The American Mink Gene Map

Introduction

The American mink ( Mustela vison ) represents the large family Mustelidae. The family belongs to the order of Carnivora and includes many dozens of species living in a broad range of biotopes, from the Arctic to tropical zones.

Carnivores are of great interest in comparative gene mapping in mammals. Giemsa banding with trypsin analysis has shown that chromosome divergence in Mustelidae, despite the considerable variation in number and morphology of chromosomes, mainly resulted from rearrangements of virtually constant elements, probably representing the remains of an ancestral karyotype ( Graphodatsky and others 1989 ; Wurster-Hill and Centerwall 1982 ). A similar situation was reported in the families Canidae and Ursidae ( Graphodatsky and others 1995 ; Nash and O'Brien 1987 ; Wurster-Hill and Centerwall 1982 ). Moreover, the comparison of the Giemsa-banding chromosomes of Carnivora and species belonging to other orders--primates, rodents, or ungulates--demonstrated that in some cases there is a similarity in the Giemsa-banding patterns for large chromosome regions of the distantly related species ( Graphodatsky 1989 ; Nash and O'Brien 1982 ). However, supplementary data on gene mapping are needed to decide whether this similarity reflects true homology and to reach reliable conclusions about the chromosome evolution of mammals ( O'Brien and others 1993 ; Wakefield and Graves 1996 ).

For a long time, the efforts of mink breeders were focused on the identification of mutations affecting coat color genes. At the time of this writing, approximately 30 coat color mutations have been identified in mink. Most of them (at least 20) are controlled by independent autosomal loci ( Nes and others 1988 ; Robinson 1975 ; Shackelford 1948 ). it is reasonable to expect that some of them are linked because a haploid set of mink chromosomes contains 14 autosomes plus the X or Y sex chromosome ( Christensen and others 1996 ; Mandahl and Fredga 1975 ). However, there has been only I description of linkage of 2 coat color mutations, Ebony ( Eb ) and royal pastel (b) ( Shackelford 1949 ). in this paper, I present the current state of gene mapping of mink, based on data obtained from traditional breeding, somatic cell hybridization, chromosome-mediated or interphase nuclei-mediated gene transfer procedures, and in situ hybridization.

Chromosomal Localization of Mink Genes Using Somatic Cell Hybrids

Chromosomal localization of mink genes was based mainly on the use of 2 panels of clones: mink-Chinese hamster hybrid cells ( Rubtsov and others 1981 ) and mink-mouse hybrid cells ( Pack and others 1992 ). Segregation analysis of mink chromosomes and markers in the hybrid cells made it possible to assign the mink genes to particular chromosomes. Assignment of a gene to a chromosome requires both a high level of concordant segregation of the gene and the chromosome and a sufficiently high level of discordant segregation of the gene and all other chromosomes ( Cowmeadow and Ruddle 1978 ; Rubtsov and others 1981 ; Wijnen and others 1977 ). Segregation analysis of mink chromosomes in the panels of mink-rodent hybrids showed that the percentage of discordant clones for any chromosome pair was more than 20% (no fewer than 5 among 23 or 25 hybrid clones) ( Pack and others 1992 ; Rubtsov and others 1981 ; Serov and others 1987 ). This degree of discordance rendered the panels reliably stable for gene mapping.

As seen in Figure 1 , the mink gene map includes 77 genes, which mark all mink chromosomes except the Y. More than 30 gene assignments made with the panel of mink-Chinese hamster cell hybrids were also supported by results from another panel of mink-mouse hybrids. Only 2 previous chromosomal localizations, those of NP and PKM2, have been revised by these data ( Pack and others 1992 ; Serov and Pack 1993 ).

Figure 1

Gene map of the American mink (Mustela vison) containing 77 biochemical loci marking all mink chromosomes, except the Y. The nomenclature of mink chromosomes is according to Christensen and others (1996), and former numbers of mink chromosomes ( Mandahl and Fredga 1975 ) appear in parentheses. AATP, ATP-ase, alpha-subunit; ACON1, aconitase-1; ACP1, acid phosphatase-1 ACP2, acid phosphatase-2; ACY, aminoacylase; ADA, adenosine deaminase; ADH2,* alcohol dehydrogenase, subunit B; ADK, adenosine kinase; AK3, adenylate kinase-3; ALDB, aldolase B; ALDC, aldolase C; A2M, alpha-2-macroglobulin; APRT, adenine phosphoribosyl-transferase; BATP, ATP-ase, beta subunit; BLVR, biliverdin reductase; CKBB, creatine phosphokinase, brain type; ENO1, enoIase-1; ES1, esterase-1; ES2, esterase-2 (presumed); ES3, esterase-3; ESD, esterase D; ESR, esterase regulator; FNP1, fibronectin pseudogene-1 (presumed); GALK, galactokinase; GAPD, glyceraldehyde-3-phosphate-dehydrogenase; GH, growth hormone; GALA, alpha-galactosidase; GLO1, glyoxalase- 1; GOT1, glutamate-oxaloacetate transaminase-l; GPI, glucosephosphate isomerase; G6PD, glucose-6-phosphate dehydrogenase; GPT, glutamate-pyruvate transaminase; GSR, glutathione reductase; HK1, hexokinase-1; HOXB, homeo box B; HPRT, hypoxanthine phosphoribosyl transferase; IDH1, isocitrate dehydrogenase-1; IDH2, isocitrate dehydrogenase-2; IGGC, immunoglobulin gamma polypeptide, constant region; IGKC, immunoglobulin kappa polypeptide, constant region; IGLC, immunoglobulin lambda polypeptide, constant region; ITPA, inosine triphosphatase; LDHA, lactate dehydrogenase A; LDHB, lactate dehydrogenase B; LPM, lipoprotein of mink; ME1, malic enzyme-1; MDH1, malate dehydrogenase-1 (NAD dependent); MPI, mannose phosphate isomerase; NF1, neurofibromatose-1; NP, purine nucleoside phosphorylase; OTC, ornithine carbamoyltransferase; PEPA, peptidase A; PEPB, peptidase B; PEPC, peptidase C; PEPD, peptidase D; PEPS, peptidase S; PGD-6, phosphogluconate dehydrogenase; PGM 1, phosphoglucomutase- 1; PGK1, phosphoglycerate kinase-1; PGP, phosphoglycolate phosphatase; PKM2, pyruvate kinase, muscle type; PRL, prolactin; PRNP, prion protein; PSP, phosphoserin phosphatase; POMC, proopimelanocortin; PP, inorganic pyrophosphatase; QDPR, quinoid dihydropterine reductase; RAD52, RAD52 protein is a homologue of Saccharomyces cerevisiae recombination and repair protein ( Bindixen and others 1994 ); 5SrRNA, 1 cluster of the genes for 5S ribosomal RNA; 5SrRNA,** another cluster of the genes for 5S ribosomal RNA; SOD1, superoxide dismutase-1; SOD2, superoxide dismutase-2; TK1, thymidine kinase-1; TPI, triosephosphate isomerase; UMPH2, uridine 5'-monophosphate phosphohydrolase-2; WARS, tryptophanyl-tRNA synthetase. *This enzyme is, in fact, aldehyde dehydrogenase-3 (ALDH3). Both ADH2 and ALDH3 have overlapping substrate specificity; hence, further study is necessary to accurately discriminate between them. **Two clusters of the 5S ribosomal RNA.

Regional Assignments of Mink Genes

In mink, 21 genes have been regionally mapped on chromosomes 1, 2, 8, and X ( Figure 1 ). These data were obtained from a variety of studies using different approaches.

Regional assignments of genes ME1, HK1, PP, GOT1, ADK, PGM1, PGD, ENO1, and G6PD were based on the use of the well-identified rearrangements of mink chromosomes 1, 2, and X ( Pack and others 1992 ; Serov and others 1987 ; Zhdanova and others 1985 , 1988 ). Two translocations involving mink chromosomes I and 8 were revealed in 2 mink-mouse hybrid clones of independent origin: t(1;8) (1qter-1 cen::8cen-8pter), containing the entire long arm of mink chromosome 1; and t(1;8) (1qter-1q21::8q24-p26), containing the distal part of the long arm of mink chromosome 1 ( Pack and others 1992 ). No mink ME1 activity was detectable in the 2 clones, suggesting the assignment of the ME1 gene to the short arm of mink chromosome 1 ( Pack and others 1992 ; Figure 1 ).

The different rearrangements of mink chromosome 2 have been identified by cytogenetic analysis of 2 mink cell lines (MV and MVTK cells) and a mink-Chinese hamster hybrid clone, CO113 ( Rubtsov and others 1981 ; Serov and others 1987 ; Zhdanova and others 1985 ). Clone CO113 contained a large deletion that involved the 2p11-pter region of the short arm. In the clone, mink HKI, PP, GOT1, and ADK activities were not detected. These results made it possible to assign the genes HK1, PP, GOT1, and ADK to the short arm of chromosome 2 and the remaining genes ( PGM1, ENO1, and PGD ) to its long arm. A total of 105 subclones were derived from CO113 ( Zhdanova and others 1985 ). Of these, mink PGD activity was present in 103; mink PGD, PGM1, and ENO1 were absent in 1 derivative; and in another derivative, mink PGM1 (but not PGD and ENO1) activity was detected. Cytogenetic analysis revealed a deletion 2q24.4-qter in the latter derivative, which allowed us to provisionally assign PGD and ENO1 to that region ( Zhdanova and others 1985 ; Figure 1 ) and PGM1 to p11.1-q24.4 of mink chromosome 2 ( Serov and others 1987 ). Subchromosomal localization of PP, GOT1, HK, and ADK was determined by a set of mink-mouse hepatoma clones ( Serov and others 1987 ). The MVTK- cells (mink cells deficient in thymidine kinase) were fused with mouse hepatoma cells, BWTG, and 19 primary hybrid clones were isolated ( Zhdanova and others 1985 ). MVTK- cells contained an intact chromosome 2 and a chromosome 2 containing a small deletion in the region 2p22-pter ( Rubtsov and others 1981 ; Serov and others 1987 ; Zhdanova and others 1985 ). Among mink-mouse hepatoma hybrid clones, 13MV1 was found to contain an intact and a deleted mink chromosome 2. Fifteen secondary clones were derived from the clone 13MV1 ( Zhdanova and others 1985 ). Because the segregation of mink PP, GOT1, and HK1 was concordant with the terminal region of the short arm, those 3 genes were provisionally assigned to 2pter-p22, whereas ADK was assigned to pter-p11.1 ( Serov and others 1987 ; Zhdanova and others 1985 ; Figure 1 ).

The order of 4 X-linked genes in mink was established using the method developed by Goss and Harris (1977 ). Segregation of the mink markers ( GALA, PGK1, HPRT, and G6PD ) was analyzed in more than 180 independent hybrid clones obtained by fusion between gamma-irradiated mink fibroblasts with Chinese hamster cells as well as in 31 mink-mouse hepatoma hybrid clones. Statistical analysis of the segregation patterns revealed the following order of the 4 genes on the mink X chromosome: GALA-PGK1-HPRT-G6PD ( Zhdanova and others 1988 ; Figure 1 ). Moreover, cytogenetic analysis of 5 mink-Chinese hybrid clones expressing mink GALA, PGK1, and HPRT, but lacking G6PD, allowed identification of small deletions involving the terminal part of the X-chromosome and the tentative assignment of G6PD to Xq15.22-qter ( Zhdanova and others 1988 ).

For subchromosomal localization of ESD, TK1, GALK, ALDC, UMPH2, and ADH2 (presumed aldehyde dehydrogenase-3, ALDH3 ) on mink chromosome 8, the method of chromosome-mediated gene transfer was used ( Gradov and others 1985 ; Pack and others 1989 , 1992 ). Mink genes were transferred into mutant mouse cells (LMTK cells deficient in thymidine kinase activity) by a chromosome-mediated transfer technique using isolated metaphase chromosomes of mink ( Sukoyan and others 1984 ). As a result, 16 independent transformants were isolated. Cytogenetic analysis demonstrated that 8 of them contained fragments of different sizes from mink chromosome 8. Analysis of ESD, TK1, GALK, ALDC, UMPH2, and ADH2 in the transformants made it possible to establish regional localization for these genes on chromosome 8 ( Figure 1 ). The close linkage of mink TK1 and GALK genes was also supported by another gene transfer experiment. Isolated nuclei from mink fibroblasts were encapsulated in artificial lipid membranes (liposomes). After treatment of murine mutant (LMTK-) cells, a set of clones carrying the selective marker gene of mink origin was isolated ( Sukoyan and others 1985 ). Five of the 14 transformants expressed mink TK1 together with GALK, but not ALDC or ESD . It is notable that gene transfer technology made it possible to establish gene order as qter -(HOXB, UMPH2)-ALDC-GALK-TK1-ADH2 on the long arm of the chromosome 8 (the former nomenclature) ( Pack and others 1989 ; Figure 1 ).

The development of the in situ hybridization technique offered an efficient technology for regional gene assignment. Since 1996, more than 50 single copy sequences (at least 1 for each mink chromosome) have been localized ( Christensen and others 1996 ). In 1998, Christensen and others (1998 ), using in situ hybridization, demonstrated that the genes for 5S ribosomal RNA (5SrRNA) are located in 3 sites on the long arm of mink chromosome 2. Two are at the proximal and the distal edges of the largest G-band, and the third site maps close the centromere. Although the use of heterologous DNA probes seldom leads to a good result, the genes for ESD and WARS were assigned to mink 8q22-25 and 2p21-24, respectively (a nomenclature of Mandahl and Fredga 1975 ), using human probes ( Graphodatsky and others 1991 ). It is important to note that the assignment of the ESD gene in the region was supported by chromosome-mediated gene transfer (see above; Gradov and others 1985 ).

The isolation of DNA fragments homologous to human or mouse genes from mink DNA libraries can provide probes for gene assignment by in situ hybridization. This strategy has already been used to assign the mink GH gene to mink 8p25-23 ( Malchenko and others 1994 ).

Gene Mapping by Combining Somatic Cell Hybridization and Breeding Tests in Mink

In a special population study of more than 40 biochemical markers of the American mink, polymorphisms have been found for the following genes: ES1, ES2, ES3, ESR, PEPB, PEPD, PI (alpha-protease inhibitor), GC (group-specific component), and LPM (high density lipoprotein) ( Borodin and others 1995 ; Mullakandov and others 1986 ; Serov and others 1987 ). Breeding tests have demonstrated that ES1, ES2, ES3, ESR, and PEPD belong to 1 linkage group with distances of 14 cM between ES1 and ES3, 10 cM between ES1 and PEPD, and 24 cM between ES3 and PEPD . No recombinants between the loci for ES1, ES2, and ESR were found. I suggest the following putative order for these loci: ES3--ES1,ES2,ESR--PEPD . The chromosomal localization for ES3 and PEPD on chromosome 7 was established using mink-rodent hybrid clones ( Pack and others 1989 ; Serov and others 1987 ); hence, the linkage group ES3-ES1,ES2,ESR--PEPD is located on the same chromosome ( Figure 1 ).

Breeding tests have demonstrated that PEPB and LPM are linked (11 cM apart) ( Yermolaev and others 1989 ). Because PEPB has already been localized on chromosome 12 ( Mullakandov and others 1986 ), we may infer that LPM is located on the same chromosome ( Figure 1 ). Since 1995, the results of studies using breeding tests have ruled out the possibility of linkage of ESI, PEPD, and PEPB with coat color mutations S (black-cross), p (silver-blue), a (Aleutian), and h (white-hedlund) (unpublished data), as well as between the PI, GC, and coat color gene Cr (Crystal) ( Borodin and others 1995 ; Trapezov 1997 ).

Comparative Gene Mapping: Mink and Other Mammals

At the time of this writing, the 77 genes that comprise the genetic map of the American mink mark all chromosomes except the Y ( Figure 1 ), making it possible to compare the mink map with the genetic maps of other mammals. The comparison shows that there are more than 10 large associations of syntenic genes common to mink, human, mouse, and other mammalian species. Some syntenic groups of mink genes are characteristic of 3 or more orders of mammals. For instance, TK1, GALK, ALDC, UMPH2, GH, HOXB, and NF1 mark a large conserved region in mink covering approximately 30 cM in the mouse gene map (World Wide Web site http://www.informatics.jax.org ). This region is common to primates, carnivores, ungulates, rodents ( Wakefield and Graves 1996 ), and insectivores ( Matyakhina and others 1997 ; Serov and others 1998 ). The presence of this large conserved group of syntenic genes in different mammalian species reflects its evolutionary conservation and probably represents the remains of ancestral genome.

Another large conserved region common to carnivores, primates, rodents, ungulates and even fishes ( Wakefield and Graves 1996 ) is marked by ITPA, ADA, and PRNP . This region covers 18 cM of the mouse gene map. Ten other conserved regions are located on mink chromosomes 1 (on both arms), 2 (on the short and the long arms), 6, 7, and 8 (on the short and the long arms), and 12 and 13 (see Wakefield and Graves 1996 ).

New data on the comparison of mink and human chromosomes using the zoological fluorescence in situ hybridization (ZOO-FISH) technique have appeared in the late 1990s. Hameister and others (1997 ) demonstrated that specific DNA probes for 22 human autosomes cross-hybridized with 32 large regions of mink chromosomes. In some cases, chromosomal DNA probes showed positive staining on only I mink chromosome region. For example, the human chromosome 9 probe hybridized only to the entire mink chromosome 9, the probe for the human chromosome 10 stained only the long arm of mink chromosome 2, and the probe for the long arm of human chromosome 17 hybridized only to the short arm of mink chromosome 5. In other cases, mink chromosomes were painted by probes derived from 2 or 3 different human chromosomes. For instance, the short arm of mink chromosome 8 was painted by a DNA probe specific to human chromosome 20, and the long arm of the same chromosome, by a DNA probe specific to human chromosome 2.

Results obtained with the zoological fluorescence in situ hybridization technique are in remarkable agreement with the gene mapping data in mink described above ( Hameister and others 1997 ). Thus, there is significant evidence in support of a previous conclusion ( Serov and others 1987 , 1991 ) concerning the existence of large conserved regions in mink common to human and mouse as well as other mammalian species. The nature of the conservation of large syntenic gene associations during evolution of mammals is unknown. However, the existence of conserved regions common to many vertebrates may be interpreted as a result of natural selection.

According to the estimates of Ehrlich and others (1997 ), rates of syntenic disruption during evolution differ significantly among mammalian lineages, from 0.05 (cat and Chinese hamster) to 0.90 (human, mouse, and rat) syntenic disruptions per million years. The rate of syntenic disruption in the mink lineage is moderate (0.30) and similar to that in baboons, chimpanzees, and cattle ( Ehrlich and others 1997 ). Early cytogenetic data showing substantial variability in chromosome rearrangements among mammalian lineages support these estimates ( Busch and others 1977 ; Graphodatsky 1989 ; Wilson and others 1974 ). However, it is unclear whether the rate of syntenic disruptions is constant in the lineage or, as suggested by Graphodatsky (1989 ), is possibly higher during the origin of higher taxonomic categories (family, order) than in late speciation within taxa. In any case, the nature of substantial variation in rates of syntenic disruptions is no less mysterious than those of the origin of the large conserved regions in mammalian genome.

Nevertheless, a syntenic group specific to mink has been found ( Khlebodarova and others 1995 ). This group includes genes GPT, PGP, and PSP located on mink chromosome 14 ( Figure 1 ), which is the smallest in the mink karyotype. The homologous human genes are all located on distinct chromosomes: 8 (GPT), 16 (PGP), and 7 (PSP) ( O'Brien 1993 ). The GPT and PSP genes are located on mouse chromosomes 15 and 5, respectively ( O'Brien 1993 ). Probably, the GPT, PGP, and PSP syntenic group has arisen de novo in the Mustelidae. A comparative analysis of the Geimsa banding with trypsin patterns of the chromosomes of more than 20 species representing 6 genera of the Mustelidae family revealed that all of them possess a chromosome similar to mink chromosome 14 ( Graphodatsky and others 1989 ). According to data of Hameister and others (1997 ), human DNA probes derived from chromosome 7 and 16, but not from chromosome 8, cross-hybridized with mink chromosome 14. These human probes painted 2 distinct chromosomes (B4pter-p and E3) in cat, suggesting that this syntenic group is probably unique for the Mustelidae but not the Felidae family.

Based on the existence of conserved regions of syntenic genes in phylogenetically distant mammalian species, comparative mapping data may be used to search for important genes in fur bearing animals. It is possible to use the gene maps as Mendeleev's periodical system. In searching for the location ora gene in mink, one should first determine whether this particular gene belongs to a syntenic group in other species. If so, it is reasonable to use the other member(s) of this syntenic group for which the location is known in mink as a candidate marker for the linkage experiment. This rationale can be applied to search for the location of the color genes.