WO1998036082A1 - Yeast artificial chromosomes for studying aberrant proto-oncogene transcription - Google Patents

Abstract

The present invention relates to a yeast artificial chromosome (YAC) which comprises a gene activation element and a proto-oncogene or oncogene which is activated by the gene activation element. In particular, the gene activation element is, or is from, an immunoglobulin locus. The invention also relates to cells which have the YAC integrated into their genome, chimeric or transgenic animals derived from such cells and the use of such YACs, cells and animals to investigate the role of tumour formation, in particular B-cell lymphomas.

Description

YEAST ARTIFICIAL CHROMOSOMES FOR STUDYING ABERRANTPROTO-ONCOGENE TRANSCRIPΗON

The present invention relates to a yeast artificial chromosome (YAC) which comprises a gene activation element and a proto-oncogene or oncogene which is activated by the gene activation element. In particular, the gene activation element is, or is from, an immunoglobulin locus. The invention also relates to cells which have the YAC integrated into their genome, chimeric or transgenic animals derived from such cells and the use of such YACs, cells and animals to investigate the role of tumour formation, in particular B-cell lymphomas .

Activation of a proto-oncogene or oncogene is a consistant feature of B-cell lymphomas. More specifically, activation of a proto-oncogene by one of the immunoglobulin (Ig) loci after chromosomal translocation is a consistent feature of Burkitt ' s lymphoma. Different subtypes of this tumour vary in the molecular architecture of the translocation region. In most cases there are no known regulatory elements of the lg locus neighbouring the oncogene and this considerably obscures the mechanism of its deregulation.

Chromosomal translocations by which the c-myc proto- oncogene on region 8q24 is juxtaposed to one of the immunoglobulin (Ig) gene loci on chromosomes 14q32

(IgH) , 2pll (lgκ) . or 22qll (Igλ) are consistently found in Burkit ' s lymphomas. The translocated and often mutated c-myc gene is expressed and deregulated at various levels - these include alternative promoters, transcription elongation and mR A stability - whereas the non-translocated allele in the tumour cells is usually transcriptionally silent (Spencer and Groudine, 1991) . In various cell systems, expression of the c-myc proto-oncogene has been shown to be tightly linked to cellular growth, differentiation and/or apoptosis (Yokoyama and Imamoto, 1987; Hoffman-Liebermann and Liebermann, 1991; Evan and ittlewood, 1993; Milner et al . , 1993; Cherney et al . , 1994) . Exogenous expression of c-myc in transgenic mice can induce neoplastic proliferation of different types of cells depending on the regulatory unit, made up by promoter and enhancer combinations, which drives transgene expression (Stewart et al . , 1984, Schoenenberger et al . , 1988; Spanopoulou et al . , 1989; Sandgren et al . , 1991). ymphoid tumours have been found in mice carrying various minigene constructs with a c-πiyc gene under the control of a murine or human IgH intron enhancer (Adams et al . , 1985; Suda et al . , 1987; Schmidt et al . , 1988; Yukawa et al . , 1989) . Although the transgene was present in up to 40 copies per genome, (Suda et al . , 1987) these results supported the notion that deregulation of c-myc in Burkitt ' s lymphoma is achieved by Ig gene sequences which subsequently control a critical step in tumorigenesis . In these constructs, the c-myc gene is necessarily (by virtue of the restricted size of the minigene) in close proximity to the intron enhancer. Such constructs therefore do not represent situations where the intron enhancer may be at a much further distance from the activated gene (such as in Burkitt ' s lymphoma) . Furthermore, the required multiple copy numbers of the minigene and their known delay in tumour production cannot be used to accurately investigate the timing and regulation of tumour development.

In Burkitt ' s lymphoma cells, however, translocation of c -myc into the vicinity of a known Ig enhancer have rarely been documented (Hayday et al . , 1984). Possible regulatory elements in the Igκ light chain locus which could be responsible for c-myc deregulation in t(2;8) translocations have been defined in vi tro (Hόrtnagel et al . , 1995; Mautner et al . , 1996), but their importance for tumour formation in vivo has still to be shown. The situation in t(8;14) translocations, being predominant in Burkitt ' s lymphoma, and concerning oncogene activation by the heavy chain locus, is even less clear and this has not been possible to investigate in detail .

On the molecular level, the breakpoints on both chromosomes involved vary considerably between individual cases of Burkitt ' s lymphoma (Croce, 1993). Generally, two constellations of the IgH/c-τπyc translocation region are found in different tumours, with their relative abundance in the population varying between different geographic regions (Shiramizu et al . , 1991; Gutierrez et al . , 1992). The endemic form involves J (joining) or D (diversity) region breakpoints and in the sporadic form different switch region breakpoints are found whilst breakpoints in the c-myc locus can be located a large distance upstream of the oncogene or can lead to the deletion of exon 1, respectively. As a result, in the vast majority of Burkitt ' s lymphoma with t(8;14) translocations, the IgH intron enhancer is either located a long distance upstream of the c-myc gene (usually more than 100 kb) or is missing altogether from the translocation product carrying the oncogene (Rabbitts and Boehm, 1991) .

In vi tro assays to measure the activity of the IgH intron enhancer coupled to a c -myc promoter have been performed with a large number of Burkitt ' s lymphoma cell lines (Jain et al . , 1993). Significant transcriptional differences were revealed between those cell lines, not correlated with the location of the endogenous translocation breakpoints, but in accordance with the suggestion that additional elements of the IgH locus are involved in c-myc deregulation. Various enhancer elements have been discovered at the 3 ' end of the IgH locus in mice (Dariavach et al . , 1991; Lieberson et al . , 1991, Matthias and Baltimore, 1993; Michaelson et al . , 1995) which in combination can effectively deregulate an adjacent c-myc promoter after transfection into Burkitt ' s lymphoma cell lines (Madisen and Groudine, 1994) . Nevertheless, it has not yet been possible to investigate the potential involvement of such crucial regulatory sequences in tumour formation in animals, nor indeed have equivalent control elements been identified in the human IgH locus .

Expression studies of large gene loci in, vivo have been facilitated by techniques to transfer yeast artificial chromosomes (YACs) into mouse embryonic stem (ES) cells or oocytes . Transloci of up to 1300 kb have been introduced into the mouse germline by spheroplast fusion of YAC containing yeast cells with ES cells and have resulted in the tissue-specific rearrangement and expression of human Ig light chain genes in mice (Zou et al . , 1996). Appropriate developmental control of the human β-globin locus present on a YAC transgene has also been reported (Gaensler et al . , 1993). Site-specific modifications of the locus have identified the function of specific regulatory sequences (Bungert et al . , 1995) and have led to the production of developmental mutants with medical implications (Peterson et al . , 1995) .

In spite of the molecular characterisation of many different chromosomal breakpoints found in Burkitt ' s lymphomas, it is still unclear which element or elements of an Ig locus is, or are, responsible for the deregulation of the translocated c-myc gene. In particular, the large distance between the oncogene and putative regulatory sequences hampers the in vivo analysis. To overcome problems concerned with manipulation and transfer of large regions, we have used YAC technology to develop a strategy for introducing a modified YAC into embryonic stem (ES) cells. These cells, when injected into blastocysts, can contribute to the development of a mouse. This is indicated initially in the coat colour contribution of the animal. Such chimeric animals are useful as a model for the in vivo study of oncogene activation, in particular after chromosomal translocation. Tumour tissue (or other transformed tissue) from such animals may be used for the in vi tro study of oncogene activation.

According to a first aspect of the present invention, there is provided a YAC which comprises a gene activation element and a proto-oncogene or oncogene which is activated by the gene activation element.

YACs are well known in the art and any can be used according to the present invention. The gene activation element carried by the YAC may be any region, coding or non-coding which activates a proto-oncogene or oncogene also carried on the YAC. It may comprise a known gene activation element, region or locus, including an enhancer or may comprise an element which was not previously known to have such an * activation' capacity.

It may include any element, region or locus inherently having such "activating' capacity. The proto-oncogene or oncogene may be expressed and/or deregulated at various levels by the gene activation element and these include alternative promoters, transcription elongation and mRNA stability.

The gene activation element is preferably an immunoglobulin locus or a region or locus involved in translocation events and can include: a heavy chain or light chain (kappa or lambda) locus or a part thereof; a variable or constant locus or a part thereof; an enhancer or activator element, in particular an Ig intron enhancer or Ig 5 ' or Ig 3 ' enhancer defined by the location of the gene transcription elements. The gene activation element may be in germ line configuration. By definition, if a proto-oncogene or an oncogene in a YAC construct (according to the present invention) is activated to produce tumours, in particular a B-cell lymphoma, then a gene activation element must be present.

The proto-oncogene or oncogene is any gene which, when expressed, can lead to the formation of a tumour and includes genes encoding carcinomas as well as benign and malignant tumours. The proto-oncogene or oncogene may, or may not, be mutated or otherwise modified.

A preferred gene according to the invention is the proto- oncogene c-myc, but any proto-oncogene or oncogene is covered by the present invention in particular the genes: svc, other myc genes, mos, fos, sis, fms, fes, fps, abl, rel, erb, myb, BRAC1, BRAC2 , src, yes, fgr, ros, ski, erbA, erbB, kit, sea, crk, mat, cbl, jun, rat, mil, Ha-ras, Ki-ras, ets, st, N-myc, L- yc, neu, N-ras, dbl, K-fgf (= hst) , mcf2, mc£3, met, onc-D, ras related, Tlym-1, -2, trk, tx-1, tx-2, tx-3, tx-4, bcl-1, bcl-2, bcr, tcl-1, T^S-1, int-1, int-2, Mlvi-1, -2, -3, Pim-1, Pvt-2, RMO-int-1, fim-1, -3, evi - 1, -2, Spi-1, fis-1, c-abl, c-myb, c-erbB, c-Ki-ras-2 , N-ras, arg, fyn, hck, lyk, lyn, tkr.

The gene activation element and/or the proto-oncogene or oncogene may be derived from any mammal, in particular humans, mice or rodents, or may be bovine, ovine or porcine derived. It may be separated from the proto- oncogene or oncogene by any distance, for example more than lOkb or even more than 50kb away.

The gene activation element may be modified and/or mutated such that activation of the proto-oncogene or oncogene is altered. In particular the gene activation element may be modified and/or mutated such that activation of the proto-oncogene or oncogene is decreased or eliminated. The modification or mutation may be by rearrangement of the gene activation element

(eg. an Ig locus) or deletion of various portions of the gene activation element. In this way, the regulation of tumour formation, brought about by expression of the proto-oncogene or oncogene can be investigated. Thus, a YAC according to the first aspect of the invention provides an excellent model system for investigating the regulation of tumour formation, in particular the regulation of B-cell lymphomas. B-cell lymphomas according to the present invention include precursor B- ly phoblastic leukemia/lymphoma, B-cell chronic lymphocytic leukemia/prolymphocytic leukemia/small lymphocytic lymphoma, lymphoplasmacytoid lymphoma/immunocytoma, mantle cell lymphoma, follicle conter lymphoma (follicular) , marginal zone B-cell lymphoma, splenic marginal zone lymphoma, hairy cell leukemia, plasmacytoma/plasma cell myeloma, diffuse large B-cell lymphoma, Burkitt' s lymphoma and high-grade B-cell lymphoma. Since some lympohomas are actually a human disease, a non-human model of the system should correctly be described as a disease-like lymphoma, for example Burkitt 's lymphoma. In this text, a non-human model of a human disease is simply referred to as its usual human disease name.

According to a second aspect of the invention there is provided a process for the transformation of cells by a YAC, (as described according to the first aspect of the invention) . The process may comprise any of the following standard techniques well known in the art: protoplast fusion using yeast spheroplasts ; micro- injection and co-precipitation with calcium phosphate; electroporation; or lipofection treatment.

The YACs may be transfected into any animal cell, preferably a mammalian cell. The YAC is preferably integrated into the genome of the host cell . The host cell may be human, murine, rodent, porcine, bovine or ovine. Most preferably the host cell is one which can be developed into a non-human animal, eg. embryonic stem cells. In such a manner the cells containing a YAC can be introduced into a non-human animal in which they can contribute to the various tissues, including germ cells.

The third aspect of the invention thus provides a host cell which comprises integrated into its genome, a YAC - according to the first aspect of the invention. Such host cells may be used to produce chimeric tissue or a chimeric or transgenic animal .

The present invention also includes, according to a fourth aspect, a process for producing chimeric tissue or a chimeric or transgenic animal from a host cell as per the third aspect of the invention. The process preferably produces tumour cells. Such tumour cells can be cultivated in vi tro, using standard cell culture techniques. Such cell cultures enable a further in vi tro model for tumour formation, control, etc as per the other aspects of the invention. The chimeric tissue material or a chimeric or transgenic animal produced is a fifth aspect of the present invention. The chimeric material refers to transgenic material which is from a chimeric animal or transgenic material from a transgenic animal. Such transgenic material or tissue is preferably tumour tissue. The production of such chimeric material (usually in tissue culture) or such an animal is by techniques well known in the art. A particularly preferred method for producing chimeric mice is by injecting embryonic stem cells according to the third aspect of the invention into a mouse blastocyst. The blastocyst is then developed to term according to standard procedures known in the art (Hogen et al . , 1994). Transgenic animals can also be produced according to techniques in the art, including nuclear transfer as described in O97/07669 and WO97/07668. The choice of producing a transgenic or a chimeric animal will usually reside in the length of time the animal takes to produce tumours and the length of time the animal is able to survive once tumours are produced. A suitable animal model may have a greater life span and be of more use if it is chimeric. However, a particularly slow tumour production model may be better if expresses from a transgenic animal.

A sixth aspect of the invention provides for the use of

YACs according to the first aspect of the invention, the use of host cells according to the third aspect of the invention, or a chimeric animal or tissue according to the fifth aspect of the invention in order to investigate the role of a distantly situated genetic control element or any other aspect of the formation. This can be done by monitoring the production and/or formation of tumours in relation to the gene activation locus, including modifications and/or mutations of the gene activation locus.

The various aspects of the invention may also be used, in a similar way, for the in vi tro or in vivo analysis of proto-oncogene or oncogene activation and/or tumour formation or for the identification of a tumour suppressing element. Cells from the tumours may be propagated in tissue culture, enabling in vi tro studies to be carried out of the effects of various treatments, e.g. drugs, chemicals, antibodies, on tumour cell growth.

Such a use may also involve in vivo analysis on chimeric tissue, including chimeric animals. The tumour suppressing elements may be a compound or composition administered to the tissue or animal which suppresses or regulates the development or formation of tumours . Alternatively the tumour suppressing element may be a modified or mutated gene activation element or region.

The present invention also covers, according to a seventh aspect, a tumour suppressing element identified as above .

The present invention is described with reference to the attached drawings, as follows:

Figure 1: a schematic representation of the human IgH/c- myc YAC. The 220 kb IgH insert depicted by a thin line contains 5 V_Hgenes (V_H2-5, V_H4-4, V_Hl-3, V_Hl-2 and V_H6-1) , the D cluster (Dl - D4 and D_HQ52 are marked) , the 6 J_H segments, the enhancer (Eμ) marked by a circle and Cμ and Cδ shown by boxes . The YAC vector arms are represented by a thick line with the arrowheads marking the telomeres, with c-myc and 2 copies of the neo^r gene in the right arm shown by boxes . Genes and gene segments are shown without their exon/intron structure. The transcriptional orientation of the c-myc gene is indicated by an arrow. The fragments of the human IgH locus used as hybridization probes (a, V_H6-1; b, Eμ; c, Cμ; d, δ M2 ; e, 3 ' IgH) and the restriction fragments in kb highlighted by these probes are shown. Only Sail or Spel restriction sites relevant for this study and BamHI or Hindlll sites in the J_H-Cμ region are shown. (B, BamHI; H, HindiII; P, Spel; S, Sail).

Figure 2: Site-specific integration of pLNA (8.9 kb) and pRBura33 (12.3kb) into the acentric arm of the HuIgH YAC. (A) Schematic representation of the homologous recombination events. The right YAC arm is shown as a hatched box with the telomere marked as an arrowhead and the insert represented as a wavy line. The c-myc exons are depicted as numbered boxes and the chromosomal breakpoint (b/p) adjacent to the Ig switch sequence

(Sμ/Sγ) is marked. URA3 and LYS2 are the selectable marker genes for yeast and neo^r refers to the tk-neo^r gene cassette. The sizes of fragments hybridizing to the c-myc or neo probe are given in kb (R, EcoRI) . (B) Southern blot analysis of yeast clone Nl containing the IgH/c-myc YAC (IgH/c-myc) , normal yeast cells (AB1380) and yeast cells harbouring the HuIgH YAC (HuIgH) .

Undigested DNA from the yeast clones was separated by pulsed-field gel electrophoresis (25 sec switch time for 32 hrs at 180 V and 3.0°C) whilst conventional TAE gels where used for separation of EcoRI and Sail digests. Filters were hybridized to c-myc, neo, or the human Eμ probe as indicated.

Figure 3 : Integration of the human IgH/c-myc YAC into the ES cell genome. DNA from the IgH/c-myc YAC containing yeast clone Nl , the ES cell clones Nl-21, Nl- 22, and Nl-25 and HM-1 ES cells was digested with Hindi11 (panel Eμ) or EcoRI (all other panels) . Yeast DNA of 2.5 x 10⁵ cells (digested in agarose blocks) or 6μg ES cell DNA (equals 10⁶ diploid genomes) was applied per lane. Hybridization to specific probes, c-myc, human Eμ and amp, or total yeast DNA, is indicated in bold and sizes are in kb. Note that hybridization with total yeast DNA also detects cross-hybridizing fragments (Davies et al . , 1993) highlighted in all ES cell clones including unmanipulated HM-1 cells.

Figure 4 : Flow cytometry analyses to characterise tumour cells. (A) Cells derived from tumours of IgH/c-myc chimeras 21-3, 21-4, 22-1 or 22-2 were stained with anti-CD45R(B220) and anti-mouse IgM. The location of the tumour samples shown is given in brackets (a, abdomen; h, head) . T indicates primary tumour tissue and C the cell line established from the tumour. (B) Cytograms of spleen, thymus and blood cells from IgHJ/c- myc chimeras (left, spleen and thymus from chimera 22-2 and blood cells from chimera 21-4) compared to normal tissues from control BALB/c mice (right) . The analyses were performed on total (ungated) cell populations.

Figure 5: Configuration of the endogenous IgH locus. Rearrangement of the mouse heavy chain genes in spleen or tumours from chimeras 21-5 and 22-1 was identified in EcoRI digests after hybridization with the mouse 3 ' J_H probe. The 6.5 kb germline fragment, obtained from Nl- 22 ES cells as well as BALB/c liver, is marked by an arrowhead. Tumour samples (T, tumour tissue; C, cultured cells derived therefrom) are designated by their origin of head [tu(h)] or abdomen [tu(a)].

Figure 6: Integrity of the IgH/c-myc translocus in different tumours. Southern blot hybridizations were performed with tumour DNA of chimeras 21-5, 21-3, and 22-1. DNA was obtained from primary tumour tissue (21-5 in all panels and 21-3 and 22-1 in panels A-C) or cultivated cells derived thereof (21-3 and 22-1 in panels D-F) . DNA from HM-1 ES cells, normal mouse liver and ES clone Nl-21 served as controls. The DNA was digested with EcoRI (A), BamHI and HindiII (B,C) or Spel

(D-F) and hybridized to the human probes notes below the panels with the molecular weight in kb. Separation was on conventional TAE gels except for panel E where pulsed-field gel electrophoresis (5 sec switch time for 27 hrs followed by 10 sec switch time for 6 hrs at 140 V and 120°C) was performed. Germline or unaltered transgene configurations are indicated by arrowheads.

Figure 7: FACS analysis of day 17 fetal liver cells from normal mice (left) and IgH/c-myc tumour mice (right) . Gated populations for analysis are: large IgM+ cells (top panels) , CD25+ cells (middle panels) and B220+ cells (bottom panels) .

Figure 8: FACS analysis of cells from bone marrow (BM) , blood, spleen and Peyer' s patches (PP) of normal and chimeric IgH/c-myc mice. The analysis shown was carried out on mice at 9 weeks of age. FL1 = staining with anti-mouse IgM; FSC-H = forward scatter, demonstrating the presence of abnormal populations comprising large IgM+ cells in the chimeric animals. The Table shows the percentage of abnormal cells in various organs at ages from 10 days to 9 weeks.

Figure 9: FACS analysis of tumour cell phenotype . The characteristics of the large IgM+ cells (left panel) are analysed by staining for B220, IgM (μ) , CD43 and IgD

(δ) . The correlation of the staining pattern with stages in B cell development (pre B-cell to mature B- cell) is shown.

Figure 10 : Molecular analysis of mouse Ig gene rearrangement in relation to tumour development. Each individual rearrangement is represented by a different subscrip symbol. (H = mouse heavy chain locus, K = mouse kappa light chain locus) . Hashed cells represents tumour commitment, non-hashed cells tumour establishment. Rearrangement at the mouse heavy chain locus is invariably present in the tumour cells, whereas rearrangement at the kappa locus is sometimes, but not always , present .

Figure 11: Southern blot analysis of rearrangement of the mouse Ig loci in tumours. Mouse H, K = heavy and light chain loci; Tumour A, H = abdomen and head. In each case, results for two mice are presented (out of many analysed) . For tumours present in the same mouse, the same H locus rearrangement is seen, with a variable second H chain or kappa rearrangement . The rearrangement of the human Ig YAC occurs later and is also variable (see e.g. mouse 2 of the right hand panel) . Mouse JH and Jk probes were used.

Figure 12: Southern blot analysis of rearrangement of mouse heavy (H) and light (K) chain loci in cloned tumour cells obtained from the IgH/c-myc chimeric mice and cultivated in vi tro . This confirms that the same H rearrangement is present in each clone, but rearrangement of the K light chain locus is variable. T = original tumour cells, Cl C2 = clones.

In order to assess possible oncogene activation signals, we mimicked a translocation region by modifying a yeast artificial chromosome (YAC) containing a 220 kb region of the human Ig heavy chain (IgH) locus by insertion of a c-myc gene about 50 kb from the IgH intron enhancer. Single copy integration of this YAC into the genome of mouse embryonic stem (ES) cells was achieved by spheroplast fusion. Chimeric mice derived from these ES cells developed monoclonal B-cell lymphomas expressing surface IgM by 4-16 weeks of age. The IgH/c-myc translocus showed different V_HDJ_H rearrangement in almost all tumours without any alterations of the distance between c-myc and the IgH intron enhancer. This mouse model can be used for the in vivo analysis of c -myc deregulation and the tumour formation capacity of the IgH/locus in aberrant rearrangements.

The activation of a c-myc gene linked to an Ig enhancer on a minigene construct has been extensively studied in transgenic mice produced by injection of DNA into oocytes (Adams et al . , 1985; Suda et al . , 1987 Schmidt et al 1988; Yukawa et al . , 1989). This technique usually results in tandem integration of multiple copies of the transgene into the genome as described for IgH intron enhancer/c-myc transgenic mice by Suda et al . ,

(1987) . Although the expression levels of small transgenes do not always correlate with their copy number, a high copy number of a c-myc transgene might nevertheless have a stimulating effect on tumour induction since c-myc amplification has been associated with a variety of human tumours (Wong et al . , 1986; Escot et al . , 1986; Brison, 1993). However, these problems do not arise in our transgenic mouse model, as single integration of the IgH/c-myc YAC into the ES cell genome was sufficient for early tumour formation in chimeric mice. This also circumvented the time- consuming task of obtaining germline transmission. Tumour formation was successful with two independently derived ES cell clones which indicated that the integration site of the transgene did not play a critical role in tumour formation. B-cell lymphomas developed early in adulthood in the vast majority of the chimeric mice derived from the IgH/c-myc ES cells whilst none of the 35 chimeras carrying the HuIgH YAC (without c-myc) as a transgene developed tumours, nor were tumours found after germline transmission of this YAC. As the c-myc transgene on its own does not provoke any tumour formation as shown by ourselves and other (Adams et al . , 1985) we conclude that c-myc activation by Ig gene sequences present on the IgH/c-myc YAC gives rise to B lymphoid tumours in our chimeric mice .

The human c-myc gene in the experiments was derived from the translocated allele in the Burkitt ' s lymphoma cell line Raji, where no Ig enhancer element is located adjacent to the oncogene (Rabbitts et al . , 1983). Therefore the only known Ig enhancer present on the IgH/c-myc translocus is the intron enhancer located about 50 kb upstream of the introduced c-myc gene. This distance is not altered in the lymphoid tumours although the translocus underwent V_HDJ_H rearrangement in most cases . The latter may indicate an open chromatin structure of the integrated translocus which could support c-myc activation and is analogous to Burkitt ' s lymphoma where the oncogene is also activated by a rearranged Ig locus. However, in the tumour of one chimeric mouse no alteration of the translocus was found. This may indicate that linkage of the oncogene to the heavy chain locus even in the absence of rearrangement is sufficient to drive c-myc activation in

B-lymphocytes . A long-distance influence of another enhancer at the 3 ' end of the mouse IgH locus has been proposed after its targeted deletion from the mouse genome (Cogne et al . , 1994) . In chimeric mice which carry a homozygous 3' enhancer deletion in all B-cells, several constant region genes located up to 100 kb upstream of the deletion seemed to be affected. This was evident from the deficiency of certain Ig isotypes in serum and a lack of germline transcripts of at least some of these genes which could not be induced in vi tro .

Recently, transfection experiments using lymphoid cell lines showed that Igκ enhancer elements can influence the expression of a c-myc gene located about 30 kb away (Mautner et al . , 1996), thereby supporting the view that large linear distances may not prevent enhancer function. However, it still needs to be shown whether this can be achieved in transgenic animals, as conflicting results between c-myc gene expression in vi tro and in vivo have been reported (Morello et al . , 1993; Lavenu et al . , 1994).

Our transgenic mouse model for the derivation and investigation of tumours differs in at least two aspects from tumour formation provoked by Eμ/ c -myc minigene constructs. First, the tumours in such mice typically involve the lymphatic system with swelling of the lymph nodes being the first sign of the illness (Adams et al . , 1985; Schmidt et al . , 1988; Suda et al . , 1987). Thymomas and other tumours located in the chest were also reported and 25% of the animals had tumour masses inside the skull adherent to the bone (Harris et al . , 1988a) . The latter feature was found in all of our tumour-bearing mice leading to a deformation of the head. Separate abdominal tumour masses and occasionally tumours in the chest were also found but never involved any lymph nodes. Second, the tumours of mice carrying

Eμ/c-myc minigenes represented different stages of B- cell development. Plasmacytomas could be induced in these mice either by introduction of an additional transgene like Eμ/v-abl (Rosenbaum et al . , 1990) or by treatment with pristane (Harris et al . , 1988b) but without any further manipulation pre B-cell stage tumours were predominant (Harris et al . , 1988b; Schmidt et al . , 1988; Suda et al . , 1987). The tumours induced by the IgH/c-myc YAC, however, uniformly consisted of sIgM⁺ B cells without exception. Interestingly,

Burkitt ' s lymphoma cells are usually also slg⁺ (typically IgM, Gunven et al . , 1987) and the tumours are formed primarily at extranodal sites (for review see Magrath,

1990) . Their predominant location seems to be dependent on different factors such as age or geographic and socioeconomic origin of patients and can include jaws (Burkitt, 1958) , probably associated with developing teeth, and abdominal sites (Gutierrez et al . , 1992). Tumour deposits in the central nervous system have also been reported (Gutierrez et al . , 1992). Therefore in terms of the location and the developmental stage of the tumour cells, the IgH/c-myc induced tumours presented here seem to be more closely related to Burkitt ' s lymphoma than the lymphoid tumours provoked by small

Eμ/c-myc transgenes . Several reasons for the phenotypic differences between the transgenic mouse models can be envisaged: (1) The different mouse strains used (BALB/c in our study, C57BL/6, C3H/HeJ, or different F₂ mice in the other studies) may cause different tumour phenotypes . An influence of the genetic background on the average lifespan of Eμ/c- yc transgenic mice (Sidman et al . , 1988) or the phenotype of the lymphomas they develop, B-cell versus T-cell (Yukawa et al . , 1989), has been reported; (2) The c-myc genes used in all studies differ from each other in terms of species (mouse/human) and the presence of mutations. Although unlikely, we cannot exclude that a mutated c-myc gene from a Burkitt ' s lymphoma cell line, as used in our YAC, initiates tumour formation of a Burkitt ' s like phenotype more easily than other c-myc genes. However, tumour activation seems to be independent of oncogene orientation after translocation. The transcriptional orientation of c-myc with respect to the Ig locus is different in IgH and IgL translocations (Rabbitts and Boehm, 1991) , presumably guided by maintaining the centromere, and in transgenic mouse lines tumours were obtained with c-myc adjacent to Eμ in either orientation

(Suda et al . , 1987; Yukawa et al . , 1989). (3) A critical point is that the differences in tumour phenotype may be caused by the large Ig gene region linked to c-myc. The predominance of preB-cell tumours in Eμ-myc minigene mice is likely to be due to the activity of Eμ early in B cell development (Gerster et al . , 1986) . In our mouse model, the distance of Eμ and c-myc and/or not yet identified control sequences in the Ig gene region might modulate Eμ enhancer activity and, thus, restrict the activation of c-myc in a different subset of B lymphocytes. Alternatively, a regulatory element different from Eμ but located in cis on the transgenic Ig gene locus might be responsible for c - yc activation. It also remains possible that c-myc activation and tumourigenesis starts during an early stage of B cell development and that the large Ig gene region favours further differentiation of the neoplastic cells to slgM positive B-cells. For tumour cells developed in Eμ/c -myc mice, a rare example for such a progression from preB to B-cell stage has been reported (Adams et al . , 1985).

A different approach to model chromosomal translocations in transgenic mice by use of the cre/loxP system has been reported recently (Smith et al . , 1995). A t(12;15) translocation was achieved in ES cells although no tumour induction in mice has yet been shown. Compared to this strategy, the introduction of translocation YACs into mice has the advantage that the methodology to modify YACs is now well established (Spencer et al . , 1993; Mckee-Johnson and Reeves, 1996; Duff and Huxley, 1996) and allows the identification of the function of control sequences on large gene loci. For this, our emphasis is on deletion strategies to identify the function of the Eμ enhancer in oncogene activation. By transferring modified translocation YACs into the mouse genome and monitoring tumour formation we will be able not only to clarify the role of distantly located control elements, but also to use this mouse model for identifying efficient strategies to suppress tumour formation.

EXAMPLES

Materials and Methods Construction of the human IgH/c-myc YAC

The HuIgH YAC contains a 200 kb human IgH region in germline configuration accommodating 5 V_H genes, all D segments, the J_H cluster, Cμ and C,§ (Wagner et al . , 1996; Brϋggemann and Neuberger, 1996). The yeast URA3 gene on a 1.1 kb Hindlll fragment (Rose et al . , 1984) was cloned into the HindiII site of pUCRB19RH7 (Rabbitts et al . , 1983; Hamlyn and Rabbitts, 1983) to obtain the 12.3 kb plasmid pRBura3 with URA3 , c-myc and the ampicillin resistance gene in the same transcriptional orientation. Co-transformation of spheroplasted yeast cells AB1380 (Burke et al . , 1987) containing the HuIgH YAC with pRBura3, linearised with Ncol, and pLNA (Davies et al . , 1996) .

Cell cul ture and mice

HM-1 ES cells (Selfridge et al . , 1992), were cultured under standard conditions (Hogan et al . , 1994) on mitotically inactivated SNL fibroblasts (McMahon and Bradley, 1990) . YAC containing yeast was spheroplasted with zymolyase-20T (ICN) and fused with ES cells using PEG 1500 (Boehringer Mannheim) as described (Davies et al . , 1996). ES cell transformants were selected in medium containing 200 μg/ml G418 (Gibco) , picked after 9-12 days and expanded for analysis. ES cells were injected into BALB/c blastocysts and reimplanted into foster animals as described (Hogan et al . , 1994) .

In order to establish cell lines, tumour tissue was disaggregated by pushing it through a 70 μm nylon mesh

(cell strainer, Falcon) and the cells were cultivated in 96 well plates on mitomycin C treated SNL fibroblast feeders in RPMl 1640 medium (ICN) supplemented with 15% fetal calf serum (FCS) , 50 μM β-mercaptoethanol , and 50 μg/ml gentamicin. After 2-3 weeks cultures were expanded and used for further analyses .

FACS analysis

For two-colour flow cytometry on a FACScan (Becton- Dickinson) phycoerythrin (PE) -conjugated anti CD45R(B220) (Sigma) and fluorescein (FITC) labelled anti-mouse IgM (Zymed) were used. Tissue samples from sacrificed animals were pushed through a nylon mesh to obtain single cell suspensions, washed with 0.8% NH₄C1 and sedimented through a FCS cushion. Cultivated tumour cells were detached from the feeder cells by treatment with phosphate-buffered saline (PBS) containing 0.2 mg/ml EDTA for 5 min. The cells were preincubated in DMEM supplemented with 10% FCS for 10 min, incubated with staining antibodies in PBS/2% FCS for 30 min and washed with PBS/2% FCS before the analysis.

DNA hybridization analysis

Genomic DNA in solution was prepared, digested and separated in agarose gels by standard procedures (Sambrook et al . , 1989). Yeast DNA was routinely prepared in agarose (Seaplaque FMC) plugs (Davies et al . , 1996) at 5xl0⁷ cells/100 μl and mammalian cells were embedded in agarose at a concentration of 5x10^s cells/100 μl . Lambda ladder (New England Biolabs) was used as size marker for pulsed-field gel electrophoresis (1% agarose gels in 0.5xTBE) using a LKB Pulsaphor 2015 equipment. For conventional agarose gel electrophoresis yeast DNA plugs were melted at 60°C before loading. DNA fragments were transferred to nylon membranes (Hybond N, Amersham) and hybridized with ³²P oligolabelled probes at 65°C in Church buffer. The following human specific probes were used: c-myc, a 1.2 kb Smal fragment from pUCRB19RH7 (Rabbitts et al . , 1983) comprising exon 1, intron 1 and part of exon 2; V_H6-1, a 0.8 kb EcoRI fragment from pUCRI 0.5/2 (Buluwela and Rabbitts, 1988); Eμ, a 0.8 kb Bglll-Hindlll fragment from the human J_H-Cμ intron (see map in Hayday et al . , 1984); Cμ, a 1.2 kb EcoRI fragment from λC75 (Milstein et al . , 1984); δM2 , a

0.35 kb PCR fragment from the human Cδ gene (Wagner et al . , 1996); and 3'IgH, a 0.4 kb EcoRI fragment from plasmid pM5-l-23 (which contains 3.4 kb of DNA adjacent to the acentric arm of the HuIgH YAC obtained by plasmid rescue; C. Mundt, unpublished) . The mouse 3 ' J_H probe was a 0.7 kb EcoRI -Xbal fragment, covering the mouse IgH intron enhancer (Gillies et al . , 1983; Banerj i et al . , 1983) . The neo probe was a 1.1 kb BamHI fragment from pLUNA (Davies et al . , 1992) and the amp probe was a 1.5 kb PvuII-Scal fragment from pBluescript (Stratagene) . Total yeast DNA used as probe was isolated from S . cerevisiae strain AB1380 and digested with Sau3A before labelling.

Example 1

Construction of the human IgH/c-myc YAC and transfer into ES cells

The HuIgH YAC contains an authentic 220 kb region of the human heavy chain locus in germline configuration, including 5 variable (V) region gene segments, the diversity (D) segment cluster, the joining (J) segments and the μ and δ constant (C) region genes (Brϋggemann and Neuberger, 1996) . The c- yc gene on a 8.6 kb EcoRI - Hindlll fragment including all 3 exons with adjacent Ig switch region sequences was derived from the Burkitt ' s lymphoma cell line Raj i (Hamlyn and Rabbitts, 1983). In order to create a translocation region (Figure 1) the c-myc gene and selectable marker genes for ES cells (neo^r) and for yeast (LYS2) were introduced site- specifically into the acentric arm of the HuIgH YAC. As shown in Figure 2A this was performed by co- transformation of YAC containing yeast with plasmids pRBura3 and pLNA (see Material and Methods) . Yeast clones growing in lysine-deficient medium were screened by Southern hybridization for homologous integration of a single copy of pRBura3 and tandem copies of pLNA which are needed for efficient transfer of the YAC into ES cells (Davies et al . , 1996). Several correct yeast clones were identified by a 10.9 kb EcoRI fragment hybridizing to the c-myc probe, two Sail fragments of

11.4 kb and 8.9 kb hybridizing to the neo probe and the expected increase in size of the undigested YAC by 30 kb

(Figure 2B) . Tandem integration of more than one copy of pRBura3 would give rise to an additional 12.3 kb c- myc specific EcoRI fragment was seen in other yeast clones (data not shown) .

The translocation locus termed IgH/c-myc YAC (yeast clone Nl) was transferred into ES cells by fusion with yeast spheroplasts followed by G418 selection. Several G418 resistant ES cell clones were obtained and Southern hybridization with probes specific for different regions of the YAC (see Figure 1) indicated that a single and complete copy was integrated into the ES cell genome of clones Nl-21 and Nl-22 (Figure 3, left panels). Single copy integration was evident from the hybridization profile of the amp probe with EcoRI digested DNA (Figure 3, lower panel on the left) . A 7.6 kb internal fragment of the modified right YAC arm and two additional fragments of variable size were detected in each clone.

These derive from the distal EcoRI fragments of the left (5.5 kb) and modified right (5.1 kb) YAC arm and their increase in size is therefore dependent on the integration site of the YAC which in the case of Nl-22 resulted in co-migrating fragments. Thus, the different fragment patters in the ES cell clones are indicative of

YAC integration at different sites in the genome. The presence of both YAC arms was further confirmed by additional digests (data not shown) whilst hybridization with total yeast DNA showed that clones Nl-21 and Nl-22 in contrast to others (eg. Nl-25) had not taken up any significant amounts of yeast DNA in addition to the YAC

(Figure 3, right panel) .

Example 2

Tumour formation in chimeric mice derived from transgenic ES cells

The ES cell clones Nl-21 and Nl-22 containing the IgH/c- myc YAC were used for blastocyst injection. A total of nine chimeric mice were born with a coat colour contribution obtained from the ES cells of 25 - 70%. By the age of 8-16 weeks six of these chimeras, four derived from Nl-21 and two from Nl-22, had developed tumours (Table 1) some with a visible onset of the illness as early as 7-8 weeks. Their illness became uniformly manifested by a hunched posture, ruffled fur and the formation of a prominent lump on the head. The latter turned out to be caused by soft, diffuse tumour tissue located intracranially adherent to the bones and clearly separate from the brain. An autopsy showed that almost all chimeras suffered from additional and larger tumour tissues, located in the abdomen and/or chest (in Figure 8, % of abnormal cells), presented as soft tumour masses with non-diffuse borders which could contribute to about 10% of their body weight. The only other abnormalities observed were a frequent enlargement of the spleen (2 to 8 fold in four animals) and/or the thymus (two animals) . Signs of the illness manifested in posture, coat, and movement were also noticed in chimera

21-1 by just 5 weeks of age but no tumours were found when the mouse was sacrificed a few days later. Chimeras 22-3 and 22-4 were tumour free (sacrificed at

28 and 37 weeks of age, respectively) and none of six non-chimeric littermates (ie. chimerism undetectable by the coat colour) showed any signs of illness (data not shown) .

For further studies, cell lines were established from intracranial as well as intraperitoneal tumour samples of chimera 21-3 (derived from ES cell clone Nl-21) and chimera 22-1 (derived from Nl-22) . The shape of these cells closely resembled that of common lymphoid cell lines (eg. the Burkitt ' s lymphoma cell line DAUDI, ECACC no. 85011437) but they were dependent on feeder cells for growth (see Materials and Methods) . In all cultures the majority of the tumour cells were loosely attached to the feeders while up to about 40% were in suspension.

Example 3

Lymphoid origin and monoclonali ty of the tumours

In order to identify the developmental stage of the tumours, flow cytometry analyses of single cell suspensions from tumour tissues and cell lines derived therefrom were performed using antibodies against the B- cell surface marker B220 (CD45R) and mouse IgM. In all tumour tissues and cell lines analysed, well over 80% of the cells stained brightly for both markers (Figure 4A) . A significantly enlarged proportion of double-positive cells was noticed in the spleen of all chimeras and, in some cases, in thymus and blood when compared with non- transgenic control tissues of BALB/c mice (Figure 4B) . In contrast, when animals carrying the HuIgH YAC without a c-myc gene (ie. the HuIgH YAC retrofitted with two copies of pLNA) were analysed, they were healthy and never developed tumours. The FACS cytogram of their spleen was always indistinguishable from non-transgenic BALB/c mice (data not shown) . Thus the alterations in the FACS profiles are indicative of the massive infiltration of the tumour cells into the various organs.

The B-lymphoid origin of the tumour cells was further characterised on the molecular level . This involved identification of rearrangements of the endogenous and transgenic (see below) IgH loci (Table 1) to investigate their clonality. The rearrangement pattern of the mouse IgH locus was determined using DNA from various tumour samples of the chimeras analysed by digestion with EcoRI and hybridization with the mouse 3 ' J_H probe (Figure 5) . All tumour samples showed rearrangement of both alleles with two altered fragments in equimolar amounts (in chimera 22-1 these are probably two co-migrating fragments) and the absence of the germline bank (Figure 5) . However, whilst distinct rearrangements were found in the different animals, tissue samples dissected from different sites in the same mouse (either primary tissue or cultivated cells thereof) showed in all cases the same rearranged fragments (Figure 5, lanes 2-3 and lanes 4-6) . This not only confirmed a mature B-cell origin of the tumours, but also indicated a monoclonal origin and the tendency of the tumours to form metastases . This is further supported by the infiltration of tumour cells in the spleen where identical rearrangements can be seen (Figure 5, first lane) .

Example 4

Presence and integrity of the translocus in the tumour cells

In view of the chimeric character of the tumour-bearing mice, tumour cells were analysed for the presence of the oncogene. For this, DNA was digested and hybridized with a c-myc specific probe which detects the human transgene as well as the endogenous mouse c-myc gene on a 10.9 kb and a 21 kb EcoRI fragment, respectively. The latter served as an internal standard to prove equal loading of the samples. The human c-myc band was present in all tumour samples (Figure 6A) and the intensity of the signal when compared with the signal obtained from a similar amount of DNA from IgH/c-myc YAC containing ES cells suggests that most if not all cells were derived from them. As this also holds true for all cell lines derived from the tumours, it is clear that the manipulated ES cells with their potential to differentiate are tumorigenic in the B-cell lineage.

The next question was the integrity of the IgH/c-myc translocus which we assessed by analysing V_HDJ_H rearrangements of the human IgH region. Hybridization of BamHI -HindiII digests with the human Eμ probe showed a germline-size fragment of 5.8 kb in tumour DNA from chimera 21-5 whilst in tumour samples from other chimeras (eg. 21-3 and 22-1) different rearrangement products were detected (Figure 6B) . As expected, identical hybridization bands were obtained from different tumour samples of the same mouse (data not shown) further verifying their monoclonal origin. However, as rearrangements could also potentially indicate gross alterations of the transgene locus, which could have deleted a region between the intron enhancer and the oncogene, we established the presence of the neighbouring BamHI -HindiII fragment by hybridization with the Cμ probe. In all tumour samples a germline- size fragment (10.5 kb) was detected (Figure 6C) . The configuration of the translocus in tumour DNA from three different chimeras was further analysed in Spel digests.

In the human IgH locus a Spel fragment of about 100 kb accommodating V_H6-1, the D region, Cμ and part of Cδ (Berman et al . , 1988 , Sato et al . , 1988, Hofker et al . , 1989; see Figure 1) would be altered by V_HDJ_H recombination. In Figure 6D hybridization with the V_H6-1 probe highlights this Spel fragment in germline configuration in Nl-21 ES cells. In tumour DNA from chimera 21-5, where no rearrangement could be detected with the Eμ probe, the size of this fragment was unaltered (Figure 6D) . Unexpectedly, no alteration of the Spel band was found in the tumour cells of chimera 22-1. The shortened BamHI -Hindlll fragment hybridizing to Eμ is therefore most likely due to a DJ_H rearrangement involving D_HQ52 (see Figure 1) , rather than V_H to DJ_H, as this would result in a deletion of just 1.4 kb from the 100 kb Spel fragment. In the tumour DNA from chimera 21-3, however, the Spel fragment was shortened to about 30 kb (Figure 6D) as would be expected for a V_HDJ_H rearrangement involving V_H6-1. In agreement with these interpretations, hybridizing EcoRI digested DNA with the V_H6-1 probe revealed germline-size fragments (0.8 kb; Buluwela and Rabitts, 1988) in chimeras 21-5 and 22-1 and a larger, rearranged fragment of 9.4 kb in chimera 21-3 (data not shown) .

To further verify the integrity of the region between the intron enhancer and c-myc in the tumour cells, Spel sites at the 3' end of the IgH/c-myc translocus (see Figure 1) were analysed. As shown in Figures 6E and 6F, fragments in germline configuration of 9.0 kb detected by the δM2 probe and 18 kb detected by the 3 ' IgH probe were present in all tumour samples analysed. From the mapping results presented in Figure 1 it is thus likely that both Spel fragments represent the complete region

3' of Cδ on the IgH/c-myc translocus TAC . The hybridization results therefore show that the 50 kb region from Eμ to c-myc is unaltered in all the tumours analysed.

Example 5

Cellular and molecular analysis of development of tumour cells in IgH/c-myc chimeric mice .

BL-type tumours or precursor tumour cells were found in 100% of chimeric mice. Cells taken from the IgH/c-myc chimeric mice at different ages were analysed by FACS for the presence of an abnormal population of large IgM+ cells. As shown in Figure 7, such cells (tumour precursors) were already evident in day 17 fetal liver, well before the appearance of tumours. In Figure 8 they are demonstrated in bone marrow, blood, spleen and but not Peyer's patches of animals from 10 days to 9 weeks of age. The surface phenotype of the tumour cells is shown in Figure 9 (positive for IgM and IgD, negative for CD43) which establishes them as encompassing cells from the pre B-cell to the mature B-cell stage of development .

Molecular analysis of rearrangement of the mouse Ig H and K loci and of the human Ig YAC-borne genes is shown in Figures 10-12. It is evident that rearrangement of the mouse H locus is mandatory for tumour development, whereas rearrangement of the K locus is variable. Thus, in Figure 11, the Southern blot confirms that the same H rearrangement is present in each tumour (abdomen, head) from the same animal, whereas there is a variable rearrangement of the allelic H locus and of the K light chain locus (e.g. mouse 2, A (abdominal tumour) versus H (head tumour) ) . This implies that one mouse IgH chain rearrangement event is necessary for c-myc gene activation, but that rearrangement of the second H allele or the K light chain locus is not essential. The rearrangement of the human YAC is similarly variable

(mouse 2) and thus probably occurs at different times in relation to c-myc activation and is by amplification nonessential . Figure 12 confirms the consistency of mouse H rearrangement and variability of K rearrangement using tumour cell clones grown in culture.

Example 6

In order to assess the importance of a particular intron enhancer in tumour activation, the Eμ enhancer was deleted from the IgH/c-myc translocus by site specific recombination (Figure 13). The IgH/c-myc/Eμ^" YAC was transferred into embryonic stem cells and chimeric mice were derived. Th tumour development in those mice was identical to tumour development in IgH/c-myc chimeric mice. References

Adams JM, Harris AW, Pinkert CA, Corcoran LM, Alexander WS, Cory S, Palmiter RD, and Brinster RL. (1985) . -Nature, 318, 533-538.

Banerji J, Olson L and Schaffner W. (1983) . Cell , 33,

729-740. Berman JE, Mellis SJ, Pollock L, Smith CL, Suh H, Heinke B, Kowal C, Surti U, Chess L, Cantor CR and Alt FW. (1988). EMBO J. , 7, 727-738.

Brison, 0. (1993). Biochem. Biophys . Acta, 1155, 25-41. Brύggemann M and Neuberger MS. (1996). Immunology today, 17, 391-397.

Buluwela L and Rabbitts TH. (1988). Eur. J. Immunol . , 18, 1843-1845.

Bungert J, Dave U, Lim K-C, Lieuw KH, Shavit JA, Liu Q and Engel JD. (1995) Genes Dev. , 9 , 3083-3096. Burke DT, Carle GF and Olson MV. (1987) . Science, 236, 806-812. Burkitt D. (1958). Br. J. Surg. , 46, 218-233. Cherney BW, Bathia K and Tosato G. (1994) .

Proc . Natl . Acad . Sci . USA, 91, 12976-12971. Cogne M, Lansford R, Bottaro A, Zhang J, Gorman J, Young F., Cheng H-L and Alt FW. (1994) Cell , 11 , 737-747.

Cohen JHM, Revillard J-P, Magaud J-P, Lenoir G,

Vuillaume M, Manel A-M, Vincent C and Bryon P-A. (1987). J^". Natl . Cancer Inst . , 78, 235-242. Croce CM. (1993). Semin Oncol . , 20:5 (Suppl 5), 31-46. Dariavach P, Williams GT, Campbell K, Pettersson S and Neuberger MS. (1991). Bur. J. Immunol . , 21, 1499- 1504.

Davies NP, Rosewall IR and Brϋggemann M. (1992) . Nucleic Acids Res . , 20, 2693-2698. Davies NP, Rosewell IR, Richardson JC, Cook GP, Neuberger MS, Brownstein BH, Norris ML and Brύggemann M.

(1993). Biotechnology, 11, 911-914. Davies NP, Popov AV, Zou X and Brύggemann M. (1996) .

Antijbody engineering. A Practical Approach . McCafferty J, Hoogenboom HR and Chiswell DJ (eds) .

IRL Press: Oxford, pp. 59-76. Duff K and Huxley C. (1996) . YAC Protocols (Methods in

Molecular Biology, 54) . Markie D (ed) . Humana Press: Totwa NJ, pp. 187-198. Escot C, Theillet C, Lidereau R, Spyratos F, Champeme M-

H, Gest J and Callahan R. (1986) Proc. Natl . Acad.

Sci . USA, 83, 4834-4838. Evan Gl and Littlewood TD. (1993) . Current Opinion Genet . Develop. , 3, 44-49. Gaensler KML. , Kitamura M and Kan YW. (1993). Proc . Natl . Acad . Sci . USA, 90, 11381-11385.

Gerster T, Picard D and Schaffner W. (1986) . Cell , 45,

45-52. Gillies SD, Morrison SL, Oi VT and Tonegawa S. (1983) . Cell , 33, 717-728.

Gutierrez MI, Bathia K, Barriga F, Blanca D, Sackmann

Muriel F, de Andrewas ML, Epelman S, Risueio C and

Magrath IT. (1992). Blood, 79, 3261-3266. Gunven P, Klein G, Klein E, Norin T and Singh S (1980) . Int. J. Cancer, 25, 711-719.

Hamlyn PH and Rabbitts TH. (1983) . Nature, 304, 135-

139.

Harris AW, Pinkert CA, Crawford M, Langdon WY, Brinster

RL and Adams JM. (1988a) . J^". Exp. Med. . , 167, 353- 371.

Harris AW, Langdon WY, Alexander WS, Kariharan IK, Rosenbaum H, Vaux D, Webbe E, Bernard 0, Crawford M, Abud H, Adams JM and Cory S. (1988b) . Curr. Top.

Microbiol . Immunol . , 141, 82-93. Hayday AC, Gillies SD, Saito H, Wood C, Wiman K, Hayward WS and Tonegawa S. (1984). Nature, 307, 334-340. Hoffman-Liebermann B and Liebermann DA. (1991) . Mol . Cell . Biol . , 11, 2375-2381.

Hofker MH, Walter MA and Cox DW. (1989) . Proc . Natl . Acad. Sci . USA, 86, 5567-5571.

Hogan B, Beddington R, Costantini F and Lacy E. (1994) .

Manipulating the Mouse Embryo . A Laboratory Manual . Cold Spring Harbour Laboratory Press .

Hόrtnagel K, Mautner J, Strobl LJ, Wolf Da, Christoph B, Geltinger C and Polack A. (1995) . Oncogene, 10, 1393-1401. Jain VK, Judde JG, Max EE and Magrath IT. (1993) . J^".

Immunol . , 150, 5418-5428. Lavenu A, Pournin S, Babinet C and Morello D. (1994) . Oncogene, 9 , 527-536.

Lieberson S, Giannini SL, Birshtein BK and Eckhardt LA.

(1991 Nucleic Acids Res . , 19, 933-937.

Madisen 1 and Groudine M. (1994) . Genes Dev. , 8, 2212- 2226.

Matthias P and Baltimore D. (1993). Mol . Cell . Bio . , 13, 1547-1553.

Magrath I. (1990). Adv. Cancer Res . , 55, 133-270. Mautner J, Behrends U, Hόrtnagel K, Brielmeier M, Hammerschmidt W, Strbl L, Bornkamm GW and Polack A. (1996). Oncogene, 12, 1299-1307. McKee-Johnson J and Reeves RH. (1996) YAC Protocols

(Methods in Molecular Biology, 54) . Markie D (ed) . Humana Press: Totowa NJ, pp. 167-186. McMahan AP and Bradley A. (1990). Cell , 6.2. 1073-1085. Michaelson JS, Giannini SL and Birshtein BK. (1995) .

Nucleic Acids Res . , 23, 975-981. Milner AE, Grand RJA, Waters CM, and Gregory CD. (1993) . Oncogene, 8, 3385-3391. Milstein CP, Deverson EV and Rabbitts TH. (1984) . Nucleic Acids Res . , 12, 6523-6535. Morello D, Lavenue A, Pournin S and Babinet C. (1993) .

Oncogene, 8, 1921-1929. Peterson KR, Clegg CH, Huxley C, Josephson BM, Haugen HS, Furukawa T and Stamatoyannopoulos G. (1993) . Proc . Natl . Acad . Sci . USA, 90, 7593-7597. Peterson KR, Li QL, Clegg CH, Furukawa T, Navas PA,

Norton EJ, Kimbrough TG and Stamatoyannopoulos G. (1995). Proc . Natl . Acad . Sci . USA, 92, 5655-5659. Rabbitts TH and Boehm T. (1991). Adv. Immunol . , 50, 119- 146. Rabbitts TH, Forster A, Baer R and Hamlyn PH. (1983) .

Nature, 306, 806-809. Rose M, Grisafi P and Botstein D. (1984) . Gene, 29, 113- 124.

Rosenbaum H, Harris AW, Bath ML, McNeall J, Webb E, Adams JM and Cory S. (1990). EMBO J. , 9 , 897-905. Sambrook J, Fritsch EF and Maniatis T. (1989^ . Molecular Cloning: A Laboratory Manual . Cold Spring Harbour Laboratory Press.

Sandgren EP, Quaife CJ, Paulovich AG, Palmiter RD and Brinster RL. (1991) . Proc. Natl . Acad. Sci . USA, 88, 93-

97. Sato T, Matsuda F, Lee KH, Shin EK and Honjo T. (1988) . Biochem . Biophys . Res . Commun . , 154, 265-271.

Schmidt EV, Pattegale PK, Weir L and Leder P. (1988) .

Proc . Natl . Acad . Sci . USA, 85, 6047-6051. Schoenenberger C-A, Andres A-C, Groner B, van der Valk M, LeMeur M and Gerlinger P. (1988) . EMBO J. , 7, 169- 175.

Selfridge J, Pow AM, McWhir J, Magin TM and Melton DW.

(1992). Som . Cell Mol . Genet . , 18, 325-336. Shiramizu B, Barriga F, Neequaye J, Jafri A, Dalla- Favera R, Neri A Guttierez M, Levine P and Magrath I. (1991). Blood, 11 , 1516-1526. Sidman CL, Marshall JD and Harris AW. (1988) . Curr. Top.

Microbiol . Immunol . , 141, 95-99. Smith AJH, De Sousa MA, Kwabi-Addo B, Heppell-Parton A, Impey H and Rabbitts P. (1995) . Nature Genet., 9, 376-385.

Spanopoulou E, Early A, Elliott J, Crispe N, Laydman H,

Ritter M, Watt S, Grosveld F and Kioussis D. (1989). Nature, 342, 185-189.

Spencer CA and Groudine M. (1991) . Adv. Cancer Res . , 56, 1-48.

Spencer F, Ketner G, Connelly C and Hieter P. (1993) . Methods: A Companion To Methods in Enzymology, 5, 161-175. Stewart TA, Pattengale PK and Leder P. (1984) . Cell , 38, 627-637.

Suda Y, Aizawa S, Hirai S, Inoue T, Furata Y, Suzuki M, Hirohashi S and Ikawa Y. (1987). EMBO J. , 6, 4055- 4065. Wagner SD, Gross G, Cook GP, Davies SL and Neuberger MS. (1996) Genomics, 35, 405-414.

Yokoyama K and Imamoto F. (1987) . Proc . Natl . Acad. Sci .

USA. , 84, 7363-7367. Yukawa K, Kikutani H, Inomoto T, Uehira M, Bin SH, Akagi K, Yamamura K-I and Kishimoto T. (1989) . J. Exp . Med. , 170, 711-726.

Zou X, Xian J, Davies NP, Popov AV and Brύggemann M. (1996). FASEB J. , 10, 1227-1232.

Claims

1. A yeast artificial chromosome which comprises a gene activation element and a proto-oncogene or oncogene activated by the gene activation element.

2. A yeast artificial chromosome as claimed in claim 1 wherein the gene activation element comprises, or is from, an immunoglobulin locus.

3. A yeast artificial chromosome as claimed in claim 2 wherein the immunoglobulin locus is an immunoglobulin heavy chain locus .

4. A yeast artificial chromosome as claimed in any one of claims 1 to 3 wherein the proto-oncogene or oncogene is c-myc.

5. A yeast artificial chromosome as claimed in any one of claims 1 to 4 wherein the gene activation locus is human, murine, ovine, bovine or porcine derived.

6. A yeast artificial chromosome as claimed in any one of claims 1 to 5 wherein the proto-oncogene or oncogene is human, murine, ovine, bovine or porcine derived.

7. A yeast artificial chromosome as claimed in any one of claims 1 to 6 wherein the gene activation element is separated from the proto-oncogene or oncogene by more than lOkb, preferably more than 50kb.

8. A yeast artificial chromosome as claimed in any one of claims 1 to 7 wherein the gene activation element is modified or mutated such that activation of the proto- oncogene or oncogene is altered.

9. A yeast artificial chromosome as claimed in claim 8 wherein the gene activation element is modified or mutated such that activation of the proto-oncogene or oncogene is decreased or eliminated.

10. A process for the transformation of a yeast artificial chromosome, as claimed in any one of claims 1 to 9, into a host cell genome, comprising fusing a host

cell with a yeast cell, the yeast cell comprising the yeast-artificial chromosome.

11. A host cell comprising integrated into its genome, a yeast artificial chromosome as claimed in any one of claims 1 to 9.

12. A host cell as claimed in claim 11 which is human, murine, ovine, bovine or porcine.

13. A host cell as claimed in claim 11 or claim 12 which is an embryonic stem cell, in particular a mouse embryonic stem cell.

14. A host cell as claimed in any one of claims 11 to 13 for use in producing chimeric tissue or a chimeric, transgenic animal .

15. A process for producing chimeric mice comprising injecting embryonic stem cells as claimed in claim 13 into a mouse blastocyst .

16. Chimeric tissue, including tumours or a chimeric or transgenic animal produced from a host cell as claimed in any one of claims 11 to 14, or produced from a process as claimed in claim 15.

17. Use of a yeast artificial chromosome as claimed in any one of claims 1 to 9, or a host cell as claimed in any one of claims 11 to 14 or chimeric tissue or a chimeric or transgenic animal as claimed in claim 16 to investigate the role of a distantly situated genetic control element .

18. Use of a yeast artificial chromosome as claimed in any one of claims 1 to 9, or a host cell as claimed in any one of claims 11 to 14 or chimeric tissue or a chimeric or transgenic animal as claimed in claim 16 for the in vivo analysis of proto-oncogene or oncogene activation and/or tumour formation.

19. Use of a yeast artificial chromosome as claimed in any one of claims 1 to 9, or a host cell as claimed in any one of claims 11 to 14 or chimeric tissue or a chimeric or transgenic animal as claimed in claim 16 for the identification of a tumour suppressing element.

20. A tumour suppressing element identified by a use as claimed in claim 19.