulate) somatic or embryonic stem (ES) cell, preferably a fibroblast or B cell, wherein one or both alleles of at least the exogenous kappa light chain genes have been incorporated.

The invention also features a method of inserting, into a non-human mammal (e.g., an ungulate) in which the endogenous IgM heavy chain genes have been disrupted, a nucleic acid (for example, a human artificial chromosome) that contains genes sufficient for the functional expression of kappa light chain of a xenogenous, for example, non-ungulate, immunoglobulins. Preferably, these immunoglobulins are human immunoglobulins produced by introduction of nucleic acid encoding these immunoglobulins or immunoglobulin chains into a non-human mammalian (e.g., ungulate) somatic cell, preferably a fibroblast, and producing cloned non-human mammals in which the nucleic acid is transmitted into the germ line.

The invention also features a method for introducing an artificial chromosome, preferably a human artificial chromosome (HAC), that contains genes that provide for immunoglobulin expression into the aforementioned homozygous knockout cells to generate non-human mammals (e.g., ungulates) that express non-ungulate immunoglobulins, preferably human immunoglobulins, in response to immunization and that undergo affinity maturation.

The invention also features methods for producing hybridomas and monoclonal antibodies using B cells derived from the above-described transgenic non-human mammals (for example, transgenic ungulates). The invention also features methods for producing non-human mammal (e.g., ungulate) antiserum or milk that includes polyclonal human immunoglobulin by providing a transgenic mammal described above that is producing polyclonal human immunoglobulins, and collecting antiserum or milk from the mammal. Such human immunoglobulin, preferably human IgG, may be used as intravenenous immunoglobulin (IVIG) for the treatment or prevention of disease in humans or to mediate protection against a pathogen

(e.g., a virus, bacterium, or toxin). The polyclonal human immunoglobulins are preferably reactive against an antigen of interest. Particularly desirable is polyclonal human immunoglobulin that has a similar heavy chain subclass ratio and/or N-glycosylation to human serum immunoglobulin. Preferably, the mammal further includes a nucleic acid encoding all or part of a xenogenous immunoglobulin gene that undergoes rearrangement and expresses a xenogenous immunoglobulin (e.g., a human immunglobulin). The nucleic acid may be contained within a chromosome fragment (e.g., ΔHAC, ΔΔHAC, κΔHAC-I, or κΔHAC-II). Suitable donor cells for any of the above aspects of the invention include, without limitation, fibroblasts, epithelial cells, endothelial cells, neural cells, keratinocytes, hematopoietic cells, melanocytes, chondrocytes, B cells, T-cells, macrophages, monocytes, mononuclear cells, cardiac muscle cells, other muscle cells, placental cells, epidermal cells, embryonic cells, and germ cells.

The invention also features a method of producing xenogenous antibodies by providing the foregoing transgenic non-human mammal (e.g., ungulate), wherein the mammal includes a nucleic acid encoding all or part of a xenogenous immunoglobulin gene that undergoes rearrangement and expresses a xenogenous immunoglobulin, administering one or more antigens of interest to the mammal, and then recovering xenogenous antibodies from the mammal.

Also featured in the invention are antibodies produced by the non-human mammals described herein. These antibodies may be formulated in purified form for administration to a subject (for example, a human patient in need thereof). The xenogenous antibodies (e.g., human antibodies) produced by the methods of the invention preferably may have a ratio of the number of galactose molecules per asparagine residue (Gal/N) and/or a ratio of sialic acid molecules per asparagine residue (SA/N) that is within at least +/- 50% (e.g., with at least +/- 45% +/- 40%, +/- 35%, +/- 30%, +/- 25%, +/- 20%, +/- 15%, +/- 10%, or +/- 5%) of the Gal/N and/or SA/N ratio from an antibody from a control mammal (e.g., a monoclonal antibody made from a cell line or purified control polyclonal antibody from a mammal). The xenogenous antibodies produced by the above methods preferably contain both a human heavy chain immunoglobulin and a human kappa chain immunoglobulin. The xenogenous antibodies produced by the above methods may be a chimeric antibody. An example of a chimeric antibody is an antibody having a human heavy chain immunoglobulin and an endogenous light chain immunoglobulin (e.g., endogenous kappa light chain or endogenous lambda light chain). Such xenogenous chimeric antibodies may be administered to a mammal in a single dose or in multiple doses in order to elicit an immunoprotective response to a pathogen (e.g., a virus, a bacterium, or a toxin).

As used herein, by "allele" is meant one member of a DNA pair that occupies a specific position on a specific chromosome.

By "artificial chromosome" is meant a mammalian chromosome or fragment thereof which has an artificial modification such as the addition of a selectable marker, the addition of a cloning site, the deletion of one or more nucleotides, the substitution of one or more nucleotides, and the like. By "human artificial chromosome (HAC)" is meant an artificial chromosome generated from one or more human chromosome(s). An artificial chromosome can be maintained in the host cell independently from the endogenous chromosomes of the host cell. In this case, the HAC can stably replicate and segregate alongside the endogenous chromosomes. Alternatively, it may be translocated to, or inserted into, an endogenous chromosome of the host cell. Two or more artificial chromosomes can be introduced into the host cell simultaneously or sequentially. For example, artificial chromosomes derived from human chromosome #14 (comprising the Ig heavy chain gene), human chromosome #2 (comprising the Ig kappa chain gene), and human chromosome #22 (comprising the Ig lambda chain gene) can be introduced. Alternatively, an artificial chromosome(s) comprising both a xenogenous Ig heavy chain gene and Ig light chain gene, such as ΔHAC, ΔΔHAC, κΔHAC-I, or κΔHAC-II may be introduced. Preferably, the heavy chain loci and the light chain loci are on different chromosome arms (i.e., on different side of the centromere). In still other preferred embodiments, the total size of the HAC is less than or equal to approximately 12, 10, 9, 8, or 7 megabases.

By- a "bovine fetus" is meant a bovine in utero that is at least 30 days post-fertilization.

By "cells derived from an embryo" is meant cells that result from the cell division of cells in the embryo.

By "chimeric embryo" is meant an embryo formed from cells from two or more embryos. The resulting fetus or offspring can have cells that are derived from only one of the initial embryos or cells derived from more than one of the initial embryos. If desired, the percentage of cells from each embryo that are incorporated into the placental tissue and into the fetal tissue can be determined using standard FISH analysis or analysis of a membrane dye added to one embryo.

By "chimeric antibody" is meant an antibody produced in a non-human mammal that contains at least one endogenous immunoglobulin chain (e.g., heavy or light chain immunoglobulin) and at least one xenogenous immunoglobulin chain (e.g., heavy or light chain). Preferably, a chimeric antibody contains a xenogenous heavy chain immunoglobulin (e.g., human heavy chain immunoglobulin) and an endogenous light chain immunoglobulin (e.g., endogenous kappa light chain or endogenous lambda light chain). By "chimeric ungulate" is meant an ungulate formed from cells from two or more embryos. The ungulate can have cells that are derived from only one of the initial embryos or cells derived from more than one of the initial embryos. If desired, the percentage of cells from each embryo that are incorporated into the placental tissue and into the fetal tissue can be determined using standard FISH analysis or analysis of a membrane dye added to one embryo.

By "chromatin mass" is meant more than one chromosome not enclosed by a membrane. Preferably, the chromatin mass contains all of the chromosomes of a cell. An artificially induced chromatin mass containing condensed chromosomes may be formed by exposure of a nucleus to a mitotic reprogramming media (e.g., a mitotic extract from, e.g., a somatic cell or oocyte). Alternatively, an artificially induced chromatin mass containing decondensed or partially condensed chromosomes may be generated by exposure of a nucleus to one of the following, as described herein: a mitotic extract containing an anti-NuMA antibody, a detergent and/or salt solution, or a protein kinase solution. A chromatin mass may contain discrete chromosomes that are not physically touching each other or may contain two or more chromosomes that are in physical contact.

If desired, the level of chromosome condensation may be determined using standard methods by measuring the intensity of staining with the DNA stain, DAPI. As chromosomes condense, this staining intensity increases. Thus, the staining intensity of the chromosomes may be compared to the staining intensity for decondensed chromosomes in interphase (designated 0% condensed) and maximally condensed chromosomes in mitosis (designated 100% condensed). Based on this comparison, the percent of maximal condensation may be determined. Preferred condensed chromatin masses are at least 20, 30, 40, 50, 60, 70, 80, 90, or 100% condensed. Preferred decondensed or partially condensed chromatin masses are less than 10% condensed.

By "days of gestation" is meant the days from the time that the oocyte or embryo is transferred into a uterus.

By "donor cell" is meant a cell from which a nucleus or chromatin mass is derived, or a permeabilized cell, that is introduced into an enucleated oocyte during an animal cloning process. By "embryo" or "embryonic" is meant a developing cell mass that has not implanted into the uterine membrane of a maternal host. Hence, the term "embryo" may refer to a fertilized oocyte; an oocyte containing a donor chromatin mass, nucleus, or reprogrammed cell; a pre-blastocyst stage developing cell mass; or any other developing cell mass that is at a stage of development prior to implantation into the uterine membrane of a maternal host and prior to formation of a genital ridge. An embryo may represent multiple stages of cell development. For example, a one cell embryo can be referred to as a zygote; a solid spherical mass of cells resulting from a cleaved embryo can be referred to as a morula, and an embryo having a blastocoel can be referred to as a blastocyst. An "embryonic cell" is a cell isolated from or contained in an embryo.

By "embryo cloning" is meant the process in which an embryo is produced from a cell or cellular materials from another animal. Embryo cloning may be performed, for example, by inserting or fusing a donor cell (for example, a permealized cell), nucleus, or chromatin mass with an oocyte. The resulting oocyte or the embryo formed from this oocyte is then transferred into the uterus of an animal, thereby producing a cloned animal.

By "enrichment or depletion of a factor" is meant the addition or removal of a naturally-occurring or recombinant factor by at least 20, 40, 60, 80, or 100% of the amount of the factor originally present in a reprogramming media (e.g., a cell extract). Alternatively, a naturally-occurring or recombinant factor that is not naturally present in the reprogramming media may be added. Preferred factors include proteins such as DNA methy ransferases, histone deacetylases, histones, protamines, nuclear lamins, transcription factors, activators, and repressors; membrane vesicles, and organelles. In one preferred embodiment, the factor is purified prior to being added to the reprogramming media, as described below. Alternatively, one of the purification methods described below may be used to remove an undesired factor from the reprogramming media.

By "fetus" is meant a developing cell mass that has implanted into the uterine membrane of a maternal host. A "fetal cell" is any cell isolated from or contained in a fetus at any stage of gestation including birth.

By "fragment" is meant a nucleic acid or polypeptide having a region of consecutive nucleic acids or amino acids that is identical to the corresponding region of another nucleic acid or polypeptide, for example, an antibody gene or antibody of the invention, but is less than the full-length sequence. An antibody fragment has the ability to bind the same antigen as the corresponding antibody based on standard assays, such as those described herein. Preferably, the binding of the fragment to the antigen is at least 20, 40, 60, 80, or 90% of that of the corresponding antibody. By "gene" is meant a hereditary unit consisting of a sequence of DNA that occupies a specific location on a chromosome and that determines a particular characteristic in an organism. A gene typically has two alleles.

By "hemizygous mutation" is meant that one allele of an endogenous gene has been mutated and the other allele has not been mutated. By "homozygous mutation" is meant that two alleles of an endogenous gene have been mutated. According to this invention, the mutation-introducing event at both alleles may or may not be the same. Accordingly, two alleles of an endogenous gene genetically targeted by two different targeting vectors would be considered a homozygous mutation.

By "homozygous knock-out non-human mammal" is meant a mammal other than a human in which the two alleles of an endogenous gene have been genetically targeted, resulting in the marked reduction or elimination of expression of a functional gene product. According to this invention, the genetic targeting event at both alleles may or may not be the same. Accordingly, a non-human mammal, in which the two alleles of an endogenous gene have been genetically targeted by two different targeting vectors resulting in the null expression of the endogenous gene, would be considered as being a homozygous knock-out non-human mammal.

By "immortalized" is meant capable of undergoing at least 25, 50, 75, 90, or 95% more cell divisions than a naturally-occurring control cell of the same cell type, genus, and species as the immortalized cell or than the donor cell from which the immortalized cell was derived. Preferably, an immortalized cell is capable of undergoing at least 2-, 5-, 10-, or 20-fold more cell divisions than the control cell. More preferably, the immortalized cell is capable of undergoing an unlimited number of cell divisions. Immortalized cells include cells that naturally acquire a mutation in vivo or in vitro that alters their normal growth- regulating process. Still other preferred immortalized cells include cells that have been genetically modified to express an oncogene, such as ras, myc, abl, bcl2, or neu, or that have been infected with a transforming DNA or RNA virus, such as Epstein Barr virus or SV40 virus (Kumar et al. (1999) Immunol. Lett. 65: 153-159; Knight et al. (1988) Proc. Natl. Acad. ScL USA 85:3130-3134; Shammah et al. (1993) J. Immunol. Methods 160: 19-25; Gustafsson and Hinkula (1994) Hum. Antibodies Hybridomas 5:98-104; Kataoka et al. (1997) Differentiation 62:201-211 ; Chatelut et al. (1998) Scand. J. Immunol. 48:659-666). Cells can also be genetically modified to express the telomerase gene (Roques et al. (2001) Cancer Res. 61 :8405-8507).

By "KHAC" is meant a human artificial chromosome containing the unrearranged locus of the human immunoglobulin K chain loci.

By "κΔHAC-I, or κΔHAC-II" is meant a recombinant chromosome including: (i) a human chromosome #14 frgment that contains unrearranged human immunoglobulin heavy chain loci; (ii) a human centromere; (iii) a human chromosome #2 fragment that is 1 -4 megabases in length and contains unrearranged kappa chain loci; and (iv) two telomere sequences.

By "knock-in mutation" is meant the insertion of an exogenous nucleic acid, optionally, encoding a polypeptide, into the chromosome of a cell.

By "mutation" is meant an alteration in a naturally-occurring or reference nucleic acid sequence, such as an insertion, deletion, frameshift mutation, silent mutation, nonsense mutation, or missense mutation. Preferably, the amino acid sequence encoded by the nucleic acid sequence has at least one amino acid alteration from a naturally-occurring sequence. Examples of recombinant DNA techniques for altering the genomic sequence of a cell, embryo, fetus, or mammal include inserting a DNA sequence from another organism (e.g., a human) into the genome, deleting one or more DNA sequences, and introducing one or more base mutations (e.g., site-directed or random mutations) into a target DNA sequence. Examples of methods for producing these modifications include retroviral insertion, artificial chromosome techniques, gene insertion, random insertion with tissue specific promoters, homologous recombination, gene targeting, transposable elements, and any other method for introducing foreign DNA. All of these techniques are well known to those skilled in the art of molecular biology. A "non-natural Iy occurring mutation" is one that is introduced artificially, for example, by recombinant means. By "non-human mammal comprising multiallellic mutations" or "multiallelic non-human mammal" is meant a mammal other than a human in which two alleles of an endogenous gene have been mutated. The mutations in the two alleles may or may not be in the same location, and may or may not be due to the same type of alteration. For example, the alleles of the gene in a multiallelic non-human mammal may be mutated by the insertion of two different polynucleotide sequences. By "non-human mammal comprising multigenic mutations" or "multigenic non-human mammal" is meant a mammal other than a human in which two or more different genes have been mutated. The mutations in the two genes may or may not be in the same location and may or may not be due to the same type of alteration. For example, two genes in a multigenic non-human mammal may be mutated by the insertion of two different polynucleotide sequences.

By "nucleus" is meant a membrane-bounded organelle containing most or all of the DNA of a cell. The DNA is packaged into chromosomes in a decondensed form. Preferably, the membrane encapsulating the DNA includes one or two lipid bilayers or has nucleoporins. By "permeabilization" is meant the formation of pores in the plasma membrane or the partial or complete removal of the plasma membrane. A "permeabilized cell" has pores in its plasma membrane or has a partial plasma membrane.

By "placenta" is meant the membranous vascular organ that develops in female mammals during pregnancy, lining the uterine wall and partially enveloping the fetus, to which it is attached by the umbilical cord.

By "purified" is meant separated from other components that naturally accompany it. Typically, a factor is substantially pure when it is at least 50%, by weight, free from proteins, antibodies, and naturally-occurring organic molecules with which it is naturally associated. Preferably, the factor is at least 75%, more preferably, at least 90%, and most preferably, at least 99%, by weight, pure. A substantially pure factor may be obtained by chemical synthesis, separation of the factor from natural sources, or production of the factor in a recombinant host cell that does not naturally produce the factor. Proteins, vesicles, and organelles may be purified by one skilled in the art using standard techniques such as those described by Ausubel et al. (Current Protocols in Molecular Biology, John Wiley & Sons, 1995). The factor is preferably at least 2, 5, or 10 times as pure as the starting material, as measured using polyacrylamide gel electrophoresis, column chromatography, optical density, HPLC analysis, or Western blot analysis (Ausubel et al., supra). Preferred methods of purification include immunoprecipitation, column chromatography such as, immunoaffinity chromatography, magnetic bead immunoaffinity purification, and panning with a plate-bound antibody. By "recloned" is meant used in a subsequent (e.g., second) round of cloning. In particular, a cell from an embryo, fetus, or adult generated from the methods of the invention may be incubated in a mitotic reprogramming media (e.g., a mitotic cell extract) to form a chromatin mass for insertion into an enucleated oocyte, as described herein. Alternatively, the cell may be permeabilized to form a permeabilized cell, incubated in a reprogramming media, and inserted into an enucleated oocyte, as described herein. Performing two or more rounds of cloning may result in additional reprogramming of the donor chromatin mass or donor cell, thereby increasing the chance of generating a viable offspring after the last round of cloning.

By "reducing the expression of an endogenous antibody" is meant reducing the amount of endogenous, functional antibodies produced by a B cell or a population of B cells. This reduction in the amount of endogenous antibodies may be due to a decrease in the amount of endogenous antibodies produced per B cell, a decrease in the number of functional endogenous B cells, or a combination thereof. Preferably, the amount of an endogenous antibody secreted by a B cell or expressed on the surface of a B cell expressing or secreting endogenous antibody is reduced by at least 25, 50, 75, 90, or 95%. In another preferred embodiment, the number of endogenous B cells in a sample from the recipient mammal, such as a blood sample, is reduced by at least 25, 50, 75, 90, or 95%.

By "reprogramming media" is meant a solution that allows the removal of a factor from a cell, nucleus, chromatin mass, or chromosome, or the addition of a factor from the solution to the cell, nucleus, chromatin mass, or chromosome. Preferably, the addition or removal of a factor increases or decreases the level of expression of an mRNA or protein in the donor cell, chromatin mass, or nucleus, or in a cell containing the reprogrammed chromatin mass or nucleus. In another embodiment, incubating a permeabilized cell, chromatin mass, or nucleus in the reprogramming media alters a phenotype of the permeabilized cell or a cell containing the reprogrammed chromatin mass or nucleus relative to the phenotype of the donor cell. In yet another embodiment, incubating a permeabilized cell, chromatin mass, or nucleus in the reprogramming media causes the permeabilized cell or a cell containing the reprogrammed chromatin mass or nucleus to gain or lose an activity relative to the donor cell. In another embodiment, incubating a permeabilized cell in the reprogramming media allows the cell to remain membrane-bounded and does not result in the condensation of the nuclear chromatin.

Exemplary reprogramming media include solutions, such as buffers, that do not contain biological molecules such as proteins or nucleic acids. Such solutions are useful for the removal of one or more factors from a nucleus, chromatin mass, or chromosome. Other preferred reprogramming medias are extracts, such as cellular extracts from cell nuclei, cell cytoplasm, or a combination thereof. Exemplary cell extracts include extracts from oocytes (e.g., mammalian, vertebrate, or invertebrate oocytes), male germ cells (mammalian, vertebrate, or invertebrate germ cells, such as spermatogonia, spermatocyte, spermatid, or sperm), and stem cells (e.g., adult or embryonic stem cells). Yet other reprogramming media are solutions or extracts to which one or more naturally-occurring or recombinant factors (e.g., nucleic acids or proteins such as DNA methyltransferases, histone deacetylases, histones, protamines, nuclear lamins, transcription factors, activators, repressors, chromatin remodeling proteins, growth factors, interleukins, cytokines, or other hormones) have been added, or extracts from which one or more factors have been removed. Still other reprogramming media include solutions of detergent (e.g., 0.01% to 0.1%, 0.1% to 0.5%, or 0.5% to 2% ionic or non-ionic detergent, such as one or more of the following detergents: SDS, Triton X-100, Triton X-1 14, CHAPS, Na-deoxycholate, n-octyl glucoside, Nonidet P40, IGEPAL, Tween 20, Tween 40, or Tween 80), salt (e.g., ~0.1 , 0.15, 0.25, 0.5, 0.75, 1 , 1.5, or 2 M NaCI or KCl), polyamine (e.g., ~1 μM, 10 μM, 100 μM, 1 mM or 10 mM spermine, spermidine, protamine, or poly-L-lysine), a protein kinase (e.g., cyclin-dependent kinase 1, protein kinase C, protein kinase A, MAP kinase, calcium/calmodulin-dependent kinase, CKl casein kinase, or CK2 casein kinase), and/or a phosphatase inhibitor (e.g., -10 μM, 100 μM, 1 mM, 1 O mM, 5O mM, 100 mM of one or more of the following inhibitors: Na-orthovanadate, Na-pyrophosphate, Na-fluoride, NlPPl , inhibitor 2, PNUTS, SDS22, AKAP149, or ocadaic acid) or nuclioplasmin. In some embodiments, the reprogramming medium contains an anti-NuMA antibody. If desired, multiple reprogramming media may be used simultaneously or sequentially to reprogram a donor cell, nucleus, or chromatin mass.

By "reprogrammed cell" is meant a cell that has been exposed to a reprogramming media. In one example, a reprogrammed cell is created by incubating the cell in a reprogramming media that allows the cell to remain membrane-bounded and does not result in the condensation of the nuclear chromatin. Preferably, at least 1 , 5, 10, 15, 20, 25, 50, 75, 100, 150, 200, 300, or more mRNA or protein molecules are expressed in the reprogrammed cell that are not expressed in the donor or permeabilized cell. In another preferred embodiment, the number of mRNA or protein molecules that are expressed in the reprogrammed cell, but not expressed in the donor or permeabilized cell, is between 1 and 5, 5 and 10, 10 and 25, 25 and 50, 50 and 75, 75 and 100, 100 and 150, 150 and 200, or 200 and 300, inclusive. Preferably, at least 1, 5, 10, 15, 20, 25, 50, 75, 100, 150, 200, 300, or more mRNA or protein molecules are expressed in the donor or permeabilized cell that are not expressed in the reprogrammed cell. In yet another preferred embodiment, the number of mRNA or protein molecules that are expressed in the donor or permeabilized cell, but not expressed in the reprogrammed cell, is between 1 and 5, 5 and 10, 10 and 25, 25 and 50, 50 and 75, 75 and 100, 100 and 150, 150 and 200, or 200 and 300, inclusive. In still another preferred embodiment, these mRNA or protein molecules are expressed in both the donor cell (i.e., the donor or permeabilized starting cell) and the reprogrammed cell, but the expression levels in these cells differ by at least 2, 5, 10, or 20-fold, as measured using standard assays (see, for example, Ausubel et al., supra).

By "substantially identical" is meant having a sequence that is at least 80, 90, 95, 98, or 100% identical to that of another sequence. Sequence identity is typically measured using BLAST^® (Basic Local Alignment Search Tool) or BLAST^® 2 with the default parameters specified therein (see, Altschul et al. (199O) J. MoI. Biol. 215:403-410; and

Tatiana et al. ( 1999) FEMS Microbiol. Lett. 174:247-250). These software programs match similar sequences by assigning degrees of homology to various substitutions, deletions, and other modifications. Conservative substitutions typically include substitutions within the following groups: glycine, alanine, valine, isoleucine, leucine; aspartic acid, glutamic acid, asparagine, glutamine; serine, threonine; lysine, arginine; and phenylalanine, tyrosine. By "N-linked glycosylation substantially similar to control antibodies" is meant a ratio of the number of galactose molecules per asparagine residue (Gal/N) and/or a ratio of sialic acid molecules per asparagine residue (SA/N) that is within at least +/- 50%, +/- 45% +/- 40%, +/- 35%, +/- 30%, +/- 25%, +/- 20%, +/- 15%, +/- 10%, or +/- 5% of the Gal/N and/or SA/N ratio from an antibody from a control mammal (e.g., a monoclonal antibody made from a cell line or purified polyclonal antibody from a control mammal). A control antibody for the glycosylation of a xenogenous human antibody (e.g., fully human antibody or chimeric human antibody produced by the method of the invention) may be control human IgG from a human donor. Methods of determining the Gal/N and SA/N ratio for an antibody sample are described herein and are also described in Kuroiwa et al. {Nature Biotechnol. 27:173-181, 2009).

By "viable offspring" is meant an animal that survives ex utero. Preferably, the mammal is alive for at least one second, one minute, one hour, one day, one week, one month, six months, or one year from the time it exits the maternal host. The animal does not require the circulatory system of an in utero environment for survival.

Other features and advantages of the invention will be apparent from the following detailed description and from the claims.

Brief Description of the Drawings FIGURE 1 is a schematic diagram depicting sequential gene targeting in bovine primary fibroblasts. Holstein fetal fibroblasts (#6939) were targeted, after which wells containing targeted cells were selected and cloned, using a chromatin transfer system to generate IgμU^"/+ fetuses. The IgμU^"/+ cell line (#3287) was then used for the production of calves and for targeting the second allele of IgμU. Once again, cells were selected and regenerated by production of fetuses. Fetuses were harvested for the production of IgμU^"'^" cell lines, gene expression analysis, and production of calves. An IgμU^''^' cell line (#4658) was transfected with a Cre-recombinase expression plasmid to remove both the neo and puro genes simultaneously, followed by a third round of chromatin transfer to generate cloned fetuses and cell lines, in which both neo and puro selection marker genes were excised. One Cre-excised IgμU^"7" fibroblast cell line (#1404) was used for a third round of gene targeting to produce triple targeted, Cre/IgμU '/PrP^"7* fetuses and cell lines. One cell line (#8334) was subjected to the fourth round of gene targeting to produce double homozygous KO (cre/IgμU 'YPrP^"'^") fetuses and cell lines and for the evaluation of PrP gene expression.

FIGURE 2 A is a schematic diagram representing the structure of IgμU constant region locus in #6939, the puro vector used for the first round of targeting, and the genomic PCR assay used for the targeting event. The targeting vector was composed of a 5 ' homologous arm (7.2 kb), a 3' homologous arm (2.0 kb), STOP cassette containing transcriptional and translational stop sequences, DT-A (diphtheria toxin A gene), and a floxed puro gene. The vector was designed to insert the knockout cassette into exon 2 of the IgμU constant region locus. In #6939 fibroblasts, polymorphic sequences were found to distinguish allele A and allele B, as indicated. Primer pairs, puroF2 x puroR2 were used to identify the first targeting event. PCR and sequencing product showed that the vector was integrated into allele B in cell line #3287 and in allele A in #2184-1 and 2 cell lines, based on the polymorphic sequences presented in the PCR product.

FIGURE 2B is a photograph representing the identification of lgμU^"/+ fetuses by genomic PCR with puroF2 x puroR2 primers. N is a negative control (mixture of the 1 st KO vector and #6939 genomic DNA) and P is a positive control (mixture of about 10⁴ copies/μl of plasmid DNA covering puroF2-puroR2 region and #6939 genomic DNA). Cell lines #2184-1, #2184-2 and #3287 were IgμUΛ

FIGURE 2C is a photograph representing the genotyping of IgμU^"/+ calves by genomic PCR with puroF2 x puroR2 primers. N is a negative control (mixture of the 1st KO vector and #6939 genomic DNA) and P is a positive control (mixture of about 10⁴ copies/μl of plasmid DNA covering puroF2-puroR2 region and #6939 genomic DNA). Out of 13 IgμU^"/+ calves born (also shown in FIGURE 2C), five were genotyped and found to be positive to the first targeting event. FIGURE 3A is a schematic diagram depicting the structure of IgμU^"/+ #3287 alleles, the neo vector used for targeting the second allele, and the genomic PCR assay for the targeting events. Primer pairs, neoF3 x neoR3 were used to identify the neo targeting event at allele A. BCμf x BCμr is a primer pair used to confirm the absence of wild-type alleles. The primers would not amplify sequence from the targeted alleles because of the presence of STOP cassettes. FIGURE 3B is a series of photographs representing the identification of IgμU^"'^" fetuses and fibroblasts by genomic PCR with puroF2 x puroR2, neoF3 x neoR3, and BCμf x BCμr primers. "P" is a positive control (mixture of about 10⁴ copies/μl of plasmid DNA covering either puroF2-puroR2 or neoF3-neoR3 region and #6939 genomic DNA). "N" is a negative control (mixture of either the 1 st KO or 2nd KO vector and #6939 genomic DNA), and #6939 is the original fibroblast cell line. Cell lines #4658, #3655, #5109, #5139, and #4554 were positive for the targeting events both at allele A (neo-targeting) and B (puro-targeting), but negative for wild-type alleles.

FIGURE 3C is a photograph showing RT-PCR analysis of IgμU expression in mRNA extracted from spleen in day 90 fetuses. Clear expression from a positive control "P" (commercially available polyA+ bovine spleen RNA) and the wild-type (#6939) fetuses (#1 , #2), but not from IgμU^"'^" fetuses, was detected.

FIGURE 3D is a series of photographs showing the genotyping of IgμU^"'^" calves by genomic PCR with puroF2 x puroR2, neoF3 x neoR3 and BCμf x BCμr primers. N is a negative control (mixture of either the 1 st KO or 2nd KO vector and #6939 genomic DNA) and P is a positive control (mixture of about 10⁴ copies/μl of plasmid DNA covering either puroF2-puroR2 or neoF3-neoR3 region and #6939 genomic DNA). The two IgμU^"'^" calves born (one of which is shown in FIGURE 3D) were genotyped and were positive for targeting events at both allele B and A of IgμU gene but were negative for the wild-type allele. FIGURE 4A is a schematic diagram representing the structure of IgμU^"'^" #4658 alleles and the genomic PCR assay for Cre-loxP mediated removal of selection marker genes. Amplification from primer pairs, CreExF x CreExR, results in a 2.5 kb fragment from the puro targeted allele, a 4.3 kb from the neo targeted allele, or a short 0.4 kb fragment when both selection marker genes are excised. FIGURE 4B is a photograph showing the identification of Cre/IgμU^"'^" fetuses and fibroblasts by genomic PCR with CreExF x CreExR primers. In #4658 cell line, prior to introduction of Cre, 2.5 kb (puro) and 4.3 kb (neo) PCR products are detected. In the five Cre/IgμU^"'^" fetuses and cell lines, these bands completely disappear and, instead, a 0.4 kb (without puro and neo) band is detected. FIGURE 5 is a schematic illustration showing the genomic organization of IgμAY.

FIGURE 6 shows the sequence of the AY and αy alleles of IgμAY. FIGURE 7 is a schematic illustration showing the AY KO vector and the ay KO vector.

FIGURE 8A depicts flow cytometry analysis in peripheral blood of IgμAY^"'^", IgμU^"7", IgμAY^"'^" IgμU^"'^", and control calves, stained with anti-B220 and anti-IgM antibodies. FIGURE 8B depicts RT-PCR analysis to detect VDJ-rearranged IgG (primer pair;

BLl 7 x βXγlP2) transcripts in PBMC of IgμAY^"'^', IgμU^''^", and IgμAY^"'^" IgμU^"'^' calves.

FIGURE 8C illustrates the immune response to OVA immunization between IgμAY^"'^", IgμU^"'^", and IgμAY^"'^" IgμU^"'^" calves.

FIGURE 9 is a schematic diagram representing the structures of human artificial chromosomes designated as kΔHAC-I and kΔHAC-II. Both artificial chromosomes contain human kappa light chain locus (Igk) and human heavy chain locus (IgH).

FIGURE 1OA is an illustration showing flow cytometry analysis in peripheral blood of HAC/IgμAY^"'^", HAC/IgμU^"'^", and HAC/IgμAY^"'^" IgμU^''^' fetuses, stained with anti-bovine B220 and anti-human IgM antibodies. FIGURE 1OB is an illustration showing flow cytometry analysis in peripheral blood of HAC/IgμAY^"'^" and HAC/IgμAY^"'^" IgμU^"'^' fetuses, stained with anti-bovine CD21 and anti-human IgM antibodies.

FIGURE 1OC is an illustration showing flow cytometry analysis in peripheral blood of HAC/IgμAY^"'^" and HAC/IgμAY^"'^" IgμU^"'^" fetuses, stained with anti-human IgM and anti-human IgK or Igλ antibodies.

FIGURE 1OD is an illustration showing flow cytometry analysis in peripheral blood of κHACII/IgμAY^"'^" and λHAC/IgμAY^"'^" fetuses, stained with anti-bovine IgM and anti-human IgM antibodies.

FIGURE 1OE is an illustration showing a comparison of CDR3 diversity between bovine and human IGHG transcripts in a κHACII/IgμAY^"'^" fetus. The figure shows the direct sequencing of RT-PCR products amplified with the primers BLl 7 x bCγl R2 for the bovine immunoglobulin heavy chain G gene and with VH(AlI)MIX x hCγl R2 for human immunoglobulin heavy chain G gene. Top, CDR3 diversity of bovine immunoglobulin heavy chain G gene; bottom, CDR3 diversity of human immunoglobulin heavy chain G gene in a KHACII/IgμAY^"'^' fetus. FIGURE 11 is an illustration showing flow cytometry analysis of the B-cell population in Ileal Peyers's patch of a KHAC/IgμAY^''^" IgμU^"'^" calf, a κHAC/IgμU^"'^" calf, and a control calf stained with anti-CD21 and hlgM (or blgM) antibodies.

FIGURE 12 is a graph demonstrating a more rapid increase in human immunoglobulin level in kappa HAC bovines than in lambda HAC bovines. The level of human IgG in bovine serum was examined in each transgenic bovine.

FIGURE 13A is a graph showing the titer of Anthrax-? A. specific hlgG antibody (units/mL) in a κHAC/IgμAY^"'^" IgμU^"7" calf (468) and in a κHAC/IgμU^"'^" calf (1495) over time. The dashed line represents the Anthrax-VA specific total hlgM titer (units/mL) in the icHAC/IgμAY^"'^" IgμU^"'^" calf (468).

FIGURE 13B is a Western blot probed with anti-blgG (H + L) (left panel) and with anti-bovine λ chain (blgλ) (right panel): lane 1 , purified fully human hIgG/hκ chain; lane 2, purified chimeric hlgG; lane 3, purified commercial blgG; and lane 4, purified commercial hlgG. FIGURE 13C is a Western blot probed with anti-hlgG (H + L) (left panel) and with anti-human K chain (right panel): lane 1 , purified fully human hIgG/hκ chain; lane 2, purified chimeric hlgG; lane 3, purified commercial hlgG; and lane 4, purified commercial blgG.

FIGURE 14 is an illustration showing chromatographs produced from capillary gel electrophoresis with helium-cadmium laser-induced fluorescent detection (CE-LIF) of recombinant monoclonal hlgG produced in CHO cells, blgG from wild-type cattle, fully human hIgG/hκ chain (liκ) from a κHAC/IgμAY^"'^" IgμU^"'^" calf, chimeric hlgG from a κHAC/IgμAY^"'^" IgμU^"'^" calf, and hlgG from human donors (polyglobin-N). S1 -S2 represents monosialyl and bisialyl acids (sialic acid content); LP represents mannose and/or afucosylation (fucosylation content); GO, G 1 , G 1 ', G2 represent gal structure (gal content); and GO-GN, G l -GN, and GIcNAc (GIcNAc content) represent branched glycosylation patterns.

FIGURE 15 is a graph of the results of an in vivo mouse protecton assay showing the number of surviving mice over time following administration of Anthrax PA -specific hlgG produced from icHAC/lgμAY^"'^" IgμU^"'^" calves: Hu, purified fully human h!gG/hκ chain from κHAC/IgμAY^"'^" IgμU^"'^" calves at V4 of PA-immunization; Chi, purified chimeric hlgG from KHAC/IgμAY^"'^" igμU^"'^* calves at V4 of PA-immunization; Mix, purified total hlgG from κHAC/IgμAY^"A IgμU^"'^" calves at V4 of PA-immunization; Bovine, hyperimmunized pooled wild-type blgG at Vl 6 of PA-immunization.

FIGURE 16 is data showing the percentage of peripheral blood leukocytes from a transgenic piglet that contain κΔHAC-I.

FIGURES 17A-17B show the sequence of the RT-PCR product of human Igμ transcript from transgenic piglet #102-1 and #104-1 containing κΔHAC-I. The V, D, and J segments of the predicted protein product are indicated.

FIGURES 18A-18B shows the sequence of the RT-PCR product of human Igκ from transgenic piglet #102-1 and #104-1 containing κΔHAC-I. The V and J segments of the predicted protein product are indicated.

Detailed Description The present invention features transgenic non-human mammals (e.g., mice, monkeys, bovines, porcines, and other ungulates), and methods of making these mammals. In particular, the invention features transgenic non-human mammals (e.g., ungulates) producing xenogenous (e.g., human) immunoglobulin that includes a kappa light chain. The invention is described in more detail below.

Methods for Producing Non-Human Mammals and Cells Containing Multiallelic or Multigenic Mutations

One method of producing multiallelic or multigenic mutations in a mammalian cell employs the use of the Cre/Lox system. The Cre/Lox system may be used to facilitate efficient targent gene deletion. In one embodiment of the invention, fetal fibroblasts carrying the targeting vector are transfected via electroporation with a Cre-containing plasmid. A Cre plasmid that contains a GFP-cre fusion gene (Gagneten et al. (1997) Nucleic Acids Res. 25:3326-3331) may be used. Use of this particular Cre plasmid allows the rapid selection of clones that contain the Cre protein. The Cre-expressing cells are selected either by FACS sorting or by manual harvesting of green fluorescing cells via micromanipulation. Cells that are green are expected to carry actively transcribed Cre recombinase and hence delete the drug resistance marker. Cells selected for Cre expression are cloned and clones analyzed for the deletion of the drug resistance marker via PCR analysis. Those cells that are determined to have undergone excision are grown to small clones, split, and one aliquot is tested in selective medium to ascertain with certainty that the drug resistance gene has been deleted. The other aliquot is used for the next round of targeted deletion. In additional methods for producing multiallelic or multigenic mutations in a mammalian cell, a xenogenous nucleic acid molecule encoding a desired polypetide may be inserted into an endogenous gene as part of the introduced mutation. For example, genes encoding antibodies of a particular species may be introduced into an endogenous gene. Preferably, human artificial chromosomes are used for this purpose, such as those disclosed in PCT Publication Nos. WO 97/07671 and WO 00/10383, each of which are herein incorporated by reference. These human artificial chromosomes also are described in a corresponding issued Japanese Patent JP 30300092. The construction of artificial human chromosomes that contain and express human immunoglobulin genes is disclosed in Shen et al. (1997) Hum. MoI. Genet. 6: 1375-1382; Kuroiwa et al. (2000) Nature Biotechnol. 18: 1086-1090; and Loupert et al. (1998) Chromosome 107:255-259. Following the stable insertion of the artificial chromosome, the cell line (e.g., a bovine fetal fibroblast) may be used as a donor cell for further gene targeting.

As an alternative to the use of human artificial chromosomes, polynucleotides encoding genes of interest may also be integrated into the chromosome using a YAC vector, BAC vector, or cosmid vector. Such vectors may be introduced into cells (e.g., fetal fibroblasts cells) using known methods, such as electroporation, lipofection, fusion with a yeast spheroplast comprising a YAC vector, and the like. Desirably, vectors containing genes of interest may be targeted to the endogenous corresponding gene loci of the cells (e.g., fetal fibroblasts), resulting in the simultaneous introduction of the gene of interest and the mutation of the endogenous gene.

Integration of a nucleic acid encoding a gene of interest may also be carried out as described in the patents by Lonberg et al. (U.S. Patent Nos. 5,545,806, 5,569,825, 5,625,126, 5,633,425, 5,661 ,016, 5,750, 172, 5,770,429, 5,789,650, 5,814,318 5,874,299, 5,877,397, and 6,300, 129, each herein incorporated by reference). In the "knock-in" construct used for the insertion of gene of interest into a chromosome of a host mammal, one or more genes and an antibiotic resistance gene may be operably-1 inked to a promoter which is active in the cell type transfected with the construct. For example, a constitutively active, inducible, or tissue-specific promoter may be used to activate transcription of the integrated antibiotic resistance gene, allowing transfected cells to be selected based on their resulting antibiotic resistance. Alternatively, a knock-in construct in which the knock-in cassette containing the gene(s) of interest and the antibiotic resistance gene is not operably linked to a promoter may be used. In this case, cells in which the knock-in cassette integrates downstream of an endogenous promoter may be selected based on the resulting expression of the antibiotic resistance marker under the control of the endogenous promoter. These selected cells may be used in the embryo cloning procedures described herein to generate a transgenic non-human mammal containing a gene of interest integrated into a host chromosome. Alternatively, an animal containing exogenous genes of interest may be mated with an animal in which the endogenous gene is inactivated.

The disruption of a gene of interest first involves the production of hemizygous gene knockout cells and the production of a fetus by embryonic cloning. Genetically targeted cells are next harvested from the resulting fetuses at any time during gestation. In a bovine, such cell harvesting desirably occurs at between 25 to 90 days of gestation, at between 35 to 60 days of gestation, at between 35 to 50 days of gestation, preferably at between 35 to 45 days, more preferably at between 38 to 43 days, and most preferably at about 40 days of gestation. In a porcine, such cell harvesting desirably occurs at between 25 to 90 days of gestation, at between 35 to 60 days of gestation, at between 35 to 50 days of gestation, and preferably between 35 to 45 days of gestation. Next, the second allele of the same gene locus, or alternatively, an allele of a different endogenous gene is targeted in the harvested cells. These cells are next used to derive fetuses, from which somatic cells such as fibroblasts may further be isolated and used for further rounds of cloning. The above steps may then be repeated until cells containing the desired multiallelic or multigenic mutations are generated. If desired, these cells may be used to produce non-human mammals, such as ungulates, as described herein.

In general, the genetic targeting events of the invention may include inactivation, removal, or modification of a gene; upregulation of a gene; gene replacement; or transgene replacement at a predetermined locus. Examples of genes that may be targeted resulting in their inactivation, removal, or modification are genes encoding antigens which are xenoreactive to humans (e.g., α-1 , 3 -galactosy transferase); antibody-encoding genes; genes in the PrP locus responsible for the production of the prion protein and its normal counterpart in non-human animals; genes which in humans are responsible for genetic disease and which in modified, inactivated, or deleted form could provide a model of that disease in animals (e.g., the cystic fibrosis transmembrane conductance regulator gene); genes responsible for substances which provoke food intolerance or allergy; genes responsible for the presence of particular carbohydrate residues on glycoproteins (e.g., the cytidine monophospho-N-acetyl neuraminic acid hydroxylase gene in non-human animals); and genes responsible for the somatic rearrangement of immunoglobulin genes, such as RAG l and RAG2.

Among genes that can be targeted resulting in their upregulation are genes responsible for suppression of complement-mediated lysis (e.g., porcine CD59, DAF, and MCP). Furthermore, and as described above, replacement of genes may also be performed. Genes that may be replaced include genes responsible for the production of blood constituents (e.g., serum albumin), genes responsible for substances that provoke food intolerance or allergy, immunoglobulin genes, and genes responsible for surface antigens.

Additional examples of genes that may be knocked out, include endogenous immunoglobulin heavy chain and/or light chain genes (e.g., IgμAY, IgμU, Igλ, and lgκ, see, U.S. Patent Application Nos. 2003/0037347, 2004/0068760, and 2006/0041945, herein incorporated by reference). In another example, an endogenous ungulate Ig J chain gene may be knocked out to prevent the potential antigenicity of the ungulate Ig J chain in the antibodies of the invention that are administered to humans. For the construction of the targeting vector, the cDNA sequence of the bovine Ig J chain region found in GenBank Accession No. U02301 may be used. This cDNA sequence may be used as a probe to isolate the genomic sequence of bovine Ig J chain from a BAC library such as RPC 1-42

(BACPAC in Oakland, CA) or to isolate the genomic sequence of the J chain from any other mammal. Additionally, the human J chain coding sequence may be introduced into the mammals of the present invention for the functional expression of human IgA and IgM molecules. The cDNA sequence of human J chain is available as GenBank Accession Nos. AH002836, M 12759, and M 12378. This sequence may be inserted into a non-human mammalian (e.g., ungulate) fetal fibroblast using standard methods, such as those described herein. For example, the human J chain nucleic acid in a HAC, YAC vector, BAC vector, cosmid vector, or knock-in construct may be integrated into an endogenous mammalian (e.g., ungulate) chromosome or maintained independently of endogenous ungulate chromosomes. The resulting transgenic non-human mammalian cells may be used in the embryo cloning methods described herein to generate the desired mammals that have a mutation that reduces or eliminates the expression of functional ungulate J chain and that contain a xenogenous nucleic acid that expresses human J chain.

In another example, if a non-human mammal, such as an ungulate, is genetically engineered to produce a human antibody, it may be desirable to also reduce or eliminate the expression of the endogenous α-(l,3)-galactosyltransferase gene, a gene that encodes an enzyme that produces the galactosyl(α-l ,3)galactose epitope. Glycosylated human antibodies modified by this carbohydrate epitope are sometimes inactivated or eliminated when administered as therapeutics to humans by recipient antibodies reactive with the epitope. To eliminate this possible immune response in transgenic bovines, the sequence of bovine α-(l ,3)-galactosyltransferase gene may be used to design a knockout construct to inactive this gene. The bovine sequence (GenBank Accession No. J04989; Joziasse et al. (1989) J. Biol. Chem. 264: 14290-14297) or the porcine α-( 1,3 )-galactosy transferase sequence (disclosed in U.S. Patent Nos. 5,821,117 and 6,153,428, herein incorporated by reference) may be used to inactivate the genes in those species or to obtain the genomic α-(l,3)-galactosyltransferase sequence from a variety of other mammals (e.g., ungulates) to generate mammals with reduced or eliminated expression of the epitope.

The endogenous PrP gene (encoding prion protein) may also be mutated or inactivated to reduce the potential risk of an infection such as bovine spongiform encephalopathy (BSE). Mutation of the bovine PrP gene is described in U.S. Patent Application No. 2005/0097627 (herein incorporated by reference) and Right et al. {Nature Biotechnology 25(1): 132-138, 2007).

The additional mutations or the gene inactivation mentioned above may be incorporated into the non-human mammals of the present invention using various methodologies. Once a transgenic non-human mammalian cell line is generated for each desired mutation, crossbreeding may be used to incorporate these additional mutations into the mammals of the present invention. Alternatively, fetal fibroblast cells that have these additional mutations can be used as the starting material for the knockout of endogenous immunoglobulin genes and/or the introduction of xenogenous immunoglobulin genes. Also, as is described herein, fetal fibroblast cells having a knockout mutation in endogenous immunoglobulin genes (e.g., IgμAY, IgμU, lgκ, or Igλ) and/or containing xenogenous immunoglobulin genes can be used as a starting material for these additional mutations or inactivations.

Gene targeting in somatic non-human mammalian cells is used as a method as described in Kuroiwa et al. {Nature Genetics, Volume 36, Number 7, page 775-780, 2004).

Targeting Constructs

Targeted gene mutation requires generating a nucleic acid construct having regions of homology to the targeted allele in the gene of interest such that integration of the construct into the genomic allele disrupts its expression. Thus, to alter a gene, a targeting vector is designed to contain three main regions. The first region is homologous to the locus to be targeted. The second region is a polynucleotide sequence (e.g., encoding a selection marker such as an antibiotic resistance protein) that specifically replaces a portion of the targeted locus. The third region, like the first region, is homologous to the targeted locus but is not contiguous with the first region in the wild-type genome. Homologous recombination between the targeting vector and the desired wild-type locus results in deletion of locus sequences between the two regions of homology represented in the targeting vector and replacement of that sequence, for example, with a drug resistance marker. The uniqueness of each vector used is in the locus chosen for gene targeting procedures and the sequences employed in that strategy. This approach may be used in all mammals, including ungulates such as, goats {Capra hircus), sheep (Ovis aries), pigs (Sits scrofά), and cattle (Bos taurus or Bos indicus). Exemplary vectors for carrying out such targeted mutation are described herein. Methods for constructing vectors that provide for homologous recombination at other targeted sites are well known to those skilled in the art. Moreover, the construction of a suitable vector is within the level of skill in the art. In order to facilitate homologous recombination, the vectors used to effect homologous recombination and inactivation of a gene of interest, respectively, contain portions of DNA that exhibit substantial sequence identity to the genes to be targeted. Preferably, these sequences have at least 98% sequence identity, more preferably, at least 99% sequence identity, and even 100% sequence identity with the targeted gene loci to facilitate homologous recombination. In preferred embodiments, the total size of the two regions of homology is approximately 9 - 9.5 kilobases and the size of the second region that replaces a portion of the targeted locus is approximately 2 kilobases.

Typically, the construct includes a marker gene that allows for the selection of desired homologous recombinants, for example, cells in which the gene of interest has been disrupted by homologous recombination. Marker genes include antibiotic resistance markers, drug resistance markers, and green fluorescent protein, among others. One neomycin resistance construct was assembled as follows. A construct designated

"pSTneoB" (Katoh et al. (1987) Cell Struct. Fund. 12:575; Japanese Collection of Research Biologicals (JCRB) Deposit No. VE039) was designed to contain a neomycin resistance gene under the control of an SV40 promoter and TK enhancer upstream of the coding region. Downstream of the coding region is an SV40 terminator sequence. The neo cassette was excised from "pSTneoB" as an Xho\ fragment. After the ends of the fragment were converted to blunt ends using standard molecular biology techniques, the blunt ended fragment was cloned into the EcoKV site in the vector, pBS246 (Gibco/Life Technologies). This site is flanked by loxP sites. The new construct, designated "pLoxP-STNeoR", was used to generate the mu knockout DNA construct. The desired fragment of this construct is flanked by loxP sites and NoU sites, which were originally present in the pBS246 cloning vector. The desired Notl fragment, which contains loxP-neo-loxP, was used for replacement of the immunoglobulin mu constant region exons. The SV40 promoter operably linked to the neomycin resistance gene activates the transcription of the neomycin resistance gene, allowing cells in which the desired No/1 fragment has replaced the mu constant region exons to be selected based on their resulting antibiotic resistance.

The strategy used herein to target genes in ungulates (i.e., removal of a portion of the coding region and intervening sequences using a vector containing regions homologous to the regions immediately flanking the removed exons) may also be used in other mammals. For example, extensive sequence analysis has been performed on one immunoglobulin heavy chain locus of sheep (Ovis aries), and the sheep locus is highly similar to the bovine locus in both structure and sequence (GenBank Accession Νos. Z71572, Z49180 through Z49188, M60441, M60440, AF 172659 through AFl 72703). In addition to the large number of cDNA sequences reported for rearranged Ovis aries immunoglobulin chains, genomic sequence information has been reported for the heavy chain locus, including the heavy chain 5' enhancer (GenBank Accession No. Z98207), the 3' mu switch region (GenBank Accession No. Z98680), and the 5' mu switch region (GenBank Accession No. Z98681 ). The complete mRNA sequence for the sheep secreted form of the heavy chain has been deposited as GenBank Accession No. X59994. This deposit contains the entire sequence of four coding exons, which are very similar to the corresponding bovine sequence. Accordingly, the GenBank sequence may be used to determine areas of high homology with the bovine IgμU sequence for the design of PCR primers. Because non-isogenic DNA was used to target bovine cells, finding areas of high homology with sheep sequence was used as an indicator that similar conservation of sequences between breeds of cow was likely. Given the similarity between the sequences and structures of the bovine and ovine immunoglobulin loci, the targeting strategies used herein to remove bovine immunoglobulin loci may be applied to the ovine system. In addition, existing information on goats {Capra hircus; GenBank Accession No. AF 140603) indicates that the immunoglobulin locus this species is also sufficiently similar to the bovine loci to utilize the present targeting strategies.

In one method of targeting cells, the targeting construct includes regulatory expression for driving expression of the marker gene, as well as a polyadenylation signal sequence. Such a construct allows for detection of the inserted sequence independent of the expression of the mutagenized gene and thus permits the identification of recombinants in silent genes (i.e., genes that are not expressed in fibroblasts). In order to determine whether the marker gene integrated into the genome by means of homologous recombination rather than through random insertion, one may use standard molecular biology techniques such as Southern blotting, PCR, or DNA sequencing.

Selection of Targeted Cells

Genetically targeted cells are typically identified using a selectable marker. If a cell already contains a selectable marker, however, a new targeting construct containing a different selectable marker may be required. Alternatively, if the same selectable marker is employed, cells may be selected in the second targeting round by raising the drug concentration (for example, by doubling the drug concentration).

As described above, targeting constructs may also contain selectable markers flanked by loxP sites to facilitate the efficient deletion of the marker using the Cre/Lox system. Thus, at some point after the gene targeting event and preferably before any embryo cloning, such excision may be performed to remove portions of genetic material from the cell. This material may be a selectable marker or an introduced genetic transcription activator. This removal may be carried out by procedures described hereinafter, or by other procedures well known in the art.

In one example, fetal fibroblasts carrying the targeting vector are transfected via electroporation with a Cre containing plasmid (e.g., a Cre plasmid that contains a GFP-Cre fusion gene as described by Gagneten et al. (1997) Nucleic Acids Res. 25:3326-3331). This allows for the rapid selection of all clones that contain a Cre protein. In this regard, cells are selected either by FACS sorting or by manual harvesting of green fluorescing cells via micromanipulation. Cells that are green are expected to carry actively transcribed Cre recombinase, which would remove the drug resistance marker. Cells selected for Cre expression are cloned and analyzed for the deletion of the drug resistance marker by PCR analysis. Following such confirmation, such cells are used for the next round of genetic targeting or for cloning.

Ungulates and Donor Cells

Ungulates include members of the orders Perissodactyla and Artiodactyla, such as any member of the genus Bos. Other preferred ungulates include sheep, big-horn sheep, goats, buffalos, antelopes, oxen, horses, donkeys, mule, deer, elk, caribou, water buffalo, camels, llama, alpaca, pigs, and elephants. Preferred cells for gene targeting include differentiated cells such as epithelial cells, neural cells, epidermal cells, keratinocytes, hematopoietic cells, melanocytes, chondrocytes, B-lymphocytes, T-lymphocytes, erythrocytes, macrophages, monocytes, fibroblasts, and muscle cells; and undifferentiated cells such as embryonic cells (e.g., stem cells and embryonic germ cells). In another preferred embodiment, the cell is from the female reproductive system, such as a mammary gland, ovarian cumulus, granulosa, or oviductal cell. Preferred cells also include those from any organ, such as the bladder, brain, esophagus, fallopian tube, heart, intestines, gallbladder, kidney, liver, lung, ovaries, pancreas, prostate, spinal cord, spleen, stomach, testes, thymus, thyroid, trachea, ureter, urethra, and uterus. Preferably, the donor cell, donor nucleus, donor chromatin mass, or reconstituted oocyte is not tetraploid.

Production of Cloned Non-Human Mammals

We have previously disclosed a variety of methods for cloning mammals (e.g., ungulates, such as bovines) that may be used to clone mammals with one or more mutations in genes encoding IgM heavy chain (see, e.g., U.S. Patent Application Publication No. 2002/0046722, herein incorporated by reference, and PCT Publication No. WO 02/051997) and having at least one unrearranged exogenous kappa immunoglobulin locus. In some of these methods, a permeabilized cell is incubated with a reprogramming media (e.g., a cell extract) to allow the addition or removal of factors from the cell, and then the plasma membrane of the permeabilized cell is resealed to enclose the desired factors and restore the membrane integrity of the cell. Some of these methods also involve the condensation of a donor nucleus (e.g., an isolated nucleus or a nucleus within a donor cell) into a chromatin mass to allow the release of nuclear components such as transcription factors that may promote the transcription of genes that are undesirable for the development of the nuclear transplant embryo into a viable offspring. If desired, the steps of any of these methods may be repeated one or more times or different reprogramming methods may be performed sequentially to increase the extent of reprogramming, resulting in greater viability of the cloned fetuses.

Other methods for the production of cloned mammals (e.g., bovines) and cloned transgenic non-human mammals are known in the art, described, for example, in U.S. Patent No. 5,995,577, assigned to University of Massachusetts, and in PCT Publication Nos. WO 95/16670; WO 96/07732; WO 97/0669; and WO 97/0668 (collectively, "the Roslin methods")- The Roslin methods differ from the University of Massachusetts techniques in that they use quiescent, rather than proliferating, donor cells. All of these patents are incorporated by reference herein in their entirety. These techniques are not limited to use for the production of transgenic bovines; the above techniques may be used for embryo cloning of other non-human mammals, including other ungulates. Following embryo cloning, production of desired animals may be affected either by mating the non-human mammals or by secondary gene targeting using the homologous targeting vector previously described.

Methods for Breeding Non-Human Mammals

In preferred embodiments of any of the above methods for generating non-human mammals or mammalian cells, an ungulate of the invention is mated with another mammal to produce an embryo, fetus, or live offspring with two or more genetic modifications. Preferably, one or more cells are isolated from the embryo, fetus, or offspring, and one or more additional genetic modifications are introduced into the isolated cell(s).

κHAC-Containing Non-Human Mammals and Cells

The invention features a transgenic non-human mammal (e.g., a mouse, bovine, ovine, porcine, or caprine) containing an artificial human chromosome that contains both human heavy chain and human kappa light chain, for example, a κΔHAC-I or κΔHAC-11. The mammal may further have a mutation that reduces the expression of functional endogenous antibody (e.g., a mutation that reduces the expression of functional IgM heavy chain or reduces the expression of functional Ig light chain).

In a related aspect, the invention features a non-human mammalian cell containing a κΔHAC-1 or κΔHAC-II. In various embodiments, the cell is selected from fibroblasts, epithelial cells, endothelial cells, neural cells, keratinocytes, hematopoietic cells, melanocytes, chondrocytes, B cells, T cells, macrophages, monocytes, mononuclear cells, cardiac muscle cells, other muscle cells, placental cells, epidermal cells, embryonic cells, and germ cells. The cell may further have a mutation in a nucleic acid encoding an Ig heavy and/or light chain.

The invention also features method of producing antibodies by (a) administering one or more antigens of interest to a non-human mammal containing a κΔHAC-I or κΔHAC-II, wherein the κΔHAC-I or κΔHAC-II undergoes rearrangement resulting in the production of antibody proteins specific for the antigen, and (b) recovering antibodies from the ungulate. Preferably, the mammal is an ungulate, for example, a bovine, ovine, porcine, or caprine, and optionally has a mutation that reduces or eliminates the expression of an endogenous antibody.

In a related aspect, the invention features a method of producing antibodies by recovering xenogenous antibodies from a non-human mammal (e.g., an ungulate) comprising a kΔHAC-I or kΔHAC-II, wherein the kΔHAC-I or kΔHAC-II undergoes rearrangement resulting in the production of antibodies (e.g., fully human antibodies and/or chimeric antibodies). The non-human mammal, which may be, for example, a bovine, ovine, porcine, or caprine, optionally has a mutation that reduces the expression of an endogenous antibody. As discussed herein, the methods of the present invention involve the introduction of mutations into somatic non-human mammalian cells, such as somatic ungulate cells. Suitable somatic cells include cells from embryos, fetuses, calves, or adult animals. Preferred cells for gene targeting include differentiated cells such as fibroblasts, epithelial cells, neural cells, epidermal cells, keratinocytes, hematopoietic cells, melanocytes, chondrocytes, B-lymphocytes, T-lymphocytes, erythrocytes, macrophages, monocytes, placental, and muscle cells. Preferred cells also include those from any organ, such as the bladder, brain, esophagus, fallopian tube, heart, intestines, gallbladder, kidney, liver, lung, ovaries, pancreas, prostate, spinal cord, spleen, stomach, testes, thymus, thyroid, trachea, ureter, urethra, and uterus. Preferably, the donor cell, donor nucleus, donor chromatin mass, or reconstituted oocyte is not tetraploid.

Cells may be derived from any non-human mammal, including an ungulate, rabbit, mouse, rat, or primate. Ungulates include members of the orders Perissodactyla and Artiodactyla, such as any member of the genus Bos. Other preferred ungulates include sheep, big-horn sheep, goats, buffalos, antelopes, oxen, horses, donkeys, mule, deer, elk, caribou, water buffalo, camels, llama, alpaca, pigs, and elephants. Most preferably, the non-human mammal is a bovine (e.g., Bos taurus or Bos indicus) or porcine.

If a cell to be genetically targeted is derived from an embryo or a fetus, the cell may be isolated at any time during the gestation period until the birth of the genetically altered non-human mammal. As discussed above, bovine cells are desirably isolated at between 25 to 90 days of gestation, between 35 to 60 days of gestation, between 35 to 50 days, preferably between 35 to 45 days, more preferably between 38 to 43 days, and most preferably at about 40 days of gestation. Ovine cells are desirably isolated at between 25 to 150 days of gestation, between 30 to 100 days, preferably between 35 to 80 days, more preferably between 35 to 60 days, and most preferably at about 40 days of gestation. Equine cells are desirably isolated at between 25 to 300 days of gestation, between 30 to 100 days, preferably between 35 to 80 days, more preferably between 35 to 60 days, and most preferably at about 40 days of gestation. Porcine cells are desirably isolated at between 25 to 110 days of gestation, between 30 to 90 days, preferably between 30 to 70 days, more preferably between 30 to 50 days, and most preferably at about 35 days of gestation. Caprine cells are desirably isolated at between 25 to 150 days of gestation, between 30 to 100 days, preferably between 35 to 80 days, more preferably between 35 to 60 days, and most preferably at about 40 days of gestation. Primate cells are desirably isolated at between 25 to 150 days of gestation, between 30 to 100 days, preferably between 35 to 80 days, more preferably between 35 to 60 days, and most preferably at about 40 days of gestation. Rodent cells are desirably isolated at between 6 to 18 days of gestation, between 8 to 16 days, preferably between 10 to 16 days, more preferably between 12 to 16 days, and most preferably at about 14 days of gestation.

The recipient cell is preferably an oocyte, a fertilized zygote, or a two-cell embryo, all of which may or may not have been enucleated. Typically, the donor and the recipient cell are derived from the same species. However, there has been success reported in achieving development from embryos reconstructed using donor and recipient cells from different species.

Methods for Producing Serum and/or Milk Which Contain Xenogenous Antibodies

The invention also provides method for producing serum and/or milk that contain xenogenous (e.g. human) polyclonal antibodies using a non-human mammal (e.g., ungulate) of the invention. One such method involves administering one or more antigens of interest to non-human mammal (e.g., ungulate) of the invention having nucleic acid encoding a xenogenous antibody gene locus. The nucleic acid segments in the gene locus undergo rearrangement resulting in the production of antibodies specific for the antigen. Antibodies are recovered from the serum and/or milk of such mammal. The antibodies may be monoclonal or polyclonal and are preferably reactive with an antigen of interest. Preferably, the antibodies are recovered from the serum or milk of the mammal (e.g., bovine).

In a related aspect, the invention provides another method for producing antibodies that involves recovering xenogenous antibodies from a non-human mammal (e.g., ungulate) of the invention having nucleic acid encoding a xenogenous antibody gene locus. The nucleic acid segments in the gene locus undergo rearrangement resulting in the production of xenogenous antibodies. The antibodies may be monoclonal or polyclonal and are preferably reactive with an antigen of interest. Preferably, the antibodies are recovered from the serum or milk of the mammal. Preferably, the mammalian antiserum or milk has polyclonal human immunoglobulins. Preferably, the antiserum or milk is from a bovine, ovine, porcine, or caprine. In another preferred embodiment, the immunoglobulins are directed against a desired antigen. In preferred embodiments, the antiserum is used as intravenous immunoglobulin (IVIG) for the treatment or prevention of disease in humans. In another preferred embodiment, an antigen of interest is administered to the non-human mammal, and immunoglobulins directed against the antigen are produced by the mammal. Preferably, the nucleic acid segments in the xenogenous immunoglobulin gene locus rearrange, and xenogenous antibodies reactive with the antigen of interest are produced. Preferably, the antiserum and/or milk contains at least 2-, 5-, 10-, 20-, or 50-fold more xenogenous antibody than endogenous antibody, or contains no endogenous antibody. If desired, hybridomas and monoclonal antibodies can be produced using xenogenous B cells derived from the above-described transgenic non-human mammal (for example, transgenic bovines). It is also contemplated that xenogenous antibodies (e.g., human antibodies) isolated from the non-human mammals may be subsequently chemically modified so that they are covalently linked to a toxin, therapeutically active compound, enzyme, cytokine, radiolabel, fluorescent label, or affinity tag. If desired, the fluorescent or radiolabel may be used for imaging of the antibody in vitro or in vivo.

Transgenic Non-Human Mammalian Cells

In one aspect, the invention provides a non-human mammalian cell (e.g., bovine or porcine cell) having a mutation (e.g., a mutation after the initial ATC codon, such as a mutation that is within 10, 20, 50, or 100 nucleotides of this codon) in one or both alleles of at least two genes encoding IgM heavy chain. Preferably, the mutations reduce or substantially eliminate the expression of functional IgM protein. In preferred embodiments, expression of functional or total IgM protein is decreased by at least 10, 20, 40, 60, 80, 90, 95, or 100%. The mutations may be hemizygous or homozygous. In some embodiments, the mutations include an insertion of a positive selection marker (e.g., an antibiotic resistance gene) into the nucleic acid. Preferably, the positive selection marker is operably linked to a xenogenous promoter. For non-human mammals or mammalian cells with an antibiotic resistance gene inserted into both alleles of a gene encoding IgM heavy chain, each allele may contain the same or a different antibiotic resistance gene. In a preferred embodiment, a negative selection marker (e.g., DT-A or Tk) is operably linked to a xenogenous promoter and is present in a vector used to disrupt an endogenous allele. The mutation may or may not include the deletion of one or more nucleotides (e.g., contiguous nucleotides) in the gene.

In preferred embodiments of the above aspect, the mammal (e.g., bovine or porcine) or mammalian cell (e.g., bovine or porcine cell) has one or more transgenes and expresses an mRNA or protein (e.g., antibody) encoded by the transgene(s). Preferred mammals contain naturally arranged segments of human chromosomes (e.g., human chromosomal fragments) or artificial chromosomes that comprise artificially engineered human chromosome fragments (i.e., the fragments may be rearranged relative to the human genome). In some embodiments, the xenogenous nucleic acid is contained within a chromosome fragment. The nucleic acid may be integrated into a chromosome of the non-human mammal (e.g., ungulate) or maintained in the mammalian cell independently from the host chromosome. In various embodiments, the nucleic acid is contained in a chromosome fragment, such as a ΔHAC, ΔΔHAC, κΔHAC-I, or κΔHAC-II. In other embodiments, the xenogenous antibody is an antibody from another genus, such as a human antibody. Preferred mammals and mammalian cells have one or more nucleic acids having a xenogenous antibody gene locus (e.g., a nucleic acid encoding all or part of a xenogenous immunoglobulin (Ig) gene that undergoes rearrangement and expresses at least one xenogenous Ig molecule) in one or more B cells. Preferably, the nucleic acid has unrearranged antibody light chain nucleic acid segments in which all of the nucleic acid segments encoding a V gene segment are separated from all of the nucleic acid segments encoding a J gene segment by one or more nucleotides. Other preferred nucleic acids have unrearranged antibody heavy chain nucleic acid segments in which either (i) all of the nucleic acid segments encoding a V gene segment are separated from all of the nucleic acid segments encoding a D gene segment by one or more nucleotides and/or (ii) all of the nucleic acid segments encoding a D gene segment are separated from all of the nucleic acid segments encoding a J gene segment by one or more nucleotides. Other preferred mammals (e.g., ungulates) have one or more nucleic acids encoding all or part of a rearranged xenogenous immunoglobulin gene that expresses at least one xenogenous immunoglobulin.

In other preferred embodiments, the light chain and/or heavy chain of the xenogenous antibody is encoded by a human nucleic acid. In preferred embodiments, the heavy chain is any class of heavy chain, such as mu, gamma, alpha, epsilon, or delta, and the light chain is a lambda or kappa light chain. In other preferred embodiments, the nucleic acid encoding the xenogenous immunoglobulin chain or antibody is in its unrearranged form. In other preferred embodiments, more than one class of xenogenous antibody is produced by the mammal. In various embodiments, more than one different xenogenous Ig or antibody is produced by the mammal. The xenogenous antibody may be a polyclonal or monoclonal antibody.

Preferably, the non-human mammal (e.g., ungulate) also has a mutation in one or both alleles of an endogenous Ig λ chain, an endogenous Ig K chain, an endogenous nucleic acid encoding prion protein, alpha-(l,3)-galactosyltransferase and/or J chain. Preferably, the mutation reduces or eliminates the expression of the endogenous alpha-(l ,3)- galactosyltransferase enzyme, galactosyl(α 1,3 galactose epitope, and/or J chain. Preferably, the mammal produces human IgA or IgM molecules containing human J chain. Preferred mammalian cells (e.g., bovine cells) include somatic cells, such as fetal fibroblasts or B cells. The process of producing a transgenic non-human mammal of the invention involves the mutation (e.g., by homologous recombination) of one or both alleles of at least one or two IgM heavy chain genes (e.g., bovine IgμU and IgμAY genes). Gene mutation may be effected by homologous recombination. In a preferred embodiment, fetal fibroblasts are targeted in vitro using a suitable homologous recombination vector. The use of fetal fibroblasts is preferred over some other somatic cells as these cells are readily propagated and genetically manipulated in tissue culture. However, the use of fetal fibroblasts is not essential to the invention, and other cells may be substituted therefor with equivalent results. Suitable somatic cells include fibroblasts, epithelial cells, endothelial cells, neural cells, keratinocytes, hematopoietic cells, melanocytes, chondrocytes, lymphocytes (B cells and T-cells), macrophages, monocytes, mononuclear cells, cardiac muscle cells, other muscle cells, granulosa cells, cumulus cells, placental cells, and epidermal cells.

Targeted gene mutation requires constructing a DNA construct having regions of homology to the targeted IgM heavy chain allele such that the construct upon integration into an IgM heavy chain allele in the non-human mammalian genome disrupts the expression thereof. Exemplary vectors for carrying out such targeted mutation of bovine IgμU and IgμAY are described in the examples that follow. Methods for constructing vectors that provide for homologous recombination at other targeted sites are well known to those skilled in the art. Moreover, in the present instance, the construction of a suitable vector is within the level of skill in the art, given especially that the sequences of Igμ genes from other mammals (e.g., sheep and goats) are known (see below). In order to facilitate homologous recombination, the vectors used to effect homologous recombination and inactivation of the IgM gene, respectively, comprise portions of DNA that exhibit substantial sequence identity to the targeted mammalian IgM heavy and Ig light chain genes. Preferably, these sequences possess at least 98% sequence identity, more preferably at least 99% sequence identity, and still more preferably are isogenic with the targeted gene loci to facilitate homologous recombination and targeted deletion or inactivation.

Typically, the construct includes a marker gene that allows for selection of desired homologous recombinants, for example, fibroblasts, wherein the IgM heavy chain gene has been disrupted by homologous recombination. Exemplary marker genes include antibiotic resistance markers, drug resistance markers, and green fluorescent protein, among others. One neomycin resistance construct was assembled as follows. A construct designated "pSTneoB" (Katoh et al. (1987) Cell Struct. Fund. 12:575; Japanese Collection of Research Biologicals (JCRB) Deposit No. VE039) was designed to contain a neomycin resistance gene under the control of an SV40 promoter and TK enhancer upstream of the coding region. Downstream of the coding region is an SV40 terminator sequence. The neo cassette was excised from "pSTneoB" as a Xho\ fragment. After the ends of the fragment were converted to blunt ends using standard molecular biology techniques, the blunt ended fragment was cloned into the EcoRV site in the vector, pBS246 (Gibco/Life Technologies). This site is flanked by loxP sites. The new construct, designated "pLoxP-STNeoR", was used to generate the mu knockout DNA construct. The desired fragment of this construct is flanked by loxP sites and No/I sites, which were originally present in the pBS246 cloning vector. The desired No/I fragment, which contains loxP-neo-loxP, was used for replacement of the immunoglobulin mu constant region exons. The SV40 promoter operably linked to the neomycin resistance gene activates the transcription of the neomycin resistance gene, allowing cells in which the desired No/I fragment has replaced the mu constant region exons to be selected based on their resulting antibiotic resistance. After a cell line is obtained in which an IgM heavy chain allele has been effectively disrupted, it is used as a donor cell to produce a cloned non-human mammalian fetus (for example, a cloned bovine fetus) and eventually a fetus or animal wherein one of the IgM heavy alleles is disrupted. Thereafter, a second round of gene targeted mutation can be effected using somatic cells (e.g., fibroblasts) derived from the fetus or animal to produce cells in which a second IgM heavy chain allele is disrupted.

The following examples are meant to illustrate the invention. They are not meant to limit the invention in any way.

EXAMPLES Example 1: Targeting of the IgμU Gene

We have developed a broadly applicable and rapid method for generating multiple gene targeting events. As discussed above, sequential application of a highly efficient targeting system and rejuvenation of cell lines by production of cloned fetuses (FIGURE 1) were employed. We chose to target the IgμU gene, which is transcriptionally silent in fibroblasts. This gene was characterized in a male Holstein fetal fibroblast cell line (#6939) to identify a polymorphic marker DΝA sequence, outside the KO vector sequence, which could be used to distinguish the two alleles (FIGURE 2A; allele A and allele B as indicated). Fetal fibroblasts from cell line #6939 were electroporated with the first KO vector (pBCμΔKOpuro; FIGURE 2A) to produce 446 wells resistant to puromycin. Wells were split on day 14 and half of the cells were used for screening by PCR (primer pairs; puroF2 x puroR2, FIGURE 2A) to identify wells containing correctly targeted cells. Initially, six wells were positive by PCR. To exclude false positive wells, all of the PCR products were subjected to bi-directional sequencing analysis with the puroF2 and puroR2 primers. Two wells (0.45%; #147, #384) were identified as being targeted correctly. Based on polymorphic differences identified by sequence analysis, the KO vector was integrated into allele A in well #384 and into allele B in well #147. The remaining cells from the two wells were used for embryonic cloning to generate fetuses and to rejuvenate the cell lines. Pregnancy rate at 40 days of gestation was 50% (15/30; two embryos per recipient) and at 60 days of gestation, six fetuses were collected and fibroblasts were re-established.

Three of six fetuses (#2184-1 , #2184-2 and #3287) were heterozygous KOs (IgμU^"/+; FIGURE 2B) as confirmed by PCR (primer pairs; puroF2 x puroR2) and sequence analysis.

Non-targeted fetuses likely resulted from non-targeted cells that co-existed with the targeted cells in the wells. Both #2184-1 and #2184-2 were derived from well #384 where the KO vector was integrated into allele A, and fetus #3287 was from well #147 where the KO vector was integrated into allele B. Cloned IgμU^~/+ embryos produced from all three regenerated cell lines were transferred to 153 recipients to produce 13 (8%) healthy IgμU^"/+ calves, confirmed by PCR (FIGURE 2C) and sequence analysis.

All three IgμU^'/+ cell lines (#2184-1 and #2184-2, targeted in allele A; #3287, targeted in allele B) were used for targeting with the second KO vector (pBCμΔKOneo; FIGURE 3A) in which the short homologous arm was replaced with a PCR-derived sequence amplified directly from allele A of the #6939 cell line. In #2184-1 and #2184-2 cell lines, a total of 1,21 1 wells, resistant to G418, were screened by PCR (primer pairs; neoF3 x neoR3; FIGURE 3A) followed by sequence analysis. Five wells were positive and, in two, the vector was integrated into the intact allele B, producing homozygous KO (IgμU^"A) cells, and in three wells the targeting vector in allele A was replaced. In #3287 cell line, 569 wells, resistant to G418, were screened by PCR (primer pairs; neoF3 x neoR3; FIGURE 3A) followed by sequence analysis. Seven wells were positive and, in six, the vector was integrated into the intact allele A producing IgμU^7" cells and, in one well, the targeting vector in allele B was replaced. Overall, the vector had a bias of 3: 1 for allele A and was more efficient for homozygous targeting when used with cell line #3287 (6/569, 1.1% compared to 2/121 1 , 0.17%), as expected. Two IgμU^"'^" wells (#76, #91), derived from cell line #3287, were selected for embryonic cloning to generate fetuses and rejuvenate the cell lines. Overall, the pregnancy rate for IgμU^"7" fetuses at 40 to 50 days of gestation was 45% (40/89). At 45 days of gestation, five fetuses derived from well #76 and 15 fetuses from well #91 were collected and evaluated. All five from well #76 (FIGURE 3B) and three out of the 15 from well #91 were positive as determined by PCR (primer pairs; puroF2 x puroR2 and neoF3 x neoR3). PCR results were confirmed by sequence analyses and negative PCR (primer pairs; bCμf x bCμr; FIGURE 3A) for the wild-type alleles (FIGURE 3B). Confirmation of a functional KO was obtained by generation of 90 day fetuses from regenerated IgμU^"7" fibroblasts and evaluation of IgμU gene expression in spleen cells. The absence of expression was confirmed by RT-PCR (primers pairs; bCμf x bCμr, FIGURE 3C). Cloned embryos were made from five IgμU^"'^" cell lines and were transferred to recipients for development to term. Two calves from this group were born recently and were confirmed to be IgμU^"'^" by PCR (FIGURE 3D) and sequence analysis, verifying that sequential gene targeting and successive rounds of cell rejuvenation are compatible with full term development of healthy calves.

Example 2: Removal of Selection Markers Using the Cre/LoxP System

Sequential targeting requires a strategy for antibiotic selection of a newly integrated targeting vector in a cell line that already contains one or multiple antibiotic selection markers. The simplest approach is to use a different selection marker gene for each targeting event. This approach, however, limits the number of targeting events that may be performed in a cell line. Another approach is to remove the selection markers using a Cre-loxP recombination system, as has been done in murine embryonic stem cells (Abuin and Bradley ( 1996) MoI Cell Biol. 16: 1851 -1856). In our regenerated IgμU targeted fibroblasts, the selection marker genes were not expressed, likely because the IgμU locus is silent in fibroblasts. Although selection marker removal was not necessary for further targeting in our IgμU^"'^" fibroblasts, we evaluated the possibility of removing the selection markers by transfection with a Cre recombinase expression plasmid. Because the intention was for transient expression of Cre recombinase, a closed circular plasmid was used and antibiotic selection was restricted to the first three days of culture. Bovine IgμU^"'^" cell line #4658 was used for transfection and 24 selected wells were evaluated by PCR for excision of the antibiotic selection genes from the targeted alleles (FIGURE 4A). Multiple wells showed evidence of excision of both puro and neo genes, and one was chosen for fetal cloning and regeneration of cell lines. Pregnancy rate at 40 to 50 days of gestation was 35% (21/60).

Five fetuses were recovered and all had both selection markers removed (FIGURE 4B). The Cre recombinase plasmid integrated into the genome in all fetuses, except #1404. These results indicate that Cre-loxP recombination can be used to remove selection markers in somatic cells. Routine use in this system will require improvements to reduce integration frequency of Cre-expression plasmid.

Example 3: Mutation of IgμAY

IgμAY was originally described in WO 05/104835 (herein incorporated by reference). IgμAY KO vectors were generated as follows. To isolate genomic DNA around exon 2 of the IgμAY gene, a DNA probe was amplified by PCR using 5'-TCTCTGGTGACGGCAAT AGC-3' (SEQ ID NO: 1) and 5'-CTTCGTGAGGA AGATGTCGG-3' (SEQ ID NO: 2) (BCμ-f2 and BCμ-r2). Using this probe, a bovine (Holstein) genomic λ phage library derived from #4658 IgμU homozygous KO cell line was screened, and 83 positive λ phage clones were identified. These clones should contain both alleles of intact IgμAY gene and both alleles of targeted IgμU gene. To distinguish intact IgμAY clones from the targeted

IgμU clones, λDNA isolated from each clone was subjected to PCR using primer pair BCμ-f2 and BCμ-r2. In the case of clones containing the targeted IgμU gene, the PCR product cannot be amplified because of presence of the KO cassette integrated at exon 2. On the other hand, the PCR product can be amplified from intact IgμAY locus; clones producing the PCR product should be ones including intact IgμAY gene, but clones not producing the PCR products should be ones including the targeted IgμU gene. Out of 83 λ phage clones, 26 produced the PCR products and these were confirmed to be clones containing intact IgμAY gene by sequence (primer AYU-F2; 5'-GGCTGACTCCCTACCTCCCCTACAC-S ' ; SEQ ID NO: 3). At least other 10 clones that did not produce the PCR products proved to contain the targeted IgμU gene, confirmed by sequence (primer AYU-F2). The foregoing demonstrated that there are at least two Igμ genes in bovine, and that one gene (which we refer to as IgμU) is disrupted in our KO cell line (#4658), but the other gene (IgμAY) is still intact (FIGURE 5).

To distinguish both alleles of IgμAY, we sequenced all the λ phage DNA using primer AYU5' (5'-CGGAGCCCCTGGAGATGAGC-S'; SEQ ID NO: 4). According to this sequencing, we found polymorphic sequences to differentiate the alleles of IgμAY gene, which we named the AY allele and ay allele (FIGURE 6). Out of the 26 clones, 5 clones contained AY allele and 21 clones contained ay allele. To construct AY- or ay-specific KO vectors, we chose #37 clone for AY and #49 for ay. Each of #37 and #49 was analyzed further by restriction mapping. The 9 kilobases of Sal\-Ba?nHl genomic fragment containing all of the CμΛFexons was subcloned into pBluescript II SK(-) in which the Kpn\ site is already replaced with an Srβ site. Then, both the bsr and STOP cassettes were inserted at the BgIW site, which is just located in exon 2 of Cμ. Both bsr and the STOP cassettes were in a sense strand-orientation related to the IgμAY gene. A diphtheria toxin gene (DT-A, Gibco) was then added to the No/I site in the pBluescript II SK(-). DT-A was inserted in forward orientation relative to the bsr gene in the targeting cassette to kill cells in which the targeting cassette was randomly integrated in the genome (pBCμAYKObsr vector; FIGURE 7). Similarly, another KO vector for ay allele containing hyg gene was constructed (pBCμayKOhyg vector; FIGURE 7).

Transfection of IgμU homozygous KO cell lines with IgμAY KO vectors was performed using the following standard electroporation protocol. The medium used to culture the bovine fetal fibroblasts contained 500 ml alpha MEM (Gibco, 12561-049), 50 ml fetal calf serum (Hy-Clone #ABL 13080), 5 ml penicillin-streptomycin (SIGMA), and 1 ml 2-mercaptoethanol (Gibco/BRL #21985-023). On the day prior to transfection, cells were seeded on a Tl 75 tissue culture flask with a confluency of 80-100%, as determined by microscopic examination. On the day of transfection, about 10⁷ bovine fibroblasts cells were trypsinized and washed once with alpha-MEM medium. After resuspension of the cells in 800 μl of alpha-MEM, 30 μg of the Sr/1-digested KO vector (pBCμAYKObsr vector) dissolved in HEPES buffer saline (HBS) containing 1 mM spermidine was added to the cell suspension and mixed well by pipetting. The cell-DΝA suspension was transferred into an electroporation cuvette and electroporated at 550 V and 50 μF. After that, the electroporated cells were plated onto thirty 48-well plates with the alpha-MEM medium supplemented with the serum. After a 48 hour-culture, the medium was replaced with medium containing 10 μg/ml of blasticidine, and the cells were cultured for 2-3 weeks to select blasticidine-resistant cells. After selection, all colonies which reached close to 100% confluency were divided into two replica plates (24-well and 48-well plates): one for genomic DNA extraction, and the other plate for embryo cloning.

Genomic DNA was extracted from the colonies to screen for the desired homologous recombination events by PCR. Genomic DNA was independently extracted from each 24-well using the PUREGENE DNA isolation Kit (Gentra SYSTEMS) according to the manufacturer's protocol. Each genomic DNA sample was resuspended in 20 μl of 10 mM Tris-Cl (pH 8.0) and 1 mM EDTA. Screening by PCR was performed using the following primer pair AYKObsrF2 (5'-GGTAGTGCAGT TTCGAATGGACAAAAGG-3'; SEQ ID NO: 5) and AYKObsrR2 (5'-TCAGGATTTGCAGCACACAGGAGTG-S '; SEQ ID NO: 6). The sequence of one primer is located in the KO vector, and the sequence of the other primer is located just outside of the integrated vector in the targeted endogenous locus. Therefore, the expected PCR product is detected only when the KO vector is integrated into the targeted locus by homologous recombination. The PCR reaction mixtures contained 17.9 μl water, 3 μl of 1 OX LA PCR buffer 11 (Mg²⁺ plus), 4.8 μl of dNTP mixture, 10 pmol of forward primer, 10 pmol of reverse primer, 2 μl of genomic DNA, and 0.3 μl of LA Taq. Forty cycles of PCR were performed by incubating the reaction mixtures under the following conditions: 85 ⁰C for three minutes, 94 ⁰C for one minute, 98 ⁰C for 10 seconds, and 68 ⁰C for 8 minutes. After PCR, the reaction mixtures were analyzed by electrophoresis. Out of 322 screened clones, 22 clones generated the expected PCR products. As a result of sequencing of the PCR products, the KO vector designed to target AY allele was exclusively integrated into the AY allele in all the clones. pBCμayKOhyg vector also was transfected to IgμU homozygous KO cell lines, except that the vector was digested with Sal\ before electroporation. As a result of screening of 453 hygromycin-resistant colonies, 29 clones were identified as a positive by PCR using the following primer pair ayKOhygF2 (5'- TGGTTGGCTTGTATGGAGCAGCAGAC-3'; SEQ ID NO: 7) and ayKOhygR2 (S'-TAGGATATGCAGCACACAGGAGTGTGG-S'; SEQ ID NO: 8). Sequencing of the PCR products demonstrated that the KO vector designed to target ay allele was exclusively integrated into the ay allele. Judging from the above results, it can be said that both the AY and ay KO vectors specifically target each allele in an allele-specific manner and produce correct targeted clones at a frequency of 7-8%.

Cloning was performed as follows. In v//rø-matured oocytes were enucleated at 20 hpm. Bovine IgμU knockout fibroblasts were trypsinized and washed in Ca/Mg Hank's Balanced Salt Solution (HBSS) and permeabilized by incubation of 50,000 - 100,000 cells in 31.25 units Streptolysin O (SLO; Sigma, St. Louis, MO) in 100 μl for 30 minutes in a 37 ⁰C H₂O bath. Cell samples were incubated with propidium iodide and observed by fluorescent microscopy to monitor permeabilization based on uptake of the dye. Permeabilized fibroblasts were washed, pelleted, and incubated in 40 μl of mitotic extract prepared from MDBK cells containing an ATP-generating system (1 mM ATP, 10 mM creatine phosphate, and 25 μg/ml creatine kinase) for 30 minutes in a 37 ⁰C H₂O bath. Cell samples were stained with Hoechst 33342 and observed by fluorescent microscopy to monitor chromatin condensation. At the end of incubation, the reaction mix was diluted with 500 μl cell culture media (alpha MEM with 10% FBS). These cells were pelleted and resuspended in TL HEPES and used for chromatin transfer in enucleated oocytes. Twelve fetuses were determined to be hemizygous igμAY KO fetuses in which the bsrKO vector is integrated into AY allele of the IgμAY gene. Likewise, eleven fetuses were determined to be hemizygous IgμAY KO fetuses in which the hygKO vector is integrated into ay allele of the IgμAY gene. These fetuses were also IgμU^"'^". One of the IgμAY^~/+/IgμU^"/~ cell lines (A227, targeted in AY allele) was used for a second targeting experiment with pBCμayKOhyg. As a result of screening of 197 hygromycin-resistant colonies, 18 clones were identified as a positive by PCR using the primer pair (ayKOhygF2 and ayKOhygR2). Sequencing of the PCR products demonstrated that the KO vector designed to target ay allele was exclusively integrated into the ay allele, producing double homozygous knockout (IgμAY^"/7IgμlT^/") cells. We produced IgμAY^'ΥlgμU^"'^" fetuses using the methods described herein. To examine whether these fetuses were B cell-deficient, we collected IgμAY^"7'/ IgμU^"'^" fetuses at 180 days of gestation. From spleen, total RNA was extracted by using RNeasy Mini kit (QIAGEN). One microliter of total RNA was subjected to first-strand cDNA synthesis (Superscript First-Strand Synthesis System for RT-PCR, Invitrogen), followed by RT-PCR. One RT-PCR reaction was carried out as below. The primer pair used (5'- CCCTCCTCT TTGTGCTGTCA-3' (BL17; SEQ ID NO: 9) and 5'- GTTCAGGCCATCATAGGAGG-3' (mBCμR2; SEQ ID NO: 10)) is compatible both with IgμAY and IgμU amplification. The PCR reaction mixtures contained 32.5 μl water, 5 μl of 1 OX Ex Taq buffer (TAKARA), 8 μl of dNTP mixture, 10 pmol forward primer, 10 pmol of reverse primer, 2 μl of the first-strand cDNA, and 0.5 μl of Ex Taq (TAKARA). Thirty five cycles of PCR were performed by incubating the reaction mixtures at the following conditions: 85 ⁰C for three minutes, 94 ⁰C for one minute, 98 ⁰C for 10 seconds, 62 ⁰C for 30 seconds, and 72 ⁰C for 1 minute. After PCR, the reaction mixtures were analyzed by electrophoresis. There were no positive PCR products in the IgμAY ' /IgμU ^"'^" fetuses. After PCR, the reaction mixtures were analyzed by electrophoresis. No expression of IgμAY or IgμU could be detected. We also performed flow cytometry analysis to detect the presence of IgM heavy chain protein. No such protein could be detected.

Example 4: Comparison Between IgμAY^~'^~, IgμU^~'^~, and IgμAY^"7" IgμU^"'^" Calves We compared three types of knockout calves, IgμAY ^"'^", IgμU^"'^" and

IgμAY ^"'^"IgμU^"'^", in terms of their ability to secrete immunoglobulin protein and to respond to antigen immunization. We first reconfirmed the B cell-deficiency in IgμAY ^"'^"IgμU^"'^" calves as opposed to IgμAY ^"'^" and IgμU^"'^" calves (FIGURE 8A). At day 1 after birth before giving them colostrums and/or intravenous immunoglobulin, we detected secreted IgM protein in sera of IgμAY ^"'^" (4-10.9 μg/ml) and IgμU^"'^" (3.6-7.4 μg/ml) calves as well as in controls (7.5-20.6 μg/ml), while no secreted IgM protein was detected in IgμAY ^"'^"IgμU^"'^" calves by ELISA. This result reveals that IgμU protein can be secreted in the absence of IgμAY, as well as IgμAY protein in the absence of IgμU. On the other hand, IgG protein was detected in the sera of IgμAY ^"'^" (8-10.5 μg/ml), IgμU^~'^~ (6.3-10.6 μg/ml), and even IgμAY ^"'^"IgμU^"'^" (4.2-10.6 μg/ml) calves. The IgG protein detected in the sera of

IgμAY ^"'^"IgμU^"'^" calves should have come from mother through placenta during gestation because IgG transcripts could not be detected in peripheral blood mononuclear cells (PBMC) of IgμAY ^"'^"IgμU^"'^" calves at all, as opposed to IgμAY ^"'^" and IgμU^"'^" calves (FIGURE 8B). To monitor their humoral immune response, we immunized the calves with ovalbumin (OVA) at 3-4 months. Neither IgM nor IgG response specific to OVA was detected in IgμAY^"'^"IgμU^"'^" calves, while IgμU^"'^" calves responded normally with respect to OVA-specific IgM and IgG, similar to control (FIGURE 8C). This is consistent with the observation that the CDR3 region of both IgμAY and IgG transcripts in IgμU^~7~ fetuses were well-diversified. In IgμAY^"7" calves, interestingly, neither OVA-specific IgM nor IgG response was observed in this immunization scheme (FIGURE 8C). The poor IgG response could be explained by the fetal data that CDR3 of IgG transcripts was less-diversified in IgμAY^"'^" fetuses. No IgM response is unexpected because CDR3 of IgμU transcripts in IgμAY^"'^" fetuses seemed to be diversified as well as those in IgμU^"7" fetuses. This may imply that even the CDR3 diversity in IgμU transcripts is not enough for responding to antigens; affinity maturation processing such as somatic hypermutation and/or gene conversion might be totally impaired in IgμAY ^"7". Because of the genomic structure of the IgμU locus, it is possible that somatic hypermutation could not efficiently occur with this locus and/or that the IgμU locus might not be linked with the pseudo V gene cluster that is believed to be necessary for gene conversion. These data demonstrate that IgμU can largely substitute for lack of IgμAY function in driving B cell development up to mmunoglobulin secreting plasma cells, but because of its genomic structure, the affinity maturation process may be impaired and IgG generation is likely regulated by some inefficient mechanism such as trans-class switch or trans-splicing, which could result in immune-irresponsiveness following antigen immunization.

Ovalbumin immunization was performed as follows. IgμAY^"7", IgμU^"7", and IgμAY ^"7" IgμU^"7" calves and control wild-type calves were immunized between 1 and 3 months of age with ovalbumin antigen (Sigma, St. Louis, MO) at 1 mg/dose formulated with Montanide ISA 25 adjuvant (Seppic, Inc., Fairfield, NJ) as water-in-oil emulsion. The calves were immunized three times at three-week intervals (primary immunization followed by first booster after three weeks and second booster after six weeks). Vaccine was administered by intramuscular injection (2 ml dose containing 1 mg/ml ovalbumin plus 1 ml of ISA-25 adjuvant) in the neck region. Serum samples were collected before each immunization (Vl , V2 and V3) and 7 days and 14 days following each immunization for antibody titer analysis. Blood was drawn into serum separator tubes (tiger-top), allowed to clot, and serum was separated by centrifugation. Serum was then aliquoted in 0.5-1 ml volumes and stored frozen until assays were performed. Anti-ovalbumin antibody titers were determined by ovalbumin specific IgM and IgG ELISAs. To determine ovalbumin-specifϊc IgG antibody titers, flat immunoMaxisorp HB bottom 96-well microtiter plates were coated by adding l OOμl per well of 5 μg/ml ovalbumin in phosphate buffered saline (PBS) at pH 7.4 and incubating overnight (12-16 hours) at 4 ⁰C. Ovalbumin-coated plates were then washed three times with 200 μl of PBS/0.05% Tween 20 buffer (PBST). Serum samples were diluted in PBST buffer in four serial dilutions.

Pooled high titer serum with a pre-determined end-point titer was used as the standard and seven 1 :2 serial dilutions from 1 :200 to 1 : 12800 were prepared in PBST buffer for the standard curve. Reciprocal of the end-point dilution was used as titer units and for the pooled standard, the end point titer was determined and assigned as 145,800 Units. A positive control serum with pre-determined titer (24,000 units) and a negative control serum with no titer were also diluted serially in PBST buffer and were used as internal controls to monitor consistency of the assays. The calibrator standard serum dilutions, positive control serum, negative control serum, and test serum samples were added in duplicate wells at 100 μl/well in ovalbumin coated plates and incubated for one hour at 37 ⁰C. Plates were washed three times with PBST buffer to remove unbound proteins and 100 μl of sheep anti-bovine

IgG-HRP labeled antibody (KPL, Gaithersburg, MD) diluted at 1 :80,000 in PBST buffer was added to each well. Plates were incubated for one hour at 37 ⁰C and washed three times with PBST. Finally, the bound anti-ovalbumin antibodies were detected by adding 100 μl/well TMB + H₂O₂ substrate mix (KPL, Gaithersburg, MD) and incubated for 10 minutes at room temperature. The reaction was stopped by adding 100 μl 10% phosphoric acid

(Labchem Inc, Pittsburg, PA) and read in microplate reader (Biotek Instruments, Inc.) at 450 nm wavelength. A four parameter standard curve was generated using seven serial dilution values and serum sample values were calculated by interpolation on the curve by KC-4 software. Average titer value from three or four test dilutions were calculated for each test serum sample.

To determine ovalbumin specific IgM antibody titers, the assay was performed similar to IgG ELISA described above with substitution of sheep anti-bovine IgM-HRP labeled antibody (KPL, Gaithersburg, MD) as secondary antibody in place of anti-bovine IgG-HRP labeled antibody. All the steps were performed as described above for anti-ovalbumin IgG ELISA. A pooled serum with an end-point IgM titer of 64,000 units for the standard curve and a positive control serum with a 16,000 units were used for IgM ELISA. For secondary antibody, sheep anti-bovine IgM-HRP antibody was used at 1 :20,000 dilution in PBST buffer. Data analysis and titer calculations were also performed as described above for IgG ELISA.

Example 5: Construction of KHACII

Human artificial chromosome that contains kappa light chain (KHAC) was constructed using a previously described chromosome-cloning system (Kuroiwa et ai, Nature Biotech. 18: 1086-1090, 2000). The human kappa light chain gene cluster is located at 2pl 1.2 of human chromosome 2. Previously, human chromosome 2 fragment defined by the cosYHZ304 and CD8A loci was not transmitted to mouse offspring (WOOO/10383, Example 96). Here, loxP sequence was integrated at the cosl38 locus. The modified human chromosome 2 fragment (cosl 38-Ig kappa - CD8A) was about 3Mb. Cell hybrids were formed by fusing the DT40 cell clone containing the above hChr2 fragment (hCF2) truncated at the CD8A locus with a DT40 cell clone (denoted "R clone") containing human chromosome 14 minichromosome (named as "SC20" vector). The SC20 vector was generated by inserting a loxP sequence at the RNR2 locus of the S20 fragment. The SC20 fragment (FERM BP-7583, the International Patent Organism Depository, National Institute of Advanced Industrial Science and Technology, Tsukuba Central 6, 1-1 , Higashi 1-Chome Tsukuba-shi, Ibaraki-ken, 305-8566, Japan) is a fragment derived from human chromosome 14 that includes the entire region of the human Ig heavy chain gene (Tomizuka et al., Proc. Natl. Acad. Sci. USA 97:722, 2000). The resulting DT40 cell hybrids contained both hChr fragments. The DT40 hybrids were transfected with a Cre recombinase-expression vector to induce Cre/loxP-mediated chromosomal translocation between hCF2 and the SC20 vector. The stable transfectants were analyzed using nested PCR to confirm the cloning of the hChr2 region, defined by the cosl 38 and CD8A loci, into the loxP-cloning site in the SC20 vector. The PCR-positive cells which were expected to contain KHAC were then isolated by FACS sorting based on the fluorescence of the encoded green fluorescent protein. Fluorescent in situ hybridization (FISH) analysis of the sorted cells was also used to confirm the presence of KHAC, which contains the hChr2 insert. The resulted human artificial chromosome is named as "κΔHAC-1". The size of hChr2 region on KΔHAC-I is about 4 Mb. The germinal transmission of κΔHAC-I was confirmed in mouse. On the other hand, a kappa artificial chromosome that contains chromosome fragment (cosYHZ304- Ig kappa - CD8A) was not transmitted into the offspring.

Similarly, κΔHAC-II was constructed by using another hChr2 fragment truncated at AC104134 locus with a loxP sequence integrated at the same cosl38 locus (FIGURE 9). This hChr2 fragment was translocated onto the SC20 fragment with Cre/loxP-mediated chromosomal translocation, as described above. The size of the hChr2 region on κΔHAC-II is about 2.2 Mb.

Example 6: Production of KHAC Bovines One use of animals having reduced IgM levels is the generation of xenogenous antibodies. A method of producing xenogenous antibody is to produce an animal having one or more human artificial chromosomes expressing antibody heavy chain and/or light chain. To this end, ΔΔHAC (λHAC), κΔHAC-I, or κΔHAC-II was transferred from DT40 cell hybrids to Chinese hamster ovary (CHO) cells using microcell-mediated chromosome transfer (MMCT) (Kuroiwa et al. (2000) Nature Biotech. 18: 1086-1090). The CHO clone containing HAC was cultured in Fl 2 (Gibco) medium supplemented with 10% FBS (Gibco), 1 mg/ml of G418, and 0.2 mg/ml of hygromycin B at 37 ⁰C and 5% CO₂. The HAC clone was expanded into twelve T25 flasks. When the confluency reached 80-90%, colcemid (Sigma) was added to the medium at a final concentration of 0.1 μg/ml. After three days, the medium was exchanged with DMEM (Gibco) supplemented with 10 μg/ml of cytochalacin B (Sigma). The flasks were centrifuged for 60 minutes at 8,000 rpm to collect microcells. The microcells were purified through 8, 5, and 3-μm filters (Costar) and then resuspended in DMEM medium. The microcells were used for fusion with bovine fibroblasts as described below. Bovine fetal fibroblasts (IgμAY^"7", IgμU ^"7", and IgμAV'lgμUT^) were cultured in α-MEM (Gibco) medium supplemented with 10% FBS (Gibco) at 37 ⁰C and 5% CO₂. The fibroblasts were expanded in a Tl 75 flask. When the confluency reached 70-80%, the cells were detached from the flask with 0.05% trypsin. The fibroblast cells were washed twice with DMEM medium and then overlayed on the microcell suspension. After the microcell-fibroblast suspension was centrifuged for five minutes at 1 ,500 rpm, PEG 1500

(Roche) was added to the pellet according to the manufacturer's protocol to enable fusion of the microcells with the bovine fibroblasts. After fusion, the fused cells were plated into six 24-well plates and cultured in α-MEM medium supplemented with 10% FBS for 24 hours. The medium was then exchanged with medium containing 0.8 mg/ml of G418. After growth in the presence of the G418 antibiotic for about two weeks, the G418 resistant, fused cells were selected. These G418-resistant clones were used for cloning, as described previously, to generate cloned fetuses at 180-200 days of gestation and calves.

Example 7: Expression of Human IgM, IgG, and Igλ in HAC/IgμAY^~/7IgμU^'/" Fetuses at 180 Days of Gestation

To examine whether HAC/lgμAY^'^IgμU^"'^" fetuses could express human immunoglobulin such as IgM, IgG, Igλ, and IgK, we collected HAC/IgμAY^"'^" IgμU^"'^" fetuses at 180 days of gestation. Total RNA was extracted from the spleen by using RNeasy Mini kit (QIAGEN). One microliter of total RNA was subjected to first-strand cDNA synthesis (Superscript First-Strand Synthesis System for RT-PCR, Invitrogen), followed by RT-PCR. To detect human IgM expression, RT-PCR reaction was carried out using primer pair; 5'-AGGCCAGCATCTGCGAGGAT-S ' (CH3-F3; SEQ ID NO: 1 1 ) and 5'-GTGGCAGAA TAGCATCG-3 ' (CH4-R2; SEQ ID NO: 12). For human IgG expression, the following primers were used: 5'-CAGGTGCAGCTGGTGCAGTCTGG-S ' (SEQ ID NO: 13), 5'- CAGGTCACCTTGAAGGAGTCTGG-3 ' (SEQ ID NO: 14), 5 '-GAGGTGCAGCTGTGG GCTGG-3' (SEQ ID NO: 15), 5'-CAGGTGCAG CTGCAGGAGTCGGG-3' (SEQ ID NO: 16), 5 '-GAGGTGCAGCTGGTGCAGTC TGG-3' (SEQ ID NO: 17), 5'-CAGGTACAGC GCAGCAGTCAGG-3' (SEQ ID NO: 18), and 5 '-CAGGTGCAGCTGGTGCAGTCTGG-S ' (SEQ ID NO: 13) (VH All Mix) in combination with 5'-CACCACGCTGCTGAGGAGTAG GT-3' (hCglR2; SEQ ID NO: 19). For human Igλ expression, the following primer pair was used: 5'-TCCTCTGAGGAGCTTCAAGC-S ' (hCL-F2; SEQ ID NO: 20) and 5'-AGGGTTTATTGAGTGCAGGG-S ' (hCL-R2; SEQ ID NO: 21). The PCR reaction mixtures contained 32.5 μl water, 5 μl of 1OX Ex Taq buffer (TAKARA), 8 μl of dNTP mixture, 10 pmol forward primer, 10 pmol of reverse primer, 2 μl of the first-strand cDNA, and 0.5 μl of Ex Taq (TAKARA). Thirty-five cycles of PCR were performed by incubating the reaction mixtures at the following conditions: 85 ⁰C for three minutes, 94 ⁰C for one minute, 98 ⁰C for 10 seconds, 60-62 ⁰C for 30 seconds, and 72 ⁰C for 1 minute. After PCR, the reaction mixtures were analyzed by electrophoresis. From a λHAC/lgμAY^IgμLT^7" fetus, human IgM, IgG, and Igλ expression was detected by RT-PCR.

To confirm cell surface expression of human immunoglobulin at the protein level, we also performed flow cytometry analysis as described above. Goat anti-human IgM-FITC (Bethyl Laboratories, Mongomery, TX), goat anti-human IgM-FITC (Serotec Inc., Raleigh, NC), or goat anti-human IgM-PE (Serotec Inc., Raleigh, NC) antibody was used to label the human slgM expressed on the peripheral blood B cells of >180 day HAC fetuses. To detect the light chains expressed on the B cells, goat anti-human lambda-FITC antibody (Bethyl Laboratories) or goat anti-human lambda-PE antibody (Serotec Inc) was used. For dual color analysis, a FITC-labeled anti-human IgM antibody and PE-labeled light chain antibody combination was used. We detected cell populations that were positive with the anti-human IgM and light chain antibodies. Furthermore, we performed sandwich ELISA analysis to detect secreted human IgG in λHAC/IgμAY^'^IgμU^"'^" calves using an affinity purified capture antibody and an appropriate HRP-enzyme labeled detection antibody. Details of the capture antibody and detection antibody for each assay are given in Table 1 below.

Table 1 Assay Standards (Calibrator) Capture Antibody Detection Antibody

Human IgG Human Reference Goat anti-human IgG, affinity Goat anti-human

ELlSA Serum purified or goat anti-human IgG-HRP conjugated

(Bethyl laboratories) IgG-Fc specific, affinity (Bethyl laboratories) purified (Bethyl laboratories)

The capture antibody diluted in coating buffer (0.05 M sodium carbonate, pH 9.6) was coated on the microtiter plates (Nunc ImmunoMaxiSorp ELISA plates) by incubating at room temperature for 1.5 hours. After coating the capture antibody, the plates were washed with phosphate buffered saline (PBS)/Tween 20 buffer 3-5 times using an automated plate washer. Appropriate standards (Calibrators) were added in serial dilutions for quantification using a standard curve. Positive controls and negative controls were included in all assays for QC check. Serum samples from λHAC/IgμAY '/IgμU^"'^" calves were then added into duplicate wells in four serial dilutions and incubated for 1 hour at room temperature. After serum immunoglubulins were captured, the plates were washed again with PBS-Tween buffer 3-5 times using automated plate washer. HRP-enzyme labeled appropriate detection antibody was added to all wells and incubated for 1 hour at room temperature. At the end of incubation, the plates were washed again with PBS/Tween buffer 3-5 times using an automated plate washer. The bound antibodies were detected by adding TMB-Substrate solution (KPL Inc, Gaithersburg, MA) and incubating for 10-20 minutes at room temperature. The reaction was stopped by addition of 10% phosphoric acid. The plates were then read on a microtiter plate reader using KC4 software. Data were analyzed by KC4 software, and values were determined by interpolation on a four- parameter standard curve. In a blood sample collected at 14 days after birth, 7.1 μg/ml of human IgG was detected by ELISA.

Example 8: Development of B cells Expressing Human Immunoglobulin in Bovine Knockout Fetuses

To determine which knockout background is suitable for efficient production of human immunoglobulin, we introduced either KHACII or λHAC into each of the bovine IgM knockout cell lines (IgμAY^~'^~, IgμlT^7" and igμAY^"'^" igμlT^7") by means of MMCT and analyzed them around 180 days of gestation. When peripheral blood mononuclear cells (PBMCs) were stained with anti-human IgM and bovine B200 antibodies, both KHACII/lgμAY^"'^" and λHAC/IgμAY^"^ fetuses showed a significant huIgM⁺/B220⁺ double positive population, whereas neither λHAC/IgμAY^IgμlT^7" nor λHAC/IgμU^"7" fetuses did (FIGURE 1 OA). To observe generation of mature B cells, PBMCs were also stained with anti-human IgM and bovine CD21 antibodies, and our results indicated that most of huIgM⁺ B cells were CD21⁺ mature B cells (FIGURE 10B). For detection of human light chain expression, cells were further stained with anti-human IgK or hlgλ. The KHACII/IgμAY^"7" fetus showed a distinct population of huIgM⁺huIgk⁺ B cells (FIGURE 1 OC). When cells were stained with both anti-human and bovine IgM antibodies, most B cells were huIgM⁺BolgM⁺ double positive (FIGURE 1 OD). In HAC/IgμAY^"'^" fetuses, coexpression with the bovine IgμU-encoding protein appears to be important for human IgM⁺ B cells to develop efficiently. This is an unexpected result because IgμAY^"^ fetuses and calves generated a significant number of bovine B cells, which should have suppressed human IgM⁺ B cell generation. We reasoned that IgμU expression, in the absence of IgμAY, could compete efficiently with human IgM without overwhelming human IgM expression, owing to its much lower expression level than that of IgμAY in the normal setting. We also examined bovine and human IGHG gene expression, particularly to CDR3 diversity. We amplified bovine and human IGHG transcripts by RT-PCR, followed by direct sequencing. The CDR3 region of human IGHG transcripts was well diversified whereas that of bovine IGHG was less diversified and rather uniform (FIGURE 10E). This indicates that, as we observed above, IgμAY^"'^" bovines do not generate well-diversified IgG.

Example 9: Production of Kappa HAC Bovines

Bovine fetal fibroblasts (IgμAY^*/'IgμU^"/~) were cultured in α-MEM (Gibco) medium supplemented with 10% FBS (Gibco) at 37 ⁰C and 5% CO₂. The fibroblasts were expanded in a Tl 75 flask. When the confluency reached 70-80%, the cells were detached from the flask with 0.05% trypsin. The fibroblast cells were washed twice with DMEM medium and then overlayed on the CHO-microcell suspension obtained above (containing a human artificial chromosome, e.g., κΔHAC-I, κΔHAC-II, or λHAC). After the microcell-fibroblast suspension was centrifuged for five minutes at 1,500 rpm, PEGl 500 (Roche) was added to the pellet according to the manufacturer's protocol to enable fusion of the microcells with the bovine fibroblasts. After fusion, the fused cells were plated into six 24-well plates and cultured in α-MEM medium supplemented with 10% FBS for 24 hours. The medium was then exchanged with medium containing 0.8 mg/ml of G418. After growth in the presence of the G418 antibiotic for about two weeks, the G418-resistant, fused cells were selected. These G418-resistant clones were used for cloning, as described previously, to generate cloned fetuses at 180-200 days of gestation and calves.

Flow cytometry analysis was performed in llial Payer's patch (IPP) in order to evaluate B cell development in the KHAC IgμAYr' lgμU^"'^" and KHAC IgμU^"A calves. KHAC IgμAY^"/"IgμU^"/" bovines showed much better B cell development (hIgM⁺/CD21⁺ mature B cells) than KHAC IgμU^"7", which was comparable to control animals (FIGURE 1 1). The data suggests that bovine B cell development through hlgM in the complete absence of blgM efficiently occurs in IPP for proliferation and expansion, which is consistent with the high level of hlgG secretion observed (described below). Example 10: Fully Human IgG Production in κΔHAC-I Cattle

We examined the expression level of light chain in the serum of transgenic animals. Lambda chain transgenic mice expressed human light chain at higher levels than kappa chain transgenic mice, whereas kappa chain transgenic bovines produced 10-100 times greater levels than lambda chain bovines. We also examined the production of fully human immunoglobulin in serum of transgenic non-human mammals. Transgenic bovines having human heavy chain genes and human lambda light chain genes produced 0.7 μg/ml - 37.8 μg/ml fully human immunoglobulin in their serum a couple of months after birth. On the other hand, transgenic bovines having human heavy chain genes and human kappa light chain genes (κΔHAC-I transgenic bovine) produced 96.4 μg/ml - 293.3 μg/ml fully human immunoglobulin in their serum at day 84 after birth (Table 2). Table 2. Expression of Human Immunoglobulin in Transgenic Animals

1) Tomizuka et al., Proc. Natl. Acad. ScL U.S.A. (2000), Volume 97, Number 2, page 722-727

2) Kuroiwa et al., Nature Biotechnology (2000), Volume 18, Number 10, page 1086- 1090

FIGURE 12 indicates a more rapid increase in human immunoglobulin levels in kappa HAC bovines than in lambda HAC bovines. This data shows that a kappa HAC bovine will be immunized earlier than a lambda HAC bovine.

Table 3 indicates heavy chain subclass expression in lambda HAC bovines and kappa HAC bovines which was determined by hlgG subclass specific ELISA. The kappa HAC bovines (κΔHAC-I) showed a hIgGl>hIgG2>hIgG3>hIgG4 distribution ratio. A lambda HAC bovine showed a different distribution ratio, with an increased hIgG2:hIgGl ratio in serum. Table 3 Heavy Chain Subclass Expressed in Transgenic Bovines

Example 11: Characterization of Ag-specifϊc hlgG Produced in KHAC IgμAY^"/"IgμU^"/" Bovines

At the age of 1 12 days, when the total hlgG and fully hIgG/hκ concentration was measured at 1.5 g/L and 0.51 g/L, respectively, the KHAC IgμAY^"'^"IgμU^"'^" calf was immunized with Anthrax PA antigen (Brey, Adv. Drug Deliv. Rev. 57: 1266-1292, 2005) to examine the antigen-specific hlgG humoral immune response (Figure 12A). In the experiments, both KHAC IgμAY^"'^"IgμU^"'^" and KHAC IgμU^"'^" control calves were immunized with Anthrax recombinant protective antigen (rPA) at 2 mg/dose formulated with Montanide ISA 206 adjuvant (Seppic) as a water-in-oil emulsion. The calves were immunized four times at 4-week intervals. Vaccine was administered by intramuscular injection (2 mL per dose containing 2 mg/mL PA plus 1 mL of ISA-206 adjuvant) in the neck region. Serum samples were collected before each immunization for antibody titer analysis. Blood was drawn into serum separator tubes, allowed to clot, and serum was separated by centrifugation. Serum was then aliquoted in 0.5-1 mL volumes and stored frozen until assays were performed. Anti-PA antibody titers were determined by PA-specific ELISA.

The data indicate that at the second vaccination with PA (V2), the KHAC IgμAY^"'^"IgμU^~'^" calf began to show PA-specific hlgG response and reached higher titer at the fourth vaccination (V4), which was comparable with PA-specific blgG titer in wild-type bovine pooled hyperimmune sera (V 16) and higher than blgG titer in the control KHAC IgμU^"'^" calf and hlgG titer in the human reference serum AVR801 (FIGURE 13A). There was no detectable PA-specific hlgG titer in the KHAC IgμU^"'^' control calf (Table 4). The data demonstrate that the KHAC IgμAY ' lgμU^''^" genotype is useful in generating high titer, antigen-specific hlgG upon antigen immunization. Table 4: Serum Anti-Anthrax PA Antibody Titers

No. of Serum IgG PA ELISA titer PA ELISA titer

Vaccinations (QΛ) (units/ml) (uπits/mg)

Human pooled Immune serum (AVR 801 ) 4 5.3 27,000 5,094

Wild type bovine pooled hyperimmune serum 16 -25.0 3,678,260 147,130 bovine control serum" 4 23.0 600.000 26.087 κHAC//GHΛf-K3«W/. t^ txwlne serumP 4 2.6 1,461.173 561.989

³IiIgG from oontrol calf 1495 "Total human IgG from calf 468

For characterization of the PA-specific hlgG produced in bovines, IgG was purified from the plasma of the KHAC IgμAY^"/"IgμU^*/'calf, collected from V4. One liter of plasma was first fractionated with caprylic acid to remove bovine plasma proteins from the immunoglobulin fraction, followed by anti-hlgG Fc VHH (camelid heavy chain antibody) sepharose column to capture hlgG-containing Ig molecules, and then by anti-blgG Fc VHH sepharose column to remove any blgG-containing Ig molecules derived from bovine IVIG administration. Finally, the IgG sample from the anti-blgG Fc VHH column flow-through was separated by anti-hlgκ VHH sepharose column to capture fully hIgG/hκ, of which yield was ~ 250 mg (~ 10% recovery of the total IgG in plasma). High performance liquid chromatography-size exclusion chromatography (HPLC-SEC), SDS-PAGE, and Western blotting using anti-bovine IgG (H + L) and anti-bovine λ light chain polyclonal antibodies was performed to check the purity of the fully hIgG/hκ fraction. There was neither bovine IgG heavy nor light chain bands detected in the fully hlgG/hκ fraction (FIGURE 13B). On the contrary, positive bands for hlgG heavy and hlgκ light chains were clearly detected by anti-human IgG (H + L) polyclonal antibodies and anti-human Igκ polyclonal antibodies, respectively (FIGURE 13C). These data indicate that the fully hIgG/hκ fraction indeed contains both hlgG heavy and hlgκ light chains and that the amount of bovine Ig molecules left in the fraction is very little if any. The human heavy chain and bovine light chain chimeric hlgG that was the flow-through fraction of the anti-hIgG/hκ fraction VHH column was also analyzed. The chimeric hlgG fraction was positive for hlgG heavy chain and bovine λ light chains, but negative for human Igκ light and bovine IgG heavy chains (FIGURES 13B-13C) and therefore, the chimeric hlgG fraction appears to comprise hlgG/bλ (or bκ)_ Example 12: Glycosylation Analysis of Antigen-Specific hlgG Produced in KHAC IgμAY ' lgμU^"'^" Bovines

It is known that IgG heavy chain is usually glycosylated at its Fab and Fc region in a species-specific manner (Raju et al., Glycobiology 10:477-486, 2000). The N-linked glycosylation in the fully IgG/hκ and chimeric hlgG fractions was analyzed (Figure 14; Table 5). N-linked oligosaccharides of the tested samples were released enzymatically and then derivatized with 2-aminobenzoic acid. The derivatized oligosaccharide mixture was purified using a solid phase extraction cartridge. The purified sample was analyzed by capillary gel electrophoresis with helium-cadmium laser induced fluorescence detection (CE-LIF). Overall, the glycosylation pattern of the fully hIgG/hκ and chimeric hlgG fractions from the KHAC IgμAY ' lgμU^"'^" calf was similar to that of polyclonal WgG control. When compared with monoclonal hlgG produced in CHO cells and polyclonal hlgG control from human donors, the glycosylation profile of hlgG produced in the KHAC IgμAY ' lgμU^"'^" calf appears to be more similar to that of polyclonal hlgG control. One minor difference observed in the KHAC IgμAY 'TgμU^''^" calf-derived hlgG from the control polyclonal hlgG, is the LP peak, which is thought to contain fucose-less sugar chain. However, the LP peak is similarly minor even in the human control. The Sl and S2 peaks contained sugar chain to which sialic acid is added. Sialic acids attached to N-linked oligosaccharides were released by acid hydrolysis and then derivatized with DMB. The derivatized sialic acids (N-acetylneuraminic acid, NANA and N-glycolylneuraminic acid, NGNA) were analyzed by reverse phase-HPLC with fluorescent detector (Table 6). Total content of sialic acid is similar between the KHAC IgμAY^"'^"IgμU^"'^" calf-derived hlgG and the control polyclonal IgG. However, the ratio of NANA/NGNA is totally different: the KHAC IgμAY ' lgμU^"'^" calf-derived hlgG has predominantly NGNA, as is seen in the control blgG (Raju et al., supra), whereas the control polyclonal hlgG has NANA exclusively. With respect to branched sugar chains (G0-G2), the contents of galactose (Gal/N) and N-acetylglucosamine (GInAc) (GO-GN and GO-G l ) are similar between the KHAC IgμAY ' lgμU^"'^" calf-derived hlgG and the control polyclonal hlgG. Table 5: N-Linked Oligosaccharide Structure and Its Distribution

Samples Oligosaccharide structure and distribution ι (%) GaIW SAW

82 81 LP GO GO-GN G1 GV G1-GN G2 blgG" 6.5 12.1 2,9 13.2 3.8 10.5 23.2 5.0 22.0 1.33 0.25 hlgGflW 9.1 14.0 0.6 24.2 4.1 15.4 15.4 0 16.7 1.06 0.32 clgG' 5.6 9.8 0.7 28.9 5.6 18.8 13.1 0 16.7 0.84 0.21

Controf 3.5 17.4 4.0 22.9 4.9 20.8 9.0 0 17.6 1.31 0.24 ^aBovine IgG from wild type cattle

"Fully Human IgG from the KHAC IgμAY ' lgμU^"'^" calf

'Chimeric human IgG from the KHAC IgμAV' lgμU^"7" calf dControl human IgG from donors eGalactose residue per N-glycan fSialic residue per N-glycan

Table 6: Sialic Acid Content (nmol/mg protein)

NANA NGNA NANA + NGNA NGNA (%) blgG* 0.81 4.29 5.10 84.1 hlgG/hκ" 0.28 6.05 6.33 95.6

Chimeric WgG' 0.22 4.76 4.98 95.5

Controf¹ 4.83 0 4.63 0

³WgG from wild type cattle "FuJIy hlg<3/rtκ from call 468 'Chimeric htgC from calf 4δβ ""Control tilgG from human donors.

Example 13: In Vitro Toxin Neutralization Assays and In Vivo Mouse Protection

Assays using Antrax PA-Specific hlgG Produced in KHAC IgμAY^{" "}IgμU^{" "} Bovines

The purified fully hIgG/hκ and chimeric hlgG fraction containing the binding activity against Anthrax PA antigen at V4 were subjected to in vitro toxin neutralization assay (TNA), as described in Hering et al. (Biologicals 1 : 17-27, 2004) and Pittman et al. (Vaccine 20: 1412-1420, 2002) (Table 7). The neutralizing activity of hlgG produced in the KHAC IgμAY ' lgμU^"7" calf, both purified hlgG/htc and chimeric hlgG, was comparable with the neutralizing activity of hyperimmunized wild-type bovine IgG, and much higher than the activity of a human reference. Table 7: Toxin Neutralization Assay

No. of IgG cone TNA TNA (EC₅O)" vaccinations (gfl) (ED₅₀)³ <μg>

Wild-type bovine pooled 16 10.4 10,090 1.0 hyperimmune purified blgG

Calf 468-derived purified 4 17.7 12377 1.4 total hlgG

Calf 468-derived purified 4 18.4 13,143 IA chimeric hlgG

Calf 468-derived purified 4 21.1 11,890 1.8 fully human hlgG/hvc-chain

Human pooled immune 4 5.3 1 11 57.0 serum (AVR 801)

*TNA ED₂₀ is the dilution of the antibody solution or serum that neutralizes 50% of total cell cytotoxicity by tie anthrax tøtin. ⁵TNA EC₅₀ is the amount (^>) of antibody required to neutralize 50% of total cell cytotoxicity by the anthβx toxin.

Anthrax PA-challenge mouse protection assays were also performed to determine the protective activity of the bovine-derived hlgG in vivo (FIGURE 15). Mice were challenged with 1 x 10⁶ Anthrax Sterne strain spores. The mice were given either 90 mg/kg of total hlgG produced in the KHAC IgμAY ' lgμU^''^" calf at Vl that contained no activity, 90 mg/kg of fully hIgG/hκ, 70 mg/kg of chimeric hlgG, 70 mg/kg of total hlgG from the KHAC IgμAY ' lgμU^"'^" calf at V4, or 50 mg/kg hyperimmunized pooled wild-type bovine IgG at V 16. IgG doses were standardized to contain equivalent toxin neutralizing activity in the purified fraction. All ten mice receiving hlgG at Vl (negative control) died, whereas both fully hIgG/hκ and chimeric hlgG completely protected all the mice. The hyperimmune pooled blgG resulted in one death out often mice. This complete protection activity was also observed with 22.5 mg/kg and 17.5 mg/kg of fully hIgG/hκ and chimeric hlgG, respectively, from the KHAC IgμAY^"/'IgμU^"/" calf at V4. This data suggests that hlgG produced in the KHAC IgμAY^"/~IgμU^"/' calf (both fully hIgG/hκ and chimeric hlgG) was fully functional and effective in neutralizing the toxin activity in vitro and in vivo. Example 14: Alternative Cloning Strategy for Production of KHAC IgμAY^~/~IgμU^~/~ Bovines

The present study is the first to report the production of an animal that underwent such a high number of recloning and genetic modifications: the KHAC IgμAY^"/*IgμU^"/* calf went through five rounds of genetic modification and seven rounds of recloning (generation 7, G7). In order to produce more calves, KHAC IgμAY ' lgμU^"'^* cell lines (G5) originated from a different primary bovine fibroblast cell line were generated. Two more KHAC IgμAY^~/TgμU^"/~ calves have been generated from these cell lines, one of which has already produced hlgG (>l g/L).

To improve calving efficiency, an IgμAY^"'^"IgμU^"/+ bull and a IgμU^"'^* heifer were generated in order to breed IgμAY^^'IgμU^"7" fetuses (they should be regarded as GO), from which fibroblast cell lines may be used to generate Gl IgμAY^'/"IgμU^"/' cell lines by means of only one more round of gene targeting on the remaining allele of the KHAC IgμAY gene, followed by HAC transfer. These breeding-derived IgμAY^"/TgμU^~/~ cell lines (G2) have produced live calves much more efficiently (5.6%). The resulting KHAC IgμAY^'/"IgμU^'Λ calves also produced >1 g/L of hlgG. From these results, it is clear that the genotype KHAC IgμAY^"/TgμU^*/" appears to be consistently useful for production of a large quantity of hlgG and that lowering G number by breeding is effective in producing more live calves.

Example 15: Production of Transgenic Bovine with κΔHAC-II

We transferred the κΔHAC-II fragment into bovine fibroblasts (IgμAY^'/"IgμU^"Λ) and generated a cell line for κΔHAC-II. The clone was used for cloning, as well as κΔHAC-1, to generate cloned calf fetuses (#1699).

The transgenic bovine (#1699) which had the κΔHAC-II fragment showed expression of human immunoglobulin in its serum.

Example 16: Human IgG Production in κΔHAC-I Pigs

To produce human immunoglobulin in porcines, the CHO clone containing KHAC (KΔHAC-I) was cultured in Fl 2 (Gibco) medium supplemented with 10% FBS (Gibco) and 0.6 mg/ml of G418 at 37 ⁰C and 5% CO₂. The clone was expanded into twelve T25 flasks. When the confluency reached 80-90%, colcemid (Sigma) was added to the medium at a final concentration of 0.1 μg/ml. After three days, the medium was exchanged with DMEM (Gibco) supplemented with 10 μg/ml of cytochalacin B (Sigma). The flasks were centrifuged for 60 minutes at 8,000 rpm to collect microcells. The microcells were purified through 8, 5, and 3-μm filters (Costar) and then resuspended in DMEM medium. The microcells were used for fusion with porcine fibroblasts as described below.

Primary porcine fetal fibroblasts were cultured in α-MEM (Gibco) medium supplemented with 10% FBS (Gibco) at 37 ⁰C and 5% CO₂. The fibroblasts were expanded in a T175 flask. When the confluency reached 70-80%, the cells were detached from the flask with 0.05% trypsin. The fibroblast cells were washed twice with DMEM medium and then overlayed on the microcell suspension obtained above. After the microcell-fibroblast suspension was centrifuged for five minutes at 1,500 rpm, PEG 1500 (Roche) was added to the pellet according to the manufacturer's protocol to enable fusion of the microcells with the porcine fibroblasts. After fusion, the fused cells were plated into six 24-well plates and cultured in α-MEM medium supplemented with 10% FBS for 24 hours. The medium was then exchanged with medium containing 0.2 mg/ml of G418. After growth in the presence of the G418 antibiotic for about two weeks, the G418-resistant, fused cells were selected. These G418-resistant clones were used for nuclear transfer to generate cloned piglets. Generation of cloned piglets was done as previously described (Sullivan et al., Biol.

Reprod. 70, 146-153, 2004) except that multiple embryos (up to 100 embryos) were transferred to a single porcine recipient. We obtained nine cloned piglets from the above G418-resistant colonies. Retention of the κΔHAC-I was examined by genomic PCR as described previously (Tomizuka et al., Nature Genet. 16:133-143, 1997) and eight of them proved to contain the HAC. Furthermore, HAC retention was also confirmed at one week after birth by FISH analysis, as described previously (Kuroiwa et al., Nature Biotech. 20, 889-894, 2002), in peripheral blood of five piglets that were positive with the HAC by the above genomic PCR. These results demonstrate that the κΔHAC-1 can be stably maintained during porcine development (FIGURE 16). In order to investigate whether hlg loci (human IgH and IgK loci) are rearranged and expressed in the piglets, total RNA was extracted from peripheral blood of five HAC- containing piglets. RT-PCR was carried out to detect V(D)J-rearranged human Igμ (heavy chain) and Igκ (light chain) transcripts and then the PCR products were subjected to either direct sequencing (FIGURES 17 A and 18A) to determine CDR3 diversity or subcloning followed by sequencing (FIGURES 17B and 18B) to see if productive Ig protein is encoded. The RT-PCR reactions were done as follows. To amplify VDJ-rearranged human Igμ transcripts, VH(AIl)MIX (equal molar mixture of VHl ; S'-CAGGTGCAGCTGGTGCAGTC TGG-3', VH2 (SEQ ID NO: 13); 5'-CAGGTCACCTTGAAGGAGTCTGG^', VH3 (SEQ ID NO: 14); 5'-GAGGTGCAGCTGGTGGAGTCTGG^', VH4 (SEQ ID NO: 15); 5'-CAGGT GCAGCTGCAGGAGTCGGG-3', VH5 (SEQ ID NO: 22); 5'-GAGGTGCAGCTGGTGC AGTCTGG-3', VH6 (SEQ ID NO: 17); and 5'-CAGGTACAGCTGCAGCAGTCAGG-3', VH7 (SEQ ID NO: 18)) as a forward primer and Cμ-2 (5'-AGGCAGCCAACGGCCA CGCT-3'; SEQ ID NO: 23) as a reverse primer were used in 40 cycles of 98 ⁰C for 10 seconds, 59 ⁰C for 30 seconds, and 72 ⁰C for 1 minute. For VJ-rearranged human Igκ transcripts, VK(AII)MIX (5'-GACATCCAGATGACCCAGTCTCC^' (SEQ ID NO: 24), 5'-GATATTGTGATGACTCAGTCTCC-3' (SEQ ID NO: 25) and 5'-GAAATTGTGTTGA CGCAGTCTCC-3' (SEQ ID NO: 26)) as a forward primer and CK (5'-CCAAGCTTCA GAGGCAGTTCCAGATTTC-3'; SEQ ID NO: 27) were used in 40 cycles of 98 ⁰C for 10 seconds, 61 ⁰C for 30 seconds, and 72 ⁰C for 1 minute.

These data demonstrate that both human Igμ and IgK loci on the HAC can undergo diversified V(D)J-rearrangement to encode productive immunoglobulin protein in porcine.

Example 17: Production of Transgenic Mice with κΔHAC-I

Microcells were produced from a hundred million cells of a CHO clone containing KΔHAC (KΔHAC-I) (described above), according to a method previously described in Tomizuka et al. {Nature Genet. 16, 133, 1997) and were suspended in 5 ml of DMEM medium. The microcells were collected with a centrifugation. Ten million mouse ES cells (TTF2) were treated with trypsin to mediate release from the culture dish, washed three times with DMEM medium, and suspended in 5 ml of DMEM medium. After mixing the microcells with mouse ES cells, the cells were centrifuged at 1250 rpm for 10 minutes. The pellet was suspended in 0.5 ml of PEG solution (5 g of polyethylene glycol (WAKO

Pure Chemicals, Japan) and 1 ml of DMSO (Sigma, U.S.A.) in 6 ml of DMEM medium) and mixed well for about 90 seconds. DMEM medium (10 mL) was added to the solution and centrifuged at 1250 rpm for 10 minutes. The pellet was suspended in 30 ml of ES medium and cultured in three plates (Corning, U.S.A., 100 mm diameter) which were sheeted with feeder cells. After 24 hours, the medium was changed to a medium with 300 microgram/ml of G418.

After a week, 32 G418-positive clones were obtained. The positive clones were confirmed using PCR check with both C kappa primer and VH3 primer. The clones were also analyzed with a probe of human COTl DNA for FISH according to a method described in Tomizuka et al. {Nature Genet. 16, 133, 1997). Kappa HAC (κΔHAC-I) in mouse ES TT2F clones was confirmed with specific COTl probes.

Chimeric mice which contained κΔHAC-I were produced with the mouse ES cells according to the method of Tomizuka et al. {Nature Genet. 16, 133, 1997). An eight cell embryo produced from a MCH (ICR) mouse (white, Nippon Crea KK, Japan) or a transgenic mouse having a heavy chain gene knockout of its endogenous immunoglobulin (Tomizuka et al., Proc. Natl. Acad. Sci. USA, 97, 722-727, 2000) were used as hosts. Two hundred-twenty embryos with wild-type TT2F and κΔHAC-I TT2F were injected into female mice and 16 chimeric mice were obtained. Four mice were almost 100% chimeric judged by the color of their hair. The chimeric mice retained κΔHAC-I , as determined by PCR analysis of their tail cells with specific primers. This result indicates that ES cells which retain κΔHAC-I are pluripotent and can produce chimeric animals.

The female chimeric mice were mated with MCH (ICR) mice (white, Nippon Crea KK, Japan). Three baby mice retained κΔHAC-I, as confirmed by PCR analysis. The serum of these mice were analyzed with ELISA (Tomizuka et al, Nature Genet., 15, 133, 1997, Proc Natl. Acad. Sci. USA. 97, 722-727, 2000). The expression of both human heavy chain polypeptide and human kappa light chain polypeptide was detected in all three transgenic mice. This indicated that κΔHAC-1 human artificial chromosome is germinally transmitted, and the transgenic mice stably produced both human heavy chain and human light chain. Other Embodiments

All publications and patents cited in this specification are incorporated herein by reference as if each individual publication or patent were specifically and individually indicated to be incorporated by reference. Although the foregoing invention has been described in some detail by way of illustration and example for purposes of clarity of understanding, it will be readily apparent to those of ordinary skill in the art in light of the teachings of this invention that certain changes and modifications may be made thereto without departing from the spirit or scope of the appended claims. What is claimed is:

Claims

Claims

1. A transgenic non-human mammal whose genome comprises a nucleic acid encoding all or part of a xenogenous kappa (K) immunoglobulin gene which undergoes rearrangement and expresses more than one xenogenous K immunoglobulin.

2. The mammal of claim 1, whose genome further comprises mutations in two genes encoding IgM heavy chain.

3. The mammal of claim 2, wherein one of said genes is IgμAY.

4. The mammal of claim 2, wherein at least one mutation is a hemizygous mutation.

5. The mammal of claim 4, wherein both mutations are hemizygous mutations.

6. The mammal of claim 2, wherein at least one mutation is a homozygous mutation.

7. The mammal of claim 6, wherein both mutations are homozygous mutations.

8. The mammal of claim 2, wherein at least one mutation is by the insertion of an exogenous sequence.

9. The mammal of claim 2, wherein said mammal produces less than 10% of endogenous IgM heavy chain, relative to a control mammal.

10. The mammal of claim 1 or 2, wherein said mammal is an ungulate.

1 1. The mammal of claim 10, wherein said ungulate is a bovine, ovine, porcine, or caprine.

12. The mammal of claim 1 1 , wherein said ungulate is a bovine.

13. The mammal of claim 1 1 , wherein said ungulate is a porcine.

14. The mammal of claim 1 or 2, wherein said xenogenous K immunoglobulin is a human K immunoglobulin.

15. The mammal of claim 1 or 2, wherein said nucleic acid is contained within a chromosome fragment.

16. The mammal of claim 15, wherein said chromosome fragment is a K human artificial chromosome (KHAC).

17. The mammal of claim 16, wherein said KHAC is κΔHAC-I or κΔHAC-II.

18. The mammal of claim 1 or 2, further comprising a mutation in one or both alleles of the endogenous K immunoglobulin gene or a mutation in one or both alleles of the endogenous λ immunoglobulin gene.

19. The mammal of claim 18 comprising a homozygous mutation in the endogenous K immunoglobulin gene.

20. The mammal of claim 18 comprising a homozygous mutation in the endogenous λ immunoglobulin gene.

21. The mammal of claim 1 , further comprising a nucleic acid encoding all or part of a xenogenous heavy immunoglobulin gene which undergoes rearrangement and expresses more than one xenogenous heavy immunoglobulin.

22. The mammal of claim 21 , wherein said xenogenous heavy immunoglobulin is human heavy immunoglobulin.

23. The mammal of claim 21, wherein said nucleic acid encoding all or part of xenogenous heavy immunoglobulin gene is a human artificial chromosome (HAC).

24. The mammal of claim 23, wherein said HAC is ΔHAC, ΔΔHAC, κΔHAC-I, or κΔHAC-11.

25. The mammal of claim 21 , wherein said mammal is an ungulate.

26. The mammal of claim 25, wherein said ungulate is a bovine, ovine, porcine, or caprine.

27. The mammal of claim 26, wherein said ungulate is a bovine.

28. The mammal of claim 26, wherein said ungulate is a porcine.

29. A transgenic non-human mammalian cell whose genome comprises a nucleic acid encoding all or part of a xenogenous kappa (K) immunoglobulin gene is capable of undergoing rearrangement and expressing a xenogenous K immunoglobulin.

30. The cell of claim 29, whose genome further comprises a mutation of two genes encoding IgM heavy chain.

31. The cell of claim 29, wherein said cell is an ungulate cell.

32. The cell of claim 31 , wherein said ungulate cell is a bovine, ovine, porcine, or caprine cell.

33. The cell of claim 29, wherein said cell is selected from fibroblasts, epithelial cells, endothelial cells, neural cells, keratinocytes, hematopoietic cells, melanocytes, chondrocytes, B-cells, T-cells, macrophages, monocytes, mononuclear cells, cardiac muscle cells, other muscle cells, placental cells, epidermal cells, embryonic cells, and germ cells.

34. The cell of claim 30, wherein said gene is IgμAY.

35. The cell of claim 30, wherein at least one mutation is a hemizygous mutation.

36. The cell of claim 35, wherein both mutations are hemizygous mutations.

37. The cell of claim 30, wherein at least one mutation is a homozygous mutation.

38. The cell of claim 37, wherein both mutations are homozygous mutations.

39. The cell of claim 30, wherein at least one mutation is by the insertion of an exogenous sequence.

40. The cell of claim 30, wherein said cell produces less than 10% of endogenous IgM heavy chain, relative to a control cell.

41. The cell of claim 29, wherein said xenogenous K immunoglobulin is a human K immunoglobulin.

42. The cell of claim 29, wherein said nucleic acid is contained within a chromosome fragment.

43. The cell of claim 42, wherein said chromosome fragment is a K human artificial chromosome (KHAC).

44. The cell of claim 43, wherein said KHAC is κΔHAC-I or κΔHAC-II.

45. The cell of claim 29, further comprising a mutation in one or both alleles of the endogenous K immunoglobulin gene, or one or both alleles of the endogenous λ immunoglobulin gene.

46. The cell of claim 45 comprising a homozygous mutation in the endogenous K immunoglobulin gene.

47. The cell of claim 45 comprising a homozygous mutation in the endogenous λ immunoglobulin gene.

48. The cell of claim 29, further comprising a nucleic acid encoding all or part of a xenogenous heavy immunoglobulin gene capable of undergoing rearrangement and expressing a xenogenous heavy immunoglobulin.

49. The cell of claim 48, wherein said xenogenous heavy immunoglobulin is human heavy immunoglobulin.

50. The cell of claim 48, wherein said nucleic acid encoding all or part of xenogenous heavy immunoglobulin gene is a human artificial chromosome (HAC).

51. The cell of claim 50, wherein said HAC is ΔHAC, ΔΔHAC, κΔHAC-I, or κΔHAC-11.

52. A method of producing xenogenous antibodies, said method comprising the steps of:

(a) administering one or more antigens of interest to a transgenic non-human mammal comprising a nucleic acid encoding all or part of a xenogenous K immunoglobulin gene wherein the nucleic acid segments in said gene locus undergo rearrangement resulting in the production of antibodies specific for said antigen; and

(b) recovering xenogenous antibodies from said mammal.

53. A method of producing xenogenous antibodies, said method comprising recovering xenogenous antibodies from a transgenic non-human mammal a nucleic acid encoding all or part of a xenogenous K immunoglobulin gene wherein the nucleic acid segments in said gene locus undergo rearrangement resulting in the production of antibodies specific for said antigen.

54. The method of claim 52 or 53, wherein said mammal is an ungulate.

55. The method of claim 54, wherein said ungulate is a bovine, ovine, porcine, or caprine.

56. The method of claim 53, wherein said nucleic acid is contained within a chromosome fragment.

57. The method of claim 56, wherein said chromosome fragment is a K human artificial chromosome (KHAC).

58. The method of claim 57, wherein said KHAC is κΔHAC-I or κΔHAC-II.

59. The method of claim 52 or 53, wherein said mammal comprises a mutation that reduces the expression of an endogenous antibody.

60. The method of claim 59, wherein said mammal comprises a mutation in two genes encoding IgM heavy chain.

61. The method of claim 60, wherein one of said genes is IgμAY.

62. The method of claim 60, wherein both mutations are homozygous mutations.

63. The method of claim 62, wherein said mammal produces less than 10% of the endogenous IgM heavy chain, relative to a control mammal.

64. The method of claim 59, wherein said mammal comprises a mutation in one or both alleles of the endogenous K immunoglobulin gene, or one or both alleles of the endogenous λ immunoglobulin gene.

65. The method of claim 64, wherein said mammal comprises a mutation in both alleles of the endogenous K immunoglobulin gene.

66. The method of claim 64, wherein said mammal comprises a mutation in both alleles of the endogenous λ immunoglobulin gene.

67. The method of claim 52 or 53, wherein said xenogenous antibodies are human antibodies.

68. The method of claim 67, wherein said human antibodies have a galactose to asparagine residue ratio (Gal/N) or sialic acid to asparagine residue ratio (SA/N) that is within at least +/- 30% of the Gal/N or SA/N ratio of control human polyclonal antibodies.

69. The method of claim 68, wherein said human antibodies have a Gal/N and SA/N ratio that is within at least +/- 30% of the Gal/N and SA/N ratio of control human polyclonal antibodies.

70. The method of claim 67, wherein said human antibodies comprise a human heavy chain immunoglobulin and a human K chain immunoglobulin.

71. The method of claim 67, wherein said human antibodies comprise a human heavy chain immunoglobulin and an endogenous K chain immunoglobulin or endogenous λ chain immunoglobulin.

72. Purified antibodies produced by the method of claim 52 or 53.

73. Purified antibodies produced by a transgenic non-human mammal of claim 1.