NZ263925A

NZ263925A - Methods for deleting chromosomal dna from animal cells

Info

Publication number: NZ263925A
Application number: NZ263925A
Authority: NZ
Inventors: Daniel G Brenner; Robert B Dubridge; Gillis R Otten
Original assignee: Cell Genesys Inc
Priority date: 1993-03-15
Filing date: 1994-03-11
Publication date: 1997-12-19
Also published as: EP0702720A1; AU6518894A; NO953634L; FI954316A0; JPH08507696A; KR960701200A; NO953634D0; FI954316A; AU684275B2; CA2158364A1; WO1994021787A1; EP0702720A4

Description

<div class="application article clearfix" id="description"> New Zealand No. 263925 International No. PCT/US94/02676 TO BE ENTERED AFTER ACCEPTANCE AND PUBLICATION Priority dates: 15.03.1993; Complete Specification Filed: 11.03.1994 Classification: (6) C12N15/64.85; C12N5/16; A61K48/00 Publication date: 19 December 1997 Journal No.: NEW ZEALAND PATENTS ACT 1953 COMPLETE SPECIFICATION Title of Invention: Method for defined deletions of DNA Name, address and nationality of applicant(s) as in international application form: CELL GENESYS, INC., 322 Lakeside Drive, Foster City, California 94404, United States of America New Zealand No. International No. 263925 PCT/US94/02676 NEW ZEALAND PATENTS ACT 1953 COMPLETE SPECIFICATION Title of Invention: Method for defined deletions of DNA Name, address and nationality of applicant(s) as in international application form: o> CELL GENESYS INC^of 322 Lakeside Drive, Foster City, California 94404, United States of America 4 WO 94/21787 PCT/US94/02676 263925 METHOD FOR DEFINED DELETIONS OF DNA INTRQPWCTIQN Technical rigid The field of this invention concerns the genomic modification of vertebrate genes by defined deletions of 5 DNA. Background The ability to manipulate a vertebrate genome by homologous recombination has made it possible to alter the 10 genetic makeup of animal cells. When suitable host cells are used, transgenic animal technology can be utilized to permanently change the germline DNA of an entire animal, The genome can be manipulated to either add or delete DNA sequences. There are a number of uses for animals or 15 cells with targeted deletions of chromosomal DNA. Genes that one does not wish expressed can be removed, and the resulting cells used in therapeutic or experimental settings. A series of deletions may be made, where a variable 20 amount of DNA is removed by holding one end-point constant, and varying the position of the second end-point. Such a series may be used for the experimental mapping of genes, as there is interest in determining a genetic map for animals, particularly humans. 25 Some DNA sequences act as negative regulators for gene expression. It may be possible to activate the expression of certain genes by deleting those upstream or downstream elements which are responsible for silencing a ia WO 94/21787 PCT/US94/02676 gene using defined deletions of DNA to remove weak promoters or enhancers, and insert stronger foreign promoters or enhancers near the transcriptional initiation site of a native gene. Alternatively, stronger promoters 5 or enhancers may be moved from a wild type gene located upstream near the target gene. In order to fully utilize this technology there is interest in methods which can create genomic deletions of a defined size, at high frequency, and in certain 10 instances, for creating deletions without the presence of vector DNA sequences being left behind. Relevant Literature Scherer and Davis, (1979) Proc. Natl. Acad. Sci. USA 76(10):4951-4955, show the replacement of chromosome segments with altered DNA sequences in yeast cells. Mansour et al. (1988) Nature 336:348-352, describe a general strategy for targeting mutations to non-selectable genes. Thomas and Capecchi (1987) Cell 51:503-512, and Thompson et al. (1989) Cell 56(2):313-321 describe site-directed mutagenesis by gene targeting to correct the HPRT gene in mouse embryo-derived stem cells. Methods of targeting genes for modifications in embryonic stem cells are found in Valancius and Smithies (1991) Mol. Cell. Bio. 11:1402-1408, and Hasty et al. (1991) Nature 350(6315):243-246. Mombaerts et al. (1991) Proc. Natl. Acad. Sci. USA 88(8):3084-3087, describe the creation of a large genomic deletion in the T cell antigen receptor beta subunit locus by gene targeting. SUMMARY OF THE INVENTION Methods and DNA constructions are provided for simplified defined deletion or mutation of vertebrate 35 chromosomal DNA using homologous recombination. The replacement targeting vector comprises markers for positive and negative selection, homologous regions for 15 20 25 30 2 targeting at a specific site, and an additional homologous region to mediate the deletion. 263925 More specifically/ the present invention provides a method of introducing a defined deletion at a chromosomal target site in a host call, said method comprising: (1) transforming viable cells with a DNA targeting construct, said construct comprising: (A) two target DNA sequences, said target sequences being homologous to DNA sequences defining said target site, separated by: (B) a first additional DNA homology (v-ADH) sequence, said first v-ADH sequence being homologous to a first chromosomal ADH sequence (c-ADH) at a site located at a distance from said target site on the same chromosome, and (C) at least one selectable marker; (2) selecting for transformed cells that express said selectable marker; (3} selecting for transformed cells that lack said selectable marker; whereby homologous recombination occurs at said target site, resulting in integration of said DNA construct, and an intramolecular recombination occurs between said first v-ADH sequence and said first c-ADH sequence, resulting in a chromosomal deletion- between said target site and said first c-ADH site. The present invention also provides a method of introducing a deletion at a mammalian immunoglobulin locus in a host call, said method comprising: (1) transforming viable mammalian cells with a DNA construct, said construct comprising: (A) two target DNA sequences, said target sequences being homologous to DNA sequences at the constant region of said mammalian immunoglobulin locus, (B) said target sequences being separated by a v-ADH, sequence, said ADH sequence being homologous to a DNA sequence at site located at a distance 3J said target site on the same chromosome, (C) a positive selectable marker, anc (D) negative selectable marker, (2) selecting for transformed cells whi ss 263925 said positive selectable marker; (3) selecting for transformed cells which lac*, said negative selectable Barker; whereby homologous recombination occurs at said mammalian immunoglobulin locus, resulting in integration of said DNA construct, and an intramolecular recombination occurs between said v-ADH and c-ADH sequence in said chromosome, resulting in a chromosomal deletion between said target site and said c-ADH site. The present invention further provides a DNA construct comprising: (A) two target DMA sequences, said target sequences being homologous to proximal DNA sequences at a target site of a chromosome; said target sequences separated toy a (B) v—ADH sequence, said v-ADH sequence being homologous to a c-ADH sequence at a chromosomal site located distal from said target homologous DMA sequences on the same chromosome; and (C) a marker for positive selection; and (D) a marker for negative selection. The present invention additionally provides a method for inactivating a target locus located an a chromosome between a target site and a chromosomal c-ADH DMA sequence at a distance from said target site on the same chromosome comprising: 1) transforming viable mammalian cells with a DNA construct, said construct comprising: (A) two target DNA sequences, said target sequences being homologous to DNA sequences at a target site on a chromosome; said target sequences separated by a (B) v-ADH sequence, said v-ADH sequence beiI5§' homologous to a c-ADH sequence at a chromosomal site located distal from said target site on the-chromosome; <C) a marker for positive selection; and 2 2 OCT 1997 (D) a marker far negative selection, (2) selecting for transformed cells thjat eaffffififrSftld positive marker; (3) selecting for transformed cells that lack said negative marker; 3a whereby homologous recombination occurs at said 263925 target site, resulting In integration of said DMA construct, and an intramolecular recombination occurs between said v-ADH sequence and said c-ADH sequence, resulting in a chromosomal deletion of at least a portion of said target locus between said target site and said cADH site, wherein said target locus is inactivated. The present invention also provides a method for producing mammalian donor cells lacking MHC class I and/or Class II antigens for use in transplantation comprising: (1) transforming viable mammalian cells with a DNA construct for inactivating Class I and/or Class II MHC antigens, said construct comprising: (A) two target DNA sequences, said target sequences being homologous to DNA sequences within the MHC, (B) said target sequences being separated by a v-ADH sequence, said ADH sequence being homologous to a c-ADH DNA sequence at site located at a distance 5' or 3' to said MIIC on the same chromosome, and - (C) a positive selectable marker, and a (D) negative selectable marker, (2} selecting for transformed cells which express said positive selectable marker; (3} selecting for transformed cells which lack said negative selectable marker; whereby homologous recombination occurs within said MHC, resulting in integration of said DNA construct, and an intramolecular recombination occurs between said v-ADH and c-ADH sequence in said chromosome, resulting in a chromosomal deletion between said target site and said cADH site inactivating Class I and/or Class II MHC - antigens. In addition, the present invtn. mammalian cell that does not express produced by the method as defined cboye. 22rtfefW' C <ntigen I'l v c,Q 3b BRIEF DESCRIPTION OF THE DRAWINGS 26 Figure 1 is a diagram of a targeting construct and target chromosome, as described in the Example, infra. Figure 2 is a diagram of a bidirectional targeting construct and target chromosome, as described in the Example, infra. Figure 3 is a diagram of successive stages of deletion, as described infra. Figure 4 is a diagram of the targeting construct for inactivation of the immunoglobulin kappa light chain J and Constant regions and design of the targeting experiment described in the Example, infra. Figure 5 is a diagram of the construction of vectors for inactivating the immunoglobulin kappa light chain J and constant regions as described in the Example, infra. Figure 6 is a diagram of the final deletion vectors for inactivation of the immunoglobulin kappa light chain J and Constant regions as described in the Example, infra. Figure 7 is an illustration of the Southern analysis of light chain J and Constant region deleted cells as described in the Exautple, infra. DESCRIPTION OF THE SPECIFIC EMBODIMENTS Methods utilizing a DNA construct for homologous targeting via replacement of DNA are used to introduce a defined lesion at a specific site in a vertebrate chromosome. By "lesion" is meant an alteration in the DNA sequences at the site. The targeting construct may be used to delete large segments of chromosomal DNA, without leaving vector DNA sequences in the targeted chromosome. The method may also be used to alter a sequence in the target chromosome, without leaving behind vector DNA sequences. The targeting construct consists of a positive selectable marker, a negative selectable marker, which in NZ. ' 2 2 OCT 1997 —I RECFJVSD ! 3c WO 94/21787 PCT/US94/02676 some instances may be the same marker, sequences with homology to the target site in the chromosome, and an additional homologous sequence from a region outside of the sequences to be targeted in the chromosome. 5 Therefore, the target site in the chromosome is defined by the homologous sequences in the vector. The additional homologous DNA sequence, (ADH additional DNA homology) is homologous to a sequence on the same chromosome as the target sequences, which additional sequence can be located 10 either upstream or downstream on the chromosome. The sequence of DNA which lies between the ADH sequence and the target sequences on the chromosome will ultimately be deleted. A preferred targeting construct for use in the 15 invention is an "omega" or "replacement" targeting vector. However, an "O" or "insertional" targeting vector may also be used in some situations where a single target sequence is sufficient for providing homologous sequences for recombination. 20 The targeting construct is introduced into a host cell, preferably after linearization, where it undergoes homologous recombination to integrate at the site of the target sequences. The arrangement of elements in the linear construct is shown in Figure 1. However, while an 25 exemplary order of functional units is shown, it should be understood that the order of the functional sequences between the target sequences may be arranged in other ways, depending on the particular use contemplated. In addition, the chromosomal sequences that are to be 30 targeted by the homologous sequences in the targeting construct may be contiguous or separated. Where separated, homologous recombination in a double cross-over event will result in an initial deletion. For convenience, the DNA elements on the targeting 35 construct that are homologous to DNA sequences in the chromosome will be referrad to as vector or "v" homology sequences, while the chromosomal counterparts will be referred to as chromosomal sequences or "c". The two v- 4 WO 94/21787 PCT/US94/02676 target sequences necessary for homologous recombination will usually lie at the extreme 5' and 3' ends of the targeting construct forming arms of homology, except when positive-negative selection is employed in the homologous 5 recombination step, e.g. a negative selectable marker may be a terminal sequence. "c-ADH" or "v-ADH" refers to homologous sequences in the chromosome or. vector used for effecting the defined deletion. A double crossover event at the target sites in the 10 chromosome results in the replacement of the chromosomal region lying between the two c-target sequences, unless the c-target sequences are contiguous, with the construct sequences lying between the two v-target arms of homology. Sequences which are not functional in the vertebrate host, 15 such as bacterial origins of replication and antibiotic resistance, may be outside or inside of the construct arms. Sequences located outside of the arms will not be integrated into the chromosome, while those sequences between the arms will be integrated. The v-ADH region and 20 selectable markers for vertebrate cells will be positioned between the two arms of homology. After introduction of the construct DNA into the target cell, the positive selectable marker allows one to apply selective pressure, and to grow only those cells 25 which contain the integrated construct. An illustration of the resulting chromosome with integrated construct is shown in Figure 1 as the "Targeted chromosome". The illustration of the "Chromosome," shows the c-target sequences as contiguous, although it should be understood 30 that they may be separated on the chromosome, as described earlier. After integration, one copy of each target sequence anc* two copies of the ADH sequence will remain. One ADH copy is chromosomal in origin (c-ADH) , the other ADH sequence is derived from the targeting construct (v-35 ADH). The two copies are separated by the chromosomal sequences and at least the negative selectable marker of the targeting construct that lie between the v-ADH and c-ADH sequences. 5 WO 94/21787 PCT/US94/02676 Once the cells have been positively selected they car be directly negatively selected by the addition of negative selection media to the chIIs, for example using gancyclovir when the marker is the Herpes thymidine kinase 5 gene. Under negative selection conditions, cells that retain the negative selectable marker positioned between the target sequences, will usually die. Cells that have lost the negative selectable marker because of an intramolecular crossover event involving the homologous c-10 ADH and v-ADH sequences are able to proliferate under negative selection conditions. Alternatively, the positively selected cells may be screened to identify clones targeted by homologous recombination and the targeted clones can be subjected to negative selection. A 15 second independent stage of negative selection may be needed depending on the absolute efficiencies of the initial intermolecular homologous recombination step and of the deletion resulting from the intramolecular crossover. As noted, the method of the invention may also 20 employ negative selection using a different negative selectable marker from that positioned between the v-target sequences to incraase the efficiency of the intermolecular homologous targeting step. Because the c-ADH and v-ADH sequences are on the same 25 nucleic acid molecule, the deletion is an intramolecular recombination event, and will have a higher efficiency than homologous recombination which is intermolecular. The deletion event contemplated in this invention thus has a much higher probability of providing large deletions of 30 DNA than using intermolecular homologous recombination directly. An example of the order of elements in the construct is depicted in Figure 1. The 5' and 3' ends will usually be bounded by the v-target sequences. The v-ADH sequence 35 and the selectable markers will lie between the v-target sequences. The v-target and v-ADH sequences will all be in the same orientation, that is, the direction of the DNA r of the individual components will be the same in the 6 WO 94/21787 PCT/US94/02676 construct as it is on the chromosome, either 5' to 3', or 3' to 5'. The negative selectable marker, transcriptionally oriented in either direction, will lie between the v-ADH 5 sequence, and one of the v-target sequences in the construct. The exact position will depend on the position of the c-ADH sequence. On the targeted chromosome, the negative marker lies between the c-ADH and the v-ADH. The position of the positive selectable marker on the 10 construct will depend on the desired final product, because the excision step may or may not leave behind the positive marker in the chromosomal locus. If the positive and negative selectable markers are both positioned between the c-ADH and the v-ADH in the targeted 15 chromosome, then both will be deleted upon application of negative drug selection. If the negative selectable marker is positioned between the c-ADH and v-ADH - and the positive marker is outside, then the excision step will leave the positive marker behind in the chromosome of the 20 deleted locus. In this situation, it may be preferred to apply both positive and negative selection during the excision step. In general, the positive selectable marker may also be oriented in either transcription direction. As with the positive selectable marker, any exogenous 25 sequences from the construct which do not lie between the c-ADH and the vADH in the targeted chromosome will remain at the locus after the final recombination event. The arrangement of elements on the chromosome will be as shown in Figure l. The two c-target sequences will be 30 adjacent to each other to facilitate the initial double crossover event or may be separated to provide for a deletion upon homologous recombination. There will desirably be no more than 20 kb nucleotide distance between the target sequences on the native chromosome, 35 usually not more than about 10 kb, preferably not more than about 1 kb. The distance between the c-ADH and c-target sequences on the chromosome will depend on the desired final product. Normally this distance will be 7 WO 94/21787 PCT/US94/02676 much larger than the distance between the two c-target sequences. It nay be as much as 4000 kb in length, usually not more than about 100 kb in length, and will usually be at least about 500 bp in length. 5 The length of homologous v-target and v-AnH sequences will be at least about 50 nucleotides in length, preferably at least about 100 nucleotides, usually at least about 0.5 kb in length, and less than about 100 kb in length, usually less than about 20 kb, and preferably 10 less than about 1 kb. The regions of homology will have at least about 90%, preferably at least about 95%, and more preferably at least about 99%, sequence identity with the sequence of the native chromosome. The v-target sequences may provide for subtle changes 15 in the region of homology, such as introduction of a restriction site, changing the sequence of an enhancer or promoter locus, introducing or removing a splice site, or the like. By subtle change is intended fewer than a total of 10 bp substitutions and/or deletions, where the 20 percentage of base pairs involved in the target sequence does not exceed 10%, usually does not exceed 5%. Various markers may be used for selection. These include the HPRT minigene (Reid, et al. (1990) Proc. Natl. Acad. Sci. 87:4299-4303), the neo gene for resistance to 25 G418, the HSV thymidine kinase gene for sensitivity to gancyclovir, the hygromycin resistance gene, etc. One may use a marker that can be employed for both positive selection and negative selection, such as HPRT, in which case the marker must be positioned in the construct as a 30 negative selectable marker. Alternatively, one may use separate markers for positive selection, such as neo. hygromycin resistance, etc. and for negative selection, such as HSV-TK,. cytosine deaminase, etc. Other elements of the construct may include sequences 35 which code for-specific primer regions which may be used in the polymerase chain reaction (PCR) to identify recombinants, the addition of one or more restriction sites which allow for identification by gel WO 94/21787 PCT/CS94/02676 electrophoresis, the removal of a restriction site at the target locus, or other modification which allows for identification of target cells which have undergone the desired modification. 5 Various techniques may be used to introduce the linear DNA into the target cell. Techniques include electroporation, calcium precipitated DNA, fusion, transfection, lipofection, and the like. The particular manner in which the DNA is introduced into the cell is not 10 critical to this invention, although electroporation is preferred. Once the target cells have been transformed, the cells may then be selected by means of the marker gene by plating in a selective medium, growth in selective 15 culture, or Southern analysis. The cells may also be analyzed using PCR by employing primers which will provide for different sized fragments depending upon whether homologous recombination has occurred. In this way, target cells which have undergone the desired modification 20 may be identified. Modified cells may also be identified by changes gene expression, if appropriate. If a deletion will remove a negative regulatory element, or silencer, then one would expect increased gene transcription and protein 25 synthesis. Alternatively, if a gene or positive regulatory element, such as an enhancer, is deleted, then one would expect to find loss of transcription and protein synthesis. The presence or absence of the gene product may be detected by using specific antibodies, by 30 functional assays for the gene product, by detecting the absence or presence of mRNA from the gene, or the like. If the gene product is a surface membrane protein, one may use monoclonal antibodies in conjunction with FACS for identification of cells with the desired phenotype. 35 The subject methodology may by used for mammalian fine-structure genetic analysis, investigation of lesions implicated in genetic diseases and manipulation of gene expression. Genes which may be targeted include beta- WO 94/21787 PCT/US94/02676 globin, enzymes of erythrocyte metabolism, the complement system, coagulation factors, dystrophin, enzymes of carbohydrate, lipid, amino acid, steroid, purine and pyrimidine metabolism, transport proteins, such as cystic 5 fibrosis transmembrane regulator, immunoglobulin genes, T cell receptor genes, histocompatibility antigens, both major and minor, and the like. The subject method can find advantage in particular situations where other techniques may not suffice. 10 Because the second step in the process is intramolecular, deletion events may be achievable which could not otherwise be achieved by homologous recombination. Thus, genes may be modified by inactivating loci, where one wishes to inactivate a particular combination of exons, as 15 are present in the immunoglobulin loci, such as the heavy and/or light chain (kappa and lambda) loci, for example to eliminate the ability of a host animal to produce endogenous immunoglobulin heavy and/or light chains. Another application is the use of the method of the 20 invention to produce "universal donor cells" which do not express surface major histocompatibility complex (MHC) antigens, for transplantation. For example, a portion or substantially all of the MHC can be deleted using the method of the invention. The Class I and Class II MHC 25 antigens are heterodimers, each consisting of an a and a /S subunit. In Class I MHC antigens, the 0 subunit is /S2~ microglobulin. Of particular interest is the inactivation of at least one, preferably both, copies of a subunit of an MHC antigen such as ^-microglobulin. Alternatively, 30 deletion of other genes that affect MHC antigen expression may be accomplished using the method. For example, genes that regulate MHC antigen expression, such as the TAP1, TAP2 genes, LMP2 and LMP7 in the Class II locus that regulate MHC antigen dependent presentation, may be 35 deleted to prevent MHC presentation on mammalian cells. Depending upon the nature of the mammalian cell, the cell lacking at least one competent MHC antigen may find use as a donor to an allogeneic host or if an embryonic 10 WO 94/21787 PCT/US94/02676 stem cell, may find use in the production of chimeric mammalian hosts which themselves could b«a used as a source of organs for transplantation. Of particular interest are methods which provide for cells lacking at least one MHC 5 antigen, Class I or Class II, preferably Class I, which cells may serve a variety of functions in a viable host. The method involves transfection of mammalian cells, particularly normal cells, of a predetermined species with the targeting construct of the invention for deleting DNA 10 in one of the loci related to the /^-microglobulin gene, the a-subunit(s) of the Class I or II MHC antigens, the 0-subunit(s) of the Class II MHC antigens, or the genes associated with regulation of expression of the MHC antigens. The targeting construct will create a deletion 15 resulting in deletion in at least one, usually both copies, of the native gene(s), so as to prevent expression of a functional MHC antigen molecule. When the deletion is made in only one copy of the gene being inactivated, the cells having a single unmutated copy of the target 20 gene are amplified and may be subjected to a second transfozonation, where the deletion may be the same or different from the first lesion. The resulting transformants are screened for the absence of a functional target antigen and the DNA of the cell may be further 25 screened to ensure the absence of a wild-type target gene. Alternatively, homozygosity as to a phenotype may be achieved by breeding hosts heterozygous for the mutation. The cells which may be subjected to transformation may be any mammalian cells of interest, which may find use 30 in cell therapy, research, interaction with other cells in vitro or the like. Cells of particular interest include, among other lineages, the islets of Langerhans, adrenal medulla cells which may secrete dopamine, osteoblasts, osteoclasts, epithelial cells, endothelial cells, T-35 lymphocytes, neurons, glial cells, ganglion cells, retinal cells, embryonic stem cells, liver cells, bone marrow cells, and myoblast (muscle) cells. The cells may be obtained from any mammalian host, including murine and 11 WO 94/21787 PCT/US94/02676 other rodents, lagomorphs, porcine, feline, bovine, canine, human, etc. Cells from bare lymphocyte syndrome patients may be isolated in accordance with conventional ways, e.g., 5 panning, affinity columns, magnetic beads, or the like. By employing monoclonal antibodies specific for the lymphoid cell type, B- or T-cell, using monoclonal antibodies for such markers as CD 3, 4, 8, 10, 15 or 19, the desired group of cells and their progenitors may be 10 isolated in a substantially homogeneous composition. The genetically defective cells may be used in the same manner as MHC antigen defective cells produced by homologous recombination. The MHC antigen deficient cells will be selected to 15 achieve a particular function and be introduced into a mammalian host or used for research or other purpose. Also of interest will be the stem cells which act as the progenitors for any of the above cells, which may be the original progenitor or a progenitor cell which is already 20 dedicated to a particular lineage. Of particular interest will be epidermal cells, such as keratinocytes, retinal epithelial cells, myoblasts, hematopoietic cells, and other cells which may be readily manipulated in vitro, maintained for long periods of time in culture and may db 25 introduced into a host, where the cells will remain viable and functional for long periods of time. For embryonic stem cells, an embryonic stem cell line may be employed or embryonic stem cells may be obtained freshly from a host such as a murine animal, e.g. a mouse, 30 rat, guinea pig, Chinese hamster or other small laboratory animals. The cells may be grown on an appropriate fibroblast-feeder layer or grown in the presence of leukemia inhibiting factor (LIF) and then used for mutation.' 35 The targeting constructs of the invention may be modified to include functional entities that may find use in the preparation of the construct, amplification, transformation of the host cell, and integration of the 12 WO 94/21787 PCT/US94/02676 construct into the host cell. Techniques which may be used include calcium phosphate/DNA coprecipitates, microinjection of DNA into the nucleus, electroporation, bacterial protoplast fusion with intact cells, 5 transfaction, or the like. The DNA may be single or double stranded, linear or circular, relaxed or supercoiled DNA. For various techniques for transforming mammalian cells, see Keovm al., Methods in Enzvmoloav (1990) 185:527-537. Once the construct has been prepared 10 and manipulated and the undesired sequences removed from the vector, e.g., the undesired bacterial sequences, the DNA construct is now ready to be introduced into the target cells. As already indicated, any convenient technique for introducing the DNA into the target cells 15 may be employed. After transformation of the target cells, many target cells are selected by means of positive and/or negative markers, as previously indicated, neomycin resistance and Acyclovir or Gancyclovir resistance. Those cells which show the desired phenotype may then be further 20 analyzed by restriction analysis, electrophoresis, Southern analysis, polymerase chain reaction or the like. The resulting transformed cells may then be selected by the absence of the target MHC antigen on the surface of the cell. This can be achieved in a variety of ways. For 25 example, one may use antibodies to any epitope of the target MHC antigen in conjunction with complement to kill any cells having the antigen. Alternatively, one may use conjugates of the appropriate antibody, particularly monoclonal antibody with a toxin, such as the A chain of 30 ricin, abrin, diphtheria toxin, or the like. Affinity chromatography may be employed, where antibodies may be used to remove cells comprising the target antigen. The resulting cells which survive should be free of at least one MHC antigen on their surface and now not be as subject 35 to transplant rejection when introduced in vivo as wild-type cells. The cells may then be grown in an appropriate nutrient medium for expansion and used in a variety of 13 WO 94/21787 PCT/US94/02676 ways. The cells may be used for transplantation, to become part of an existing tissue, or may be grown to form tissue for transplantation into a non-syngeneic host. For example, with keratinocytes, the cells may be used for 5 replacement of skin in the case of burns, where keratinocytes may be grown to form a continuous layer prior to application. Similarly, the keratinocytes may be used in the case of plastic surgery to replace skin removed from the host for use at another site, other uses 10 for the keratinocytes include transplantation in decubitus ulcers. In the case of islets of Langerhans, they may be grown and introduced into capsules or otherwise for insertion into a host for the production of insulin. In 15 the case of retinal epithelial cells, they may be injected into the subretinal space of the eye to treat visual disorders, such as macular degeneration. In the case of immune cells, they may be injected into the bloodstream or elsewhere to treat immune deficiency. In the case of 20 myoblasts, they may be injected at various siteB to treat muscle wasting diseases, such as Duchenne muscular dystrophy. For organ transplants, non-syngeneic tissue such as xenogeneic grafts of heart or liver may be performed between related species. 25 Depending upon the nature of the cells, the therapy involved, and the disorder, the cells may be employed as films, introduced in containers for maintenance at a particular site, or as solid masses impregnated in inert matrices or independent or a matrix. The number of cells 30 administered will vary widely, depending upon the particular application and the manner in which the cells are administered. Administration may be by injection, topical application, incision and placement, in the appropriate location. 35 Another situation is where one wishes to introduce a sequence at a target site which is refractory to homologous recombination. By employing the subject methodology one can introduce the sequence at a locus 14 WO 94/21787 PCT/US94/02676 distant from the refractory site and by use of the ADH sequences Can delete the region between the site of insertion and the refractory locus, even including the refractory locus. In this manner, one may introduce an 5 enhancer, promoter, exon or the like, into the refractory site, for example to activate the expression of a gene of interest. Thus, the subject methodology can be used to introduce a sequence of interest at a site proximal to a locus to be influenced by the sequence of interest, where 10 the locus is refractory to homologous recombination, and one wishes to delete a region proximal to the refractory locus while bringing an exogenous sequence in proximity to the refractory locus, or the like. The target cells may be any of a variety of 15 vertebrate cells, particularly animal cells, more particularly mammalian cells. Of particular interest are embryonic stem cells, which may be used to create transgenic animals, and thereby introduce the altered chromosome into the germline of a host animal. Embryonic 20 cells of particular interest include rodent cells, e.g. mouse, rat and guinea pig. The method may be used to effect deletions in the chromosome of various sizes. Relatively small deletions (500 bp to 15 kb) can be achieved in a defined manner. 25 Such small deletions can also be achieved by standard targeting methodology, by choosing target regions for recombination that are separated by the desired deletion. If standard methodology is used, however, a positive marker must then be tolerated in the final locus. When no 30 extra marker is desired in the final locus, then the method of this invention will permit one to effect a deletion without leaving behind any exogenous sequences. Large deletions, up to 4000 kb, can be achieved by the methodology of this invention. The size of deletions 35 that can be made by standard methodology are restricted and vary in efficiency. The method of the invention is far more powerful than conventional techniques, because it relies, not on intermolecular homologous recombination for 15 WO 94/21787 PCT/US94/02676 the excision step, but rather upon intrachromosomal recombination, using negative selection. Large deletion events resulting from the subject invention are more likely to occur at frequencies within the scope of a 5 typical experiment, as compared to a deletion event resulting from a standard targeting experiment. In certain instances# deletion of large stretches of DNA may delete genes essential to the viability of the cell into which the targeting construct is introduced. In 10 such cases, the essential genes that may be lethal recessive genes, may be added back into the chromosome via the targeting construct or by secondary transfection with other vectors. The essential genes must be added back prior to rendering the deletion homozygous. 15 The method of the subject invention may have a number of embodiments. Shown in Figure 2 is one embodiment of the invention, wherein a second negative selectable marker and ADH sequence are included in the construct. The second vADH will not normally share any sequence identity 20 with the first v-ADH. The second c-ADH site will be located on the same chromosome as the first c-ADH and the c-target sequences, with the c-target sequences being located between the two c-ADH sequences. After the initial targeting event, negative selection will be 25 applied. This will select for those cells which have deleted the sequence between the first c-ADH and v-ADH. A second round of negative selection, desirably for a different marker, will select for those cells which have deleted the sequence between the second v-ADH and c-ADH. 30 In this way, the deletion will be bidirectional from the site of the initial targeting event. Alternatively, a series of v-ADH and negative selection marker sequences can be arranged in a sequential manner on the targeting construct to delete in successive 35 stages chromosomal DNA sequences in one direction proceeding from the site of the initial targeting event on the chromosome (Figure 3). Thus, increasingly larger '■ segments of contiguous DNA may be deleted using the method 16 WO 94/21787 PCT/US94/02676 of the invention. The targeting construct can be constructed with an arrangement of elements as described herein to accomplish bidirectional and staged successive unidirectional deletions to effect particular deletions in 5 the chromosome. In another embodiment, the ADH sequence may be a repetitive sequence. In that case, the end point for the deletion could be at a number of different sites, where ever the repetitive element is found. In this way, a 10 series of nested deletions could be constructed, with a constant starting point for the deletion, and a variable amount of DNA deleted from that site. Repetitive sequences of interest include the Alu, LI, a-satellite, telomeric or sub-telomeric repeat sequences found in man, 15 and analogous repeat sequences found in other mammals. The ADH sequence may be homologous to one of the target sequences, but incorporating a genetic lesion or modification. The genetic lesion or mutation may be an insertion, substitution or deletion in the DNA sequence. 20 The number of altered nucleotides will usually be less than about 20 bp, more usually less than about 10 bp, and will be at least 1 bp. Chromosomal sequences will usually not intervene between the ADH and target sequences in this case, therefore the only sequences which are deleted will 25 be vector sequences. In this way, a genetic modification can be introduced without leaving behind marker and vector sequences. The following examples are offered by way of illustration and not by way of limitation. 30 EXPERIMENTAL 35 Inactivation of Murine Immunoglobulin Kappa Light Chain J and Cpngtant Regjpng Design pf the targeting experiment The targeting vector was designed as a replacement type vector initially to delete the constant region as 17 WO 94/21787 PCT/US94/02676 well as the J region of the kappa locus and replace it with three elements through homologous recombination using regions of homology flanking the constant region (Figure 4). A diphtheria toxin gene (A chain) flanking either or 5 both regions of homology was included in some cases as a negative selectable marker. The three elements consisted of the G418 resistance drug marker, an ADH sequence of mouse DNA homologous to a region of the kappa locus located upstream of the J region, and a thymidine kinase 10 gene. As a result of the inclusion of the ADH sequence in the vector, this initial targeting placed a second copy of the ADH sequence in the locus. This duplication was then used to effect a deletion of the sequences between the segments by applying selective pressure. In this case the 15 cell deletes the thymidine kinase gene that lies between the two segments in order to survive gancyclovir selection. B. Construction of the targeting vector 20 The regions of homology were derived from a 129 mouse fetal liver genomic library (Stratagene, San Diego, CA) which was screened using two probes. The first probe was a 1.6 kb Hpal/BamHI fragment (Steinmetz and Zachau, (1980), Nuqleic Research 8:1693-1706) that spans the 25 constant region. A lambda phage clone which hybridized to this probe was identified and used to purify and isolate phage DNA. Analysis of this DNA showed that the Hpal/BamHI probe hybridized to a 5.6 kb Sphl/BamHI fragment which was then subcloned between the SphI and 30 BamHI sites of the plasmid pUC218 to give the plasmid pUC218/5.6kappa. This subclone contained the J region, an intronic enhancer element and the constant region of the kappa light chain locus. The second probe was a 0.8 kb EcoRI fragment (Van Ness jajfc &1. (1981), Cell 27:593-602) 35 that lies 2.8 kb upstream of the J region. Phage DNA from a lambda clone positive for this probe showed that the probe hybridized to a 5.5 kb SacI fragment which was 18 QPR 07 '95 12 = 55PM FLEHR HOHEPCH TcHT P. 2/2 . ... P2Sg94/0|6?6 • ^OTAPRms subcloned into the Sacl site of pBlusacript SKI (Stratagene) to give the plasmid pSK.A5'K (Figure 5). The inactivation vectors which contained a 5' region of homology, a thymidine kinase gene, an ADH, a neomycin 5 resistance gene and a 3' region of homology (Figure 6) flanked in some instances by diphtheria toxin genes were constructed from three plasmids (Figure 5) containing: (a) the 5' fragment of homology with or without the diphtheria toxin gene (DT) driven by the mouse 10 phosphoglycerate kinase gene (FGK) promoter as a negative selectable marker, (b) the herpes thymidine kinase gene (tk) driven by the mouse phosphoglycerate kinase gene (PGX) promoter as a negative selectable marker along with the ADH and the G418 selectable neomycin (neo) gene from 15 pKClNeo (Thomas and Capecohi (1987), cell Si:503-12), and (c) the 3' fragment of homology with or without the pGK driven DT gene. These three plasmids (Figure 6) were constructed from pSK.A, pSK.B, and pSK.C, respectively, all derived from the plasmid pBluescript 8X1 by 20 modification of the polylinker. The polylinker of the plasmid pBluescript sxx was modified by cloning between the KpnZ and Sacl sites a synthetic polylinker defined by the oligonucleotides 5' -GCATATGCCTGAGGGTAAGCATGCGGTAC CGAATTCTATAAGCTTGCGGCCGCAGCT-3 ' AMD 5' -GCGGCCGCAAGCTTAT 25 AGAATTCGGTACCGCATGCTTACCTCAGGCATATGCGTAC-3' to create the plasmid psx.A, 5'-GAGCTCGGATCCTATCTCGAGGAATTCTATAAGCTTCA TATGTAGCT-3' and 5' -ACATATGAAGCTTATAGAATTCCTCGAGATAGGATC CHAGCTCGTAC-3' to create the plasmid pSX.B, 5'-AAGCTTATAG AATTCGGTACCTGGATCCTGAGCTCATAGCGGCCGCAGCT-3' and 5'-GCGGCC 30 GCTATGAGCTCAGGATCCAGGTACCGAATTCTATAAGCTTG TAC-3' to create the plasmid pSK.C. A diphtheria toxin gene cassette was created in which the gene was flanksd by the PGX promoter and the bovine growth hormone polyadenylation signal (Woychik fli aI* 35 f 19841:. Proc. Nqtl. Acad. Sal. P.S.A., 81:3944-3948; Pfarr Al. (1986), &HA 5:115-122). A 2.3 kb Xbal/IcoRI fragment from pTH-l (Maxwell &1. (1986), Cancar Sjflgj- __ 4<:4660-4664) containing the diphtheria toxin A chjafchZ. PATENT OFFICE 19 AMENDED 9CET - 4 JUL 1995 hfcv" WO 94/21787 PCT/US94/02676 driven by the human metallothionein (hMTII) promoter was cloned into pBluescript SKI cut with Xbal and EcoRI to give the plasmid pSK.DT. The hMTII promoter of pSK.DT was replaced with the PGK promoter from pKJl (Tybulewicz sSi 5 al. (1991), Cell 65:1153-1163). A 0.5 kb Xbal/PstI fragment from pKJl was joined to a 3.1 kb Xbal/Ncol fragment from pSK.DT using a Pstl/Ncol adapter formed from the oligonucleotides 5'-GGGAAGCCGCCGC-3' and 5'-CATGGCGGC GGCTTCCCTGGA-3' to give the plasmid pSK.pgkDT. A 248 bp 10 fragment containing the bovine growth hormone polyadenylation signal, obtained by PCR amplification of bovine genomic DNA using the oligonucleotide primers 5 ' -CAGGATCCAGCTGTGCCTTCTAGTTG-3' and 5' -CTGAGCTCTAGACC CATAGAGCCCACCGCA-3', was Cloned into pCRlOOO (Invitron 15 Corp., San Diego, CA). The polyadenylation sequence was then cloned behind the DT gene as a Hindlll/pvull fragment into pSK.pgkDT cut with Hindlll and Hpal to give the plasmid pSK.pgkDTbovGH. The DT gene cassette from pSK.pgkDTbovGH was moved as a 2.1 kb EcoRI/Hindlll 20 fragment into pSK.A cut with EcoRI and NotI using a Hindlll/NotI adapter formed from the oligonucleotides 5'-AGCTGGAACCCCTTGC-3' and 5'-GGCCGCAAGGGGTTCC-3' to give the plasmid pSK.A/DT. Between the SphI and Bsu365 sites of both pSK.A and pSK.A/DT the 5' region of homology for 25 the kappa locus was cloned. For this purpose a 4.0 kb Sphl/Bsu361 fragment resulting from a partial Bsu36I digest followed by a complete SphI digest of plasmid subclone pUC218/5.6kappa was ligated to pSK.A or pSK.A/DT to give the plasmids pSK.A/5'K and pSK.A/DT/5'K, 30 respectively. In the plasmid, pSK.A/DT/5'K, the 5'-end of the DT gene and kappa fragment were adjacent to each other running in the opposite transcriptional orientations. The PGKtk gene from the plasmid pKJtk (Tybulewicz £l. (1991), Cell 65:1153-1163) was cloned as a 2.7 kb 35 EcoRI-/Hindi 11 between the unique EcoRI and Hindlll sites of pSK.B to give pSK.B/TK. A 0.8 kb EcoRI fragment used - for the ADH was cloned from pSK.S'kappa and was ligated into the EcoRI site of pSK.B/TK to give pSK.B/(TK/0.8K) 20 APR 07 '95 12J56PM FLEHR HOHEfiCH TEST PCT/0S9iv/5fl>76 IPEA/US 07 APR 1995 such that the 5'-end of the tk gene and kappa fragment were adjacent to each other running in opposite transcriptional orientations. The 1.1 kb neo gene from pMClNeo was. cloned as an XhoX/BamHI fragment between the 5 same sites of pSK.B/ (TX/0.8K) to give pSK.B/(TK/O.8K/Neo). The plasmid pSK.C/3'K containing the 3' fragment of homology was constructed by ligating pSK.C digested with BamHX and treated with alkaline phosphatase to the 1.1 kb BgllX/BamHX fragment isolated from pUC218/5.6kappa. In 10 psx.C/3'K, the kappa fragment was oriented such that transcription proceeded from the Sacl in the plasmid polylinker in the direction of the KpnX site. The 2.1 kb DT cassette from pSK.pgkDTbovGH was cloned as an ScoRI/HindiXX fragment into the same sites of pSK.C to 13 give pSK.C/3'X/DT. Three-part ligations were carried out to construct the final targeting plasmids (Figure 6). The 4.0 kb NotX/NdeX fragment from pSX.A/S'X, the 4.8 kb NdeX/SacX fragment from pSK.B/(TX/0.8K/Neo) (obtained by a Sacl 20 partial followed by an NdeX digestion of the plasmid), and i the 4.0 kb SacX/NotI fragment from pSX.C/3'K were isolated and ligated together to create pK.(TK/0.8K/Neo). The 6.1 kb NotX/NdeX fragment from pSX.A/DT/5'K, the 4.8 kb NdeX/Sacl fragment from pSK.B/(TK/0.8K/Neo), and 4.0 kb 25 SacX/NotX fragment from pSK.C/3'K were isolated and ligated together to create pK.DT/ (TX/0.8X/Neo). The 6.1 kb Notl/Ndel fragment from pSX.A/DT/5'K, the 4.8 kb NdeX/SacX fragment from pSK.B/(TK/0.8X/Neo), and 6.1 kb SacX/NotX fragment from pSK.C/3'K/DT (obtained by a Sacl 30 partial followed by a NotX digestion of the plasmid) were isolated and ligate together to create pK.DT/ (TK/0.8X/ Neo)/DT. For electroporation, the purified plasmid DNAs were first cut with FvuX or ApaLX, then extracted with phenol/chloroform and precipitated by the addition of 35 ethanoi before centrifugation. The resultant DNA pellets were reeuspended at a concentration of 1 mg/ml in 10 mM Tris-HCl, 1 mM EDTA(TE). WO 94/21787 PCT/US94/02676 Introduction of DNA into eella The embryonic stem cell line E14-1, a subclone of E14 (Hooper et al., (1987) Nature 326:292-295), was cultured in DMEM with 4.5 g/1 glucose (JRH Biosciences, Irvine, CA) 5 supplemented with 15% heat inactivated fetal calf serum, recombinant murine leukemia inhibitory factor (ESGRO from Gibco BRL, MD, 1000 units/ml), /8-mercaptoethanol (0.1 mM), glutamine (2 mM) and penicillin (100 U/ml)/streptomycin (0.1 mg/ml) and grown at 37°C in 5% CO]. The cells were 10 cultured on mitomycin-treated primary embryonic fibroblast feeder layers essentially as described by Koller and Smithies, (1989), supra. The embryonic fibroblasts were prepared from day 14 embryos carrying the homozygous targeted mutation of /^-microglobulin (Koller et al., 15 (1990) Science 248:1227-30). These feeder cells are capable of growth in media containing G418. At 80% confluency, the ES cells were prepared for electroporation by trypsinization, concentration by brief centrifugation and resuspension in HEPES-buffered saline at 2 X 107 20 cells/ml. The cells were equilibrated at room temperature, and DNA (20 ng) linearized (as described above) was added. The mixture was electroporated at 960 HF and 250 V with a BioRad Gene Pulser. The cells were left to stand at room tempierature for 10 minutes before 25 plating onto a 4 x 10 plate of mitomycin C treated feeders (3 x 106 feeder cells/plate). After incubation at 37°c for 48 hr, the cells were fed media containing G418 at 150 fig/ml (effective concentration). 30 £. Analysis of constant region-targeted ES cells Aft ex- 7-10 days under drug selection with G418, the individual surviving colonies were each picked and dissociated in a drop of trypsin in a 96-well plate and then incubated at 37°C for 2 min. The cells from each 35 colony were transferred into a well of a 24-well plate containing mitomycin C treated feeder cells and selective media With G418 at 150 /xg/ml. After an additional 5-8 days, 20% of the cells in each well were frozen and the 22 WO 94/21787 PCT/US94/02676 remainder were used to prepare genomic DNA. The cells were lysed with 0.4 ml of 10 mM Tris-HCl (pH 7.5), 100 mM NaCl, 10 mM EDTA, 1% SDS, and proteinase K (1 mg/ml) by incubation overnight at 50°C. The DNA was purified by 5 phenol extraction and ethanol precipitation. The DNA pellets were washed with 70% ethanol, dried and resuspended in TE (20 pi). Southern analysis was carried out using Bgl II digested genomic DNA from each sample. A 2.3 kb fragment 10 was detected from the native ES cell locus, while a larger 4.9 kb fragment was detected from a targeted ES cell locus (Figure 7), using as a probe the 1.2 kb BamHI/Bgl II fragment isolated from the original phage DNA contiguous with the fragment used for the 3' homology in the 15 targeting vector. The fragment increased in size because the Bgl II site in the Bgl II/BamHI fragment was lost in the targeting plasmid due to the joining of a Bgl II site to a BamHI site in the ligation, and a new Bgl II site located in the thymidine kinase gene was introduced into 20 the targeted locus. From a screen by the Southern analysis described above, of a total of 103 clones derived from experiments using three different targeting plasmids, 5 cell lines were identified which carried the intended mutation (Table 25 1). Table 1 CK Light Chain Targeting Result in E14-1 Cells Construct Nvunber Screened by Southern Number of Confirmed Targeted Clones Clone Designatio n Frequency | of Targeting pK.(TK/0. 8K/Neo) 44 2 625,691 1/22 pK.DT(TK/ 0-8/Neo) 42 2 604,611 1/21 pK.DT(TK/ | 0.8K/Neo) 1 17 1 653 1/17 23 WO 94/21787 PCT/US94/02676 Further analysis of genomic DNA produced from 4 of the positive clones (clones 625, 604, 611 and 653) after being thawed and expanded, re-confirmed the initial 5 observations. Using a second probe, a 1.7 kb Hindlll/Bgl II fragment which spanned the J region of the kappa locus, the correct integration pattern was checked for homologous targeting at the 5' end of the targeting vector. Thus, using this probe with an EcoRI digest of the genomic DNA, 10 a 15 kb fragment was detected from the unmodified allele. In contrast, a 7.8 kb fragment from the targeted allele was observed as a result of the introduction of a new EcoRI site in the thymidine kinase gene during the homologous integration (Figure 7). 15 E. In vitro excision of J region DNA from targeted clones In order to effect the desired deletion firom the homologously targeted kappa locus, cells from clone 653 were plated on feeder cells at a density of 0.5-1 x 106 20 cells/10 cm dish in the presence of both gancyclovir (2 MM) and 6418 (150 /zg/ml). After growth for 5 days in the presence of both drugs, clones were picked as described above into 24-well plates and grown under G418 selection alone. After an additional 5-8 days, 20% of the cells in 25 each well were frozen and the remainder used to prepare genomic DNA as previously described. £Ls Analysis of J/constant region deleted ES cells Southern analysis was carried out using BamHI 30 digested genomic DNA from each sample. Using as a probe the 0.8 kb EcoRI fragment used as the ADH in the targeting vectors, as 12.7 kb fragment was detected from the native ES cell locus, while a larger 15.8 kb fragment was detected from the constant region-targeted ES cell locus 35 (Figure 7) using DNA from clone 653. The fragment increased in size because of the insertion of the tk gene, the ADH, and the neo gene into the 12.7 kb BamHI fragment. There was also a new BamHI site introduced at 'the 3' end 24 WO 94/21787 PCT/US94/02676 of the neo gene. Using DNA from the J/constant region deleted cells, a 5.5 kb fragment was detected from the modified locus in addition to the 12.7 kb fragment from the untargeted allele as predicted from analysis of the 5 restriction map. From this screen by Southern analysis of 2 clones produced from 1.5 x 10* ES cells plated (clone 653), one cell line (clone 653B) was identifier' which carried the intended deletion of the J and constant regions. 10 Further analysis of genomic DNA produced from clone 653B after being thawed and expanded re-confirmed the initial observations. Using the 0.8 kb EcoRI fragment, the deletion was checked with two other restriction digests which should cut outside of the excised region on 15 the 5' and 3' ends of the targeting vector. Thus using this probe with a Bgl II digest of the genomic DNA from the unexcised clone 653, a 2.6 kb fragment was detected from both the unmodified and modified alleles, whereas an additional 4.9 kb fragment was observed from the targeted 20 allele only (Figure 7). This 4.9 kb fragment was the same as that detected with the 1.2 kb BamHI/Bgl II fragment used previously. Using DNA from clone 653B, a BglII digest revealed a 5.8 kb fragment in addition to the 2.6 kb fragment from the unmodified allele. A Sacl digest of 25 clone 653 DNA probed with the 0.8 kb EcoRI fragment showed a 5.5 kb fragment from both the unmodified and modified alleles and a 3.1 kb fragment from the targeted allele only (Figure 7). The 5.5 kb fragment was also detected in DNA from clone 653B and an additional 2.0 kb fragment. 30 The 5.8 kb Bgl II fragment and the 2.0 kb Sacl fragment were consistent with an analysis of the predicted restriction map for a precise excision step in which 10.3 kb of DNA were deleted including the J region, the tk gene, and one copy of the ADH. 35 G. Generation of Germline Chimeras The unmodified E14-1 cells contributed to the germline at a high frequency after injection into C57BL/6J 25 WO 94/21787 PCT/US94/02676 blastocysts. The cells from the targeted ES cell line 653B which were grown on primary feeder layers as described, were trypsinized and resuspended in injection medium, which consisted of DMEM supplemented with 15% 5 fetal calf serum, 20 mM HEPES (pH 7.3), antibiotics and 0-mercaptoethanol. The ES cells (10-15) were injected into eaich blastocyst, and injected blastocysts (10) were transferred to a pseudopregnant female mouse, 5 into each uterine horn. Chimeric pups are identified by chimeric 10 coat color. Chimeric males are bred to C57BL/6J females, and germline transmission of the 129 derived ES cell (unmodified or targeted) is detected by the agouti coat color of the F1 offspring. All publications and patent applications mentioned in 15 this specification are indicative of the level of skill of those skilled in the art to which this invention pertains. All publications and patent applications arc herein incorporated by reference to the same extent as if each individual publication or patent application was 20 specifically and individually indicated to be incorporated by reference. The invention now being fully described, it will be apparent to one of ordinary skill in the art that many changes and modifications can be made thereto without departing from the spirit or scope of the appended claims. 26 WO 94/21787 PCT/US94/02676 SEQUENCE LISTING (1) GENERAL INFORMATION : (i) APPLICANTt Brenner, Daniel 6. Dubridge, Robert B. Otten, Gillie R. (ii) TITLE OP INVENTION! Method for Defined Deletions of DNA (iii) NUMBER OP SEQUENCES t 12 (iv) CORRESPONDENCE ADDRESS: (A) ADDRESSEE: Flehr, Hohbach, Test, Albritton fi Herbert (B) STREET} Pour Embarcadero center, Suite 3400 (C) CITY: San Francisco (D) STATE: California (E) COUNTRY: USA (P) ZIP: 94111 (v) COMPUTER READABLE FORM: (A) MEDIUM TYPE: Floppy disk (B) COMPUTER: IBM PC compatible (C) OPERATING SYSTEM: PC-DOS/MS-DOS (D) SOFTWARE: Patentln Release #1.0, Version #1.25 (vi) CURRENT APPLICATION DATA: (A) APPLICATION NUMBER: PCT/US94/ (B) PILING DATE: ll-MAR-1994 (C) CLASSIFICATION: (viii) ATTORNEY/AGENT INFORMATION: (A) NAME: Rowland, Bertram I. (B) REGISTRATION NUMBER: 20,015 (C) REFERENCE/DOCKET NUMBER: FP-57537/BIR CELL-015 (ix) TELECOMMUNICATION INFORMATION: (A) TELEPHONE: (415) 494-8700 (B) TELEFAX: (415) 494-8771 (C) TELEX: 910 277299 (2) INFORMATION FOR SEQ ID NO:l: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 57 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: both (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (ix) FEATURE: (A) NAME/KEY: misc feature (B) LOCATION: complement (1..53) (D) OTHER INFORMATION: /note* "Complementary to SEQ ID NO:2, bases 1 to 52." (ix) FEATURE: (A) NAME/XEY: misc feature (B) LOCATION: complement (2..7) (D) OTHER INFORMATION: /note- "Ndel restriciton site." (ix) FEATURE: (A) NAME/KEY: misc feature (B) LOCATION: complement (8..14) 27 WO 94/21787 PCT/US94/02676 (D) OTHER INFORMATION: /note- "Bsu36I restriction site." (ix) FEATUREt (A) NAME/KEY: misc feature (B) LOCATION: complement (19..24) (D) OTHER INFORMATION: /note* "SphI restriction site." (ix) FEATUREi (A) NAME/KEY: misc feature (8) LOCATION: compTement (25..30) (D) OTHER INFORMATION: /note- "Kpnl restriction site." (ix) FEATURE: (A) NAME/KEY: misc feature (B) LOCATION: complement (31..36) (0) OTHER INFORMATION: /note- "EcoRI restriction site." (ix) FEATURE: (A) NAME/KEY: misc feature (B) LOCATION: complement (40 .45) (D) OTHER INFORMATION: /njjfce ■ "Hindlll restriction site." (ix) FEATURE: (A) NAME/XEY: misc feature (B) LOCATION: compTement (46..53) (D) OTHER INFORMATION: /note- 'NotI restriction site." (xi) SEQUENCE DESCRIPTION: SEQ ID NO:l: GCATATGCCT GAGGGTAAGC ATGCGGTACC GAATTCTATA AOCTTGCGOC CGCAGCT 57 (2) INFORMATION FOR SEQ ID NO:2: (1) SEQUENCE CHARACTERISTICS: (A) LENGTH: 56 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: both (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (ix) FEATURE: (A) NAME/KEY: misc feature (B) LOCATION: compTement (1..52) (D) OTHER INFORMATION* /note- "Complementary to SEQ ID NO:l, bases 1 to 53." (ix) FEATURE: (A) NAME/KEY: misc feature (B) LOCATION: compTement (1..8) (D) OTHER INFORMATION: /note- "NotI restriction site." (ix) FEATURE: (A) NAME/KEY: misc feature (B) LOCATION: compTement (9..14) (D) OTHER INFORMATION: /note- "Hindlll restriction site." (ix) FEATURE: (A) NAME/KEY: misc feature (B) LOCATION: compTement (18..23) (D) OTHER INFORMATION: /note- "EcoRI restriction cite." (ix) FEATURE: (A) NAME/XEY: misc feature (B) LOCATION: compTement (24..29) (D) OTHER INFORMATION: /note- "Kpnl restriction site." 28 WO 94/21787 PCT/US94/02676 (ix) FEATURE} (A) NAME/KEY 1 miSC £•«ture (B) LOCATIONt complement (30..35) (D) OTHER INFORMATION: /note- "SphI restriction site." (ix) FEATURE} (A) NAME/KEY: misc feature (B) LOCATION: complement (39..45) (D) OTHER INFORMATION: /note- "Bsu36I restriction site." (ix) FEATURE: (A) NAME/KEY: misc feature (B) LOCATION: complement (46..51) (0) OTHER INFORMATION: /note- "Ndel restriction site." (Xi) SEQUENCE DESCRIPTION! SEQ ID NO:2: GCGGCCGCAA GCTTATAGAA TTCGGTACCO CATQCTTACC TCAOOCATAT OCOTAC 56 (2) INFORMATION FOR SEQ ID NOi3i (i) SEQUENCE CHARACTERISTICS: (a) LENOTHi 47 base pairs (B) TYPE: nucleic acid (c) strandedness! both (D) TOPOLOGY: linear (ii) MOLECULE TYPEi CDNA (ix) FEATURE! (A) NAME/KEY: misc feature (B) location! complement (1..43) (D) OTHER INFORMATION! /note- "Complmentary to SEQ ID NO!4, baees 1 to 43." (ix) FEATURE! (a) name/key: misc feature (B) location! compTement (1..6) (D) OTHER INFORMATION! /note- "Sacl restriction site." (ix) FEATURE! (a) name/key: misc feature (B) LOCATION: compTement (7..12) (D) OTHER INFORMATION! /note- "BamHI restriction site." (ix) FEATURE: (a) name/key: misc feature (B) location: compTement (16..21) (D) OTHER INFORMATION! /note- "Xhol restriction site." (ix) FEATURE: (a) name/key! misc feature (B) location: compTement (22..27) (D) OTHER INFORMATION: /note- "EcoRI restriction stie." (ix) FEATURE: (a) name/key! misc feature (B) location: compTement (31..36) (D) OTHER INFORMATION: /note- "Hindlll restriction sits." (ix) FEATURE: (A) NAME/KEY: misc feature (B) LOCATION: compTement (37..42) (D) OTHER INFORMATION: /note- "Ndel restriction site." 29 WO 94/21787 (xi) SEQUENCE DESCRIPTION« SEQ ID NO:3: GAGCTCGGAT CCTATCTCGA OGAATTCTAT AAGCTTCATA TGTACCT PCT/US94/02676 47 (2) INFORMATION FOR SEQ ID NO:4: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 47 baa* pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: both (D) TOPOLOGY: linaar (ii) MOLECULE TYPE: cDNA (ix) FEATURE: (A) NAME/KEY: misc feature (B) LOCATION: compTement (1..43) (D) OTHER INFORMATION: /not*- "Complmentary to SEQ ID NO:03, bases 1 to 43." (ix) FEATURE: (A) NAME/XEY: misc feature (B) LOCATION: compTement (2..7) (D) OTHER INFORMATION: /note- "Ndel restriction site." (ix) FEATURE: (A) NAME/KEY: misc feature (B) LOCATION: compTement (8..13) (D) OTHER INFORMATION: /note- "Hindlll restriction site." (ix) FEATURE: (A) NAME/KEY: misc feature (B) LOCATION: compTement (17..22) (D) OTHER INFORMATION: /note- "EcoRI restriction site." (ix) FEATURE: (A) NAME/KEY: misc feature (B) LOCATION: compTement (23..28) (D) OTHER INFORMATION: /note- "Xhol restriction site." (ix) FEATURE: (A) NAME/KEY: misc feature (B) LOCATION: compTement (32..37) (D) OTHER INFORMATION: /note- "BamHI restriction site." (ix) FEATURE: .. (A) NAME/KEY: misc feature (B) LOCATION: compTsment (38..43) (D) OTHER INFORMATION: /note- "Sacl restriction site." (xi) SEQUENCE DESCRIPTION: SEQ ID NO:4: ACATATGAAG' CTTATAGAAT TCCTCGAGAT AGOATCCGAG CTCOTAC 47 (2) INFORMATION FOR SEQ ID NO:5: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: SO base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: both (D) TOPOLOGY: linear (ii) MOLECULE TYPE: CDNA 30 WO 94/21787 PCT/US94/02676 (ix) FEATUREi (A) NAME/KEY: misc feature (B) LOCATION: compTement (1..46) (D) OTHER INFORMATION: /note- "Complementary to SEQ ID NO:6, bases 1 to 46." (ix) FEATURE: (A) NAME/KEY: misc feature (B) LOCATION: complement (1..6) (D) OTHER INFORMATION: /note- "Hindlll restriction stie." (ix) FEATURE: (A) NAME/KEY: misc feature (B) LOCATION: complement (10..IS) (D) OTHER INFORMATION: /note- "EcoRI restriction sits." (ix) FEATURE: (A) NAME/KEY: misc feature (B) LOCATION: complement (16..21) (0) OTHER INFORMATION: /note- "Kpnl restriction site." (ix) FEATURE: (A) NAME/KEY: misc feature (B) LOCATION: complement (23..28) (D) OTHER INFORMATION: /note- "BariHI restriction site." (ix) FEATURE: (A) NAME/KEY: misc feature (B) LOCATION: compTement (30..35) (D) OTHER INFORMATION: /note- "Sacl restriction site." (ix) FEATURE: (A) NAME/KEY: misc fsature (B) LOCATION: compTement (39..46) (D) OTHER INFORMATION: /note- "NotI restriction site." (xi) SEQUENCE DESCRIPTION: SEQ ID NO:5: AAGCTTATAG AATTCGGTAC CTOGATCCTG AGCTCATAGC GGCCGCAOCT 50 (2) INFORMATION FOR SEQ ID NO:6: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 50 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: both (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (ix) FEATURE: (A) NAME/KEYi misc feature (B) LOCATION: compTement (1..46) (D) OTHER INFORMATION: /note- "Complementary to SEQ ID NO:5, bases 1 to 46." (ix) FEATURE: (A) NAME/KEY: misc fsature (B) LOCATION: compTement (1..8) (D) OTHER INFORMATION: /note- "NotI restriction site." (ix) fZATURE: (A) NAME/KEY: misc feature (B) LOCATION: compTement (12..17) (D) OTHER INFORMATION: /note- "Sacl restriction site.". 31 WO 94/21787 PCT/US94/02676 (ix) FEATURE: (A) NAME/KEY: misc feature (B) LOCATION> compTement (19..24) (D) OTHER INFORMATIONi /note- "BamHI restriction site." (ix) FEATURE! (A) NAME/KEY: misc feature (B) LOCATION: complement (26..31) (D) OTHER INFORMATION: /note- "Kpnl restriction site." (ix) FEATURE: (A) NAME/KEY: misc feature (B) LOCATION: complement (32..37) (D) OTHER INFORMATION: /note- "EcoRI restriction stie." (ix) FEATURE: (A) NAME/KEY: misc feature (B) LOCATION: complement (41..46) (D) OTHER INFORMATION: /note- "Hindu: ..estriction site." (xi) SEQUENCE DESCRIPTION: SEQ ID NO:6: GCGGCCCSCTA TGAGCTCAOO ATCCAGGTAC CGAATTCTAT AAOCTTGTAC 50 (2) INFORMATION FOR SEQ ID NO:7: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 13 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: both (D) TOPOLOGY: linear (ii) MOLECULE TYPE: CDNA (ix) FEATURE: (A) NAME/KEY: misc feature (B) LOCATION: complement (1..13) (D) OTHER INFORMATION: /note- "Complmentary to SEQ ID NO:8, bases 5 to 17." (xi) SEQUENCE DESCRIPTION: SEQ ID NO:7: GGGAAGCCGC CGC 13 (2) INFORMATION FOR SEQ ID NO:8: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 21 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: both (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (ix) FEATURE: (A) NAME/KEY: misc feature (B) LOCATION: compTement (5..17) (D) OTHER INFORMATION: /note- "Complementary to SEQW ID NO;7, bases 1 to 13." (xi) SEQUENCE DESCRIPTION: SEQ ID N0:8i CATGGCGGCG GCTTCCCTGC A 21 32 WO 94/21787 PCT/US94/02676 (2) INFORMATION FOR SEQ ID NO: 9: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 26 bass pairs (B) TYPE: nucleic acid (C) STRANDEDNESSi single (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (xi) SEQUENCE DESCRIPTION! SEQ ID NOi9: CAGGATCCAG CTGTGCCTTC TAGTTG 26 (2) INFORMATION FOR SEQ ID NO:10: (i) SEQUENCE CHARACTERISTICS: (A) LENGTHi 30 baae pairs (B) TYPEs nucleic acid (C) STRANDEDNESSi single (0) TOPOLOGY: linear (ii) MOLECULE TYPE: CDNA (xi) SEQUENCE DESCRIPTION: SEQ ID N0:10: CTGAGCTCTA GACCCATAOA GCCCACCGCA 30 (2) INFORMATION FOR SEQ ID NO:ll: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 16 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: both (D) TOPOLOGY: linear (ii) MOLECULE TYPE: CDNA (ix) FEATURE: (A) NAME/XEY: misc fsature (B) LOCATION: compTement (5..16) (D) OTHER INFORMATION: /note- "Complementary to SEQ ID NO:12, bases S to 16." (xi) SEQUENCE DESCRIPTION: SEQ ID N0:11: AGCTGGAACC CCTTGC 16 (2) INFORMATION FOR SEQ ID NO:12: (i) SEQUENCE CHARACTERISTICS: (A) LENGTH: 16 base pairs (B) TYPE: nucleic acid (C) STRANDEDNESS: both (D) TOPOLOGY: linear (ii) MOLECULE TYPE: cDNA (ix) FEATURE: (A) NAME/XEY: misc feature (B) LOCATION: compTement (5..16) (D) OTHER INFORMATION: /note- "Complementary to SEQ ID NO:11, bases 5 to 16." 33 WO 94/21787 PCT/US94/02676 (xi) SEQUENCE DESCRIPTION: SEQ ID NO:12: GGCCGCAAGG GGTTCC 16 34 </div>

Claims

<div class="application article clearfix printTableText" id="claims"> 263925 WO 94/21787 PCT/US94/0Z<576 WHAT TS CLAIMED TS:

1. A method of introducing a defined deletion at a chromosomal target site in a host cell, said method 5 comprising: (1) transforming viable cells with a DNA targeting construct, said construct comprising: (A) two target DNA sequences, said target sequences being homologous to DNA sequences defining said target 10 site, separated by: (B) a first additional DNA homology (v-ADH) sequence, said first v-ADH sequence being homologous to a first chromosomal ADH sequence (e-ADHJ at a site located at a distance from said target site on the same chromosome, and 15 (C) at least one selectable marker; (2) selecting for transformed cells that express said selectable marker; (3) selecting for transformed cells that lack said selectable marker; 20 whereby homologous recombination occurs at said target site, resulting in integration of said DNA construct, and an intramolecular recombination occurs between said first v-ADH sequence and said first c-ADH sequence, resulting in a chromosomal deletion- between said 25 target site and said first c-ADH site.

2. A method according to Claim l, wherein said at least one selectable marker comprises one marker for positive selection and a different marker for negative 30 selection.

3. a method according to Claim 2, wherein said marker for positive selection is positioned an_juM^M>NA construct between said marker for «Sjgttiva selection^and 35 said v-ADH sequence. \ _ ^ 22 OCT W

4. A method according to Cia|i»-^T^g^«i^5i said | ADH sequence comprises a mammalian repetltlvf^-PNaTsequonce. 35 263 925 WO M/21787 PCT/US94/02676

5. A method according1 to Claim 1, wherein said DNA construct further comprises a second v-ADH sequence, and a second markar for negative selection, said second v-ADH sequence being homologous to a c-ADH sequence at a second 5 c-ADH site located on the same chromosome as said target site, said target site being located between said first c-ADH sita and said second c-ADH site on said chromosome.

6. A method according to Claim 2, wherein said DNA 10 construct further comprises (1 + x) v-ADH additional vADH sequencas, and (l + x) markers for negative selection, said (1 + x) v-ADK sequences being homologous -te-_(l + x) additional c-ADH sequences at (l + x) additional c-ADH sites located in series, each c-ADH sequence,adjacent the 15 next in the series, on one side of said target site on the same chromosome, wherein "x" is any number of additional DNA homology sequences, whereby successive deletions of chromosomal DNA adjacent to the target site are accomplished. 20

7. A method according to Claim 2, wherein said distance between said target site and said c-ADH site is from about 500 bp to about 4000 kb in length. 25

8. A method according to claim 7, wherein included in the chromosomal sequence between said target cite and said v-ADK sequence is a DNA sequence encoding a transcriptional enhancer and/or silencer. 30

9. A method according to claim 2, wherein said vADH sequence is homologous to one of said target sequences, and differs by at least 1 bp.

10. A method according to Claim 1, whereinsaid 35 marker is selected from the group con»«jt^!^^^h^S>ST] minigene, the neo gene, the Hsv thymidine kinase" gene, hygromycin resistance, and HPRT. 1 2.2 t)Gt 1997 36 26 3 9 2 5 WO 94/21787 PCT/US94/02676

11. A method of introducing a delation at a mammalian immunoglobulin locus in a host cell, said method comprising: (1) transforming viable mammalian cells with a DMA 5 construct, said construct comprising: (A) two target DNA sequences, said target sequences being homologous to DMA sequences at the constant region of said mammalian immunoglobulin locus, (B) said target sequences being separated by a v-ADH 10 sequence, said ADH sequence being homologous to a c-ADH DNA sequence at site located at a distance 5' or 3' to said target site on the same chromosome, and (C) a positive selectable marker, and a (D) negative selectable marker, 15 (2) selecting for transformed cells which express said positive selectable marker; (3) selecting for transformed cells which lack said negative selectable marker; whereby homologous recombination occurs at said 20 mammalian immunoglobulin locus, resulting in integration of said DNA construct, and an intramolecular recombination occurs between said v-ADH and c—ADH sequence in said chromosome, resulting in a chromosomal deletion between said target site and said c-ADH site. 25

12. A method according to Claim 11, wherein said immunoglobulin locus is selected from the group consisting of the immunoglobulin heavy and light chain loci. 30

13. A method according to Claim 11, wherein said immunoglobulin locus is a light chain locus.

14. A method according to claim 13, wherein said immunoglobulin locus is a kappa chain locus. 35

15. A method according to Claim nf '"j immunoglobulin locus is a lambda chainj 1 2 2 OCT 1997 37 263925 WO 94/21787 PCTAJS94/02676

16. A ONA construct comprising: (A) two target DNA sequences, said target sequences being homologous to proximal DNA sequences at a target site of a chromosome; said target sequences separated by a 5 (B) v-ADH sequence, said v-ADH sequence being homologous to a c-ADH sequence at a chromosomal site located distal from said target homologous DNA sequences on the same chromosome; and (C) a marker for positive selection; and 10 (D) a marker for negative selection.

17. A DNA construct according to Claim 1&, wherein said v-ADH sequence comprises a mammalian repetitive DNA sequence. 15

18. A DNA construct according to Claim 16, wherein said construct further comprises a second DNA v-ADH sequence, said second v-ADH sequence being homologous to, a c-ADH sequence at a second chromosomal site located on the 20 same chromosome as said target site, said target site being located between said first c-ADH site and said second c-ADH site on said chromosome, and a second marker for negative selection. 25

19. A DNA construct according to claim 16, wherein the chromosomal sequence between said target site and said c-ADB sequence is from about 500 bp to about 4000 kb in length. 30

20. A DNA construct according to Claim 16, wherein the chromosomal sequence between said target site and said v-ADH sequence further comprises DNA sequences encoding a transcriptional enhancer and/or silencer. 35

21. A construct according to Claim 16, wherein salft marker is selected from the group cons LaNfcJEn^A'ESMTttiJFWPBT 22 dfif 199? 263925 WO 94/2TO7 PCTAJS94/02676 minigene, the naa gene, the HSV thymidine kinase gene, hygromycin resistance, and HPRT.

22. A method for inactivating a target locus located 5 on a chromosomt between a target sita and a chromosomal c- ADH DNA sequence at a distance from 3aid target sita on the saiae chromosome comprising: 1) transforming viable mammalian cells with a DNA construct, said construct comprising: 10 (A) two target ONA sequences, said target sequences being homologous to DNA sequences at a target sita on a chromosome; said target sequences separated by a (B) v-ADH sequence, said v-ADH sequence b«5fc§* homologous to a c-ADH sequence at a chromosomal site 15 located distal from said target site on the same chromosome; (C) a marker for positive selection; and (D) a marker for negative selection, (2) selecting for transformed cells that express said 20 positive marker; (3) selecting for transformed cells that lack said negative marker; whereby homologous recombination occurs at said target site, resulting in integration of said DMA 25 construct, and an. intramolecular recombination occurs between said v-ADH sequence and said c-ADH saquence, resulting in a chromosomal deletion of at least a portion of said target locus between 3aid target site and said cADH site, wherein said target locus is inactivated. 30

23. A method according to claim 22, wherein said target locus is the MHC.

24. A method according to Claim 22, wherein said 35 target locus is the /^-microglobulin subunit locus. 263 925 WO 94/21787 PCT/US94/0Z676

25. A method for producing mammalian donor calls lacking MHC class I and/or Class II antigsns for use in transplantation comprising: (1) transforming viable mammalian calls vith a DMA 5 construct for inactivating class I and/or Class IX MHC antigens, said construct comprising: (A) two target DMA sequences/ said target sequences being homologous to DNA sequences within the MHC, (B) said target sequences being separated by a v-ADH 10 sequence,. Mid ADH sequence being homologous to a c-ADH DNA sequence at site loeated at a distance 3' or 3' to said MIIC on the sane chromosome, and (C) a positive selectable marker, and a (D) negative selectable marker, 15 (2} selecting for transformed cells which express said positive selectable marker; (3) selecting for transformed cells which lack said negative selectable marker; whereby homologous recombination occurs within said 20 MHC, resulting in integration of said DNA construct, and an intramolecular recombination occurs between said v-ADH and c-ADH sequence in said chromosome, resulting in a chromosomal deletion between said target site and said cADH site inactivating Class I and/or class II MHC 25 antigens.

26. A method according to Claim 25, wherein said chromosomal deletion comprises the ^-microglobulin locus. 30

27. A mammalian cell that does not express surface MHC antigen produced by the method according to Claim 25.

28. A mammalian cfell according to claim 27, wherein said cell is murine. 35

29. A mammalian cell according to Claim 27, wherein said cell is human- | n.z. patent office 40 2 2 OCT 1997 263 925

30. A method as defined in claim 1 of introducing a defined deletion at a chromosomal target site in a host cell substantially as herein described with reference to any example thereof

31. A method as defined in claim 11 of introducing a deletion at a mammalian immunoglobulin locus in a host cell substantially as herein described with reference to any example thereof.

32. A DNA construct as defined in claim 16 substantially as herein described with reference to any example thereof

33. A method as defined in claim 22 for inactivating a target locus located on a chromosome between a target site and a chromosomal c-ADH -BNA- sequence at a distance from said target site on the same chromosome substantially as herein described with reference to any example thereof.

34. A method as defined in claim 25 for producing mammalian donor cells lacking MHC Class I and/or Class II antigens for use in transplantation substantially as herein described with reference to any example thereof.

35. A normal mammalian cell as defined in claim 27 that does not express surface MHC antigen substantially as herein described with reference to any example thereof end of claims n.z. t 22 OCT 1997 received 41 </div>