P. van Vooren

Gene-based anchoring of the rat genetic linkage and cytogenetic maps: new regional localizations, orientation of the linkage groups, and insights into mammalian chromosome evolution.

Abstract. In order to generate anchor points connecting the rat cytogenetic and genetic maps, the cytogenetic position of 62 rat markers (including 55 genes) already localized genetically was determined by fluorescence in situ hybridization. Whenever possible, markers located near one end of the linkage groups were included. These new localizations allowed us to unambiguously orient the 20 autosomal and the X chromosome linkage groups. The position of the centromere in the linkage map could also be determined in the case of several metacentric chromosomes. In addition, the regional localization of 15 other rat genes was determined. These new data bring useful information with respect to comparative mapping with the mouse and the human and to mammalian evolution. They illustrate, for instance, that groups of genes can remain syntenic during mammalian evolution while being subjected to intrachromosomal rearrangements in some lineages (synteny is conserved while gene order is not). This analysis also disclosed cases of synteny conservation in one the two rodent species and the human, while the synteny is split in the other rodent species: such configurations are likely examples of lineage-specific interchromosomal rearrangements associated with speciation.

Assignment of the cyclin-dependent kinase inhibitor genes Cdkn2a and Cdkn2b to rat chromosome bands 5q32-->q34 and 5q31-->q33, respectively by fluorescence in situ hybridization, using small PCR-generated probes.

The cyclin-dependent kinase inhibitor 2A gene (CDKN2A, also named MTS1, INK4A) is a tumor suppressor gene, frequently deleted or mutated in tumors or tumor cell lines (Kamb et al., 1994; Pollock et al., 1996). This gene is peculiar in that it encodes two distinct proteins, p16 and p19ARF, formed from transcripts initiated at distinct promoters (the p19 ARF promoter being upstream from the p16 promoter) (Mao et al., 1995; Quelle et al., 1995). The CDKN2B gene is located upstream from the p19 ARF promoter (in the human and the mouse), and encodes p15 which, like p16 (but unlike p19 ARF) is an inhibitor of the cyclin D-dependent kinases 4 and 6. The human genes Cdkn2a and Cdkn2B map at 9p21 (Kamb et al., 1994). The two mouse genes (Cdkn2a and Cdkn2b) map at 4C3–C6 (Quelle et al., 1995). The rat Cdkn2a and/or the Cdkn2b genes were found to be deleted in cell lines monosomic for chromosome 5, or carrying deletions in this chromosome (Knapek et al., 1995; Zhou et al., 1997; Hino et al., 1995), which is homologous to both human chromosome arm 9p and mouse chromosome 4 (Szpirer et al., 1990). On the other hand, analysis of somatic cell hybrids suggested that a transformation suppressor gene (Sai1) is localized in the rat chromosome region 5q23→q33 (Szpirer et al., 1994). Cdkn2a and Cdkn2b are thus Sai1 candidate genes, and it is therefore interesting to determine directly their regional chromosome position.

Chromosomal mapping of the rat Slc4a family of anion exchanger genes, Ae1, Ae2, and Ae3.

Cardiovascular Research Center, Massachusetts General Hospital and Harvard Medical School, Charlestown, Massachusetts, 02129, USA Department of Internal Medicine, University of Michigan, 1560 MSRB II, 1150 West Medical Center Drive, and Ann Arbor Veterans Administration Hospital, Ann Arbor, Michigan, 48109, USA Department of Molecular Biology, Universite Libre de Bruxelles, Rhode-St-Genese, Belgium Molecular Medicine and Renal Units, Beth Israel Hospital and Depts. of Cell Biology and Medicine, Harvard Medical School, Boston, Massachusetts, 02215, USA

Assignment1 of SND1, the gene encoding coactivator p100, to human chromosome 7q31.3 and rat chromosome 4q23 by in situ hybridization

Originally identified in Epstein-Barr virus transformed B cells where it binds to EBNA-2 (a transcription factor encoded by EBV), the coactivator p100 also binds to the RNA polymerase II initiation factor TFIIE and also interacts with the oncoproteins c-myb and v-myb (Tong et al., 1995; Dash et al., 1996; Leverson et al., 1998). These and other results suggest that p100 and its rat homologue p105 play a role in the regulation of cell proliferation and in cellular transformation. In this study we assigned the gene encoding the coactivator p100 (symbol: SND1, staphylococcal nuclease domain containing 1) to human chromosome band 7q31.3 and its rat homolog (Snd1) to rat chromosome band 4q23.

Assignment of AR1, transcription factor 20 (TCF20), to human chromosome 22q13.3 with somatic cell hybrids and in situ hybridization

The mechanism of activation of genes encoding stromelysins is under intensive investigation since these enzymes are thought to contribute to tumor metastasis (reviewed by McDonnell et al., 1994). Stromelysins are metalloproteinases with relatively broad substrate specificity and can invade tumors causing release and movement of cells. The released cells can circulate, lodge into a capillary, permeate the walls of blood vessels, and proliferate to form a new tumor mass (McDonnell et al., 1994). A previous study has identified a mouse transcription factor (SPBP) that activates expression of the stromelysin-1 gene, possibly in response to mitogens and growth factors (Sanz et al., 1995; Kirstein et al., 1996). Here we report isolation and chromosomal localization of a human cDNA encoding a transcription factor (AR1) that shows extensive sequence identity to the mouse SPBP.

Assignment1 of the rat pleiomorphic adenoma genes (Plag1, Plagl1, Plagl2) by in situ hybridization and radiation hybrid mapping

PLAG1 (pleiomorphic adenoma gene 1), PLAGL1 and PLAGL2 (PLAG-like 1 and PLAG-like 2) constitute a subfamily of C2H2 zinc finger transcriptional regulators (Kas et al., 1998). Activation of PLAG1 on chromosome 8q12 is the most frequent gain of function mutation found in pleomorphic adenomas of the salivary glands (Kas et al., 1997b). PLAGL1 was independently isolated as ZAC, a gene that shares with the tumour suppressor p53 the ability to regulate apoptosis and cell cycle progression and also as LOT1, a gene with lost or decreased expression in malignantly transformed ovarian epithelial cells and breast tumour cells (Abdollahi et al., 1999; Bilanges et al., 1999). PLAGL1/ZAC/LOT1 is a strong candidate gene for transient neonatal diabetes (Kamiya et al., 2000). In the human, PLAG1 maps to chromosome 8q12, PLAGL1 to 6q24→q25 and PLAGL2 to 20q11 (Kas et al., 1997b, 1998 and unpublished data). We determined the regional location of the three rat homologous genes. Materials and methods

Localization of new, microdissection- generated, anonymous markers and of the genes Pcsk1, Dhfr, Ndub13, and Ccnb1 to rat chromosome region 2q1

The centromeric region of rat chromosome 2 (2q1) harbors unidentified quantitative trait loci of genes that control tumor growth or development. To improve the mapping of this chromosome region, we microdissected it and generated 10 new microsatellite markers, which we included in the linkage map and/or radiation hybrid map of 2q1, together with other known markers, including four genes: Pcsk1 (protein convertase 1), Dhfr (dihydrofolate reductase), Ndub13 (NADH ubiquinone oxidoreductase subunit b13), and Ccnb1 (cyclin B1). To generate anchor points between the different maps, the gene Ndub13 and the microsatellite markers D2Ulb25 and D2Mit1 were also localized cytogenetically. The radiation map generated in region 2q1 extends its centromeric end of about 150 cR.

Assignment of the hepatocyte nuclear factor 3 genes to rat chromosome bands 6q23-->q24 (alpha: Hnf3a), 3q41 (beta: Hnf3b) and 1q21-->q22 (gamma: Hnf3g) by in situ hybridization.

The Hnf3 genes encode liver-enriched transcription factors which were shown to bind to the transthyretin and ·1-antitrypsin gene promoters (Lai et al., 1991; Costa et al., 1989). They are expressed early during development and appear to be involved in regulation of endodermal differentiation (Ang et al., 1993). The Hnf3·, -ß and -Á genes were previously mapped on mouse chromosomes 12, 2 and 7 and on human chromosomes 14q12→q13, 20p11 and 19q13.2→q4 (Avraham et al., 1992; Mincheva et al., 1996). Using fluorescence in situ hybridization (FISH), we assigned the Hnf3 genes to rat chromosome regions 6q23→q24 (Hnf3a), 3q41 (Hnf3b) and 1q21→q22 (Hnf3g). These results extend the homology between the chromosomes carrying the Hnf3 genes. Materials and methods

Rat gene mapping in the post-genome sequencing era: the continued utility of cell hybrids to localize rat genes (Cks2, Ephb4, Fabp5, Il13ra1, Rpl10, Ssr4)

Finding the position of a gene is now easily done when the genome sequence is available: the gene position is generally found by a simple query of genomic databases such as those available at the Ensembl browser or the NCBI. We were interested in determining the position of 125 cancer-related rat genes and we found that the position of most of these genes (110) could indeed be identified in this manner. However, in 15 cases, the gene position was not available in these databases, or the results were ambiguous. We then explored a more specialized database, namely the Rat Genome Database, and experimentally mapped these genes using standard and radiation cell hybrids. The 15 genes in question could be localized unambiguously. In four cases, the radiation cell hybrids were indispensable: the sequence of these four genes could not be found in the rat genome sequence. On the basis of the sample we examined, it thus appears that a classical gene mapping method is still required to localize about 3% of the rat genes, as if 3% of the rat gene sequences were lacking in the current rat genome sequence.

Assignment of the gene encoding renin binding protein (Renbp) to rat chromosome Xq37 by in situ hybridization and radiation hybrid mapping

The renin-angiotensin system (RAS) plays a crucial role in the regulation of blood pressure (Laragh, 1995). Within the RAS, the physiological relevance of renin binding protein (RnBP) has been controversial for several years (Schmitz et al., 2000). Originally, RnBP was identified as a protein in the kidney that was capable of binding renin in renal homogenates giving rise to a complex designated high molecular weight renin (Takahashi et al., 1983). It was subsequently proposed that RnBP might act as an endogenous cellular renin inhibitor. Recent studies demonstrated that human RnBP is the enzyme N-acetyl-D-glucosamine (GlcNAc) 2-epimerase (Takahashi et al., 1999). Moreover, gene-targeting experiments in the mouse excluded a specific interaction between RnBP and renin in the kidney or in the circulating RAS of the mouse (Schmitz et al., 2000). So far, the gene encoding RnBP has been mapped on chromosome X in different mammals: Xq28 in human (van den Ouweland et al., 1994), X 29.53 cM in mouse (http:// www.ncbi.nlm.nih.gov/LocusLink/LocRpt.cgi?l=19703) and X in canine (Stoy et al., 1998). As mapping data for the rat homolog Renbp is still lacking, it is unclear whether this gene is located in blood pressure quantitative trait loci (QTL) termed BP/SP-2 and SS-X on rat chromosome X. BP/SP2 has been identified in an intercross between the stroke prone spontaneously hypertensive rat (SHRSP) and the normotensive Wistar-Kyoto (WKY) rat (Hilbert et al., 1991) and SS-X has been characterized in the Sabra hypertension-prone rat (Yagil et al., 1999). We therefore set out to test the chromosome location of Renbp by performing chromosomal mapping analysis in the rat using fluorescence in situ hybridization (FISH) and radiation hybrid (RH) mapping.