Mouse chromosomes can be examined by three types of staining methods. The classical approach, in which the entire chromosome is more or less uniformly stained by orcein, Giemsa or Feulgen, can be used to determine the number of chromosomes (2n = 40 in Mus musculus), the position of the centromere (at one end of each musculus chromosome), and the presence and position of secondary constrictions. This approach has been largely replaced by the general banding methods (Q, G and R), which can be used for the same purposes but also produce a consistent pattern of light and dark bands along the length of each chromosome. The banding pattern is different for each mouse chromosome and thus permits its accurate identification ( 1, 2). In addition, special banding methods can be used to observe particular regions of chromosomes, such as the centromeric heterochromatin or nucleolus organizer regions. The special methods are most effective when combined with a general banding method by which the chromosomes are identified. Each of these three types of staining has been used to demonstrate differences between species or strains of Mus ( Table 1).
The general banding techniques are, and will remain for the foreseeable future, the most powerful tools in cytogenetic analysis. The first of these methods to be discovered was quinacrine fluorescence (Q-) banding ( 3), but the same banding patterns are revealed by Giemsa staining after treatment with trypsin (G-banding). Based on the general banding patterns, a standard karyotype of the mouse has been adopted ( 2) and a nomenclature for individual bands proposed ( 4). Figure 1 shows a karyotype of Mus musculus stained with quinacrine. The intensity of fluorescence varies along the length of each chromosome, producing a banding pattern that is consistent and highly characteristic for each chromosome, and which remains unaltered if a segment of chromosome is transferred to a different location. There are no Q-brilliant regions in musculus. Some of the chromosomes in the cell shown in Figure 1 have a secondary constriction, which probably represents the nucleolus organizer region. More specific stains for this region are discussed below.
The ability to identify each mouse chromosome has been utilized in a number of ways, The Q-banding technique was used to identify the chromosomes in a series of translocations involving known linkage groups, making possible the assignment of each linkage group to a specific chromosome, as well as the location of the centromeric end of linkage groups ( 1, 5, 6). Q- and G-banding were used to determine where the albino locus is located on chromosome 7 by locating a band that was deleted in the C25H deletion, which is about 6 cM long ( 7). Methods are being developed to study the chromosomes of prometaphase cells, which have a greater number of bands ( 8); these would make possible the recognition of deletions even smaller than the C25H. G-banding techniques have been used to identify the presence of specific chromosome changes in malignancies, for example, the association of trisomy 15 with leukemia in the AKR mouse ( 9).
Figure 2 is a karyotype of Mus cervicolor ( 10). The banding patterns of the chromosomes of musculus and cervicolor are the same except for one striking difference: each cervicolor chromosome has a brilliant fluorescent centromeric region separated from the remainder of the chromosome by a dull region. This difference reflects an evolutionary change in the satellite DNA ( 11, 12), which is concentrated in this region of a mouse chromosomes. Musculus has a single satellite DNA that is dully stained by quinacrine; cervicolor has two satellite DNAs that have different molecular properties and stain differently, one being Q-brilliant and the other Q-dull. As shown in Figure 2, these satellite DNAs are located in adjacent regions near the centromere.
If we assume that there are 30,000 genes in the mouse, the average chromosome would have about 1,500 genes; however, it has enough DNA to code for 50 to 100 times that many genes. Much of this DNA may be the transcribed spacers coding for heterogeneous nuclear RNA that is degraded in RNA processing. More may be nontranscribed spacer regions, which can be quite heterogeneous ( 13) and may be clustered. There is some evidence that the Q-bright or G-dark bands consist primarily of such DNA, with the genes concentrated in the bands that are dully stained by these methods (the reverse or R-bands). as methods of DNA sequencing and analysis are improved, variations in nontranscribed DNA may provide major sources of information about differences between mouse sublines.
All species of Mus examined thus far have the same general chromosome banding patterns ( 10, 14, 15, 16) ( Figure 3). This is not unexpected for the closely related subspecies Mus musculus musculus and M. musculus molossinus, but it is surprising that the chromosomes of M. cervicolor, M. caroli, M. cookii, M. booduga fulvidentris, and M. dunni show no sign of inversions, reciprocal translocations, or centric fusion translocations. Inversions have been specifically looked for in wild mice that are fairly closely related to musculus, but none have been found (T.H. Roderick, personal communication). Although the unbanded Giemsa-stained karyotype of M. dunni ( Figure 4) clearly differs from that of M. musculus in having short arms on virtually every autosome, as well as a metacentric X and a very large Y, G- and C-banding have shown that the differences are entirely due to the presence of short arms composed of constitutive heterochromatin, with no detectable change in the banding pattern of any chromosome's long arm ( 16). This is an important observation because differences of this magnitude observed in the pre-banding era were usually interpreted as the result of inversions or translocations.
The translocations that have been observed in wild mice, as well as those that have occurred spontaneously in laboratory stocks, have all been of the centric fusion type ( 6, 17, 18, 19). The tobacco mouse, Mus poschiavinus, has 2n = 26 chromosomes, including seven pairs of biarmed chromosomes ( 17). These have been identified by their Q-banding pattern ( Figure 5) and in this case, too, the patterns are identical to those of musculus chromosomes, showing that the biarmed chromosomes were produced by centric fusion. Multiple centric fusion chromosomes have been found in homozygous form in mice from several mountain valleys in Italy and Switzerland ( 17, 18, 19). Comparison of the various populations indicates that, in general, different chromosomes are represented in the fusion chromosomes of mice from different locations. Figure 6 shows pairing of meiotic chromosomes in an F1 between a mouse of the T Rov 1-6 stock (caught near Roverdo in the Cal Mesolcina) and poschiavinus (caught in the Valle di Poschiavo about 75 miles away). Only one of the fusion chromosomes is identical in the two stocks of mice, and virtually every chromosome is included in one or another of the 13 biarmed chromosomes.
C-banding involves a rather nonspecific type of staining reaction: treating chromosome preparations with alkali and then staining with Giemsa. The C-band method stains only the centromeric heterochromatin, and a general banding technique must be used on the same cells in order to identify the chromosome. When this is done, differences are seen between wild mice and laboratory strains ( 20) and between even closely related sublines. C57L/J and C57BL/J, for example, differ markedly in the amount of C-band material on five different pairs of chromosomes, and 129 and AKR differ in the amount on four pairs of chromosomes ( 21). In the inbred lines examined thus far, eight chromosomes have been shown to have such variants ( 22) and if we add those found in molossinus virtually every chromosome has a C-band variant. In an F1 between one molossinus strain and C57BL/6J, differences were observed between eight pairs of homologs ( 14) ( Figure 7). Since these variants are at the centromeric end of the chromosome, they provide useful markers in monitoring stocks for changes in this region. C-banding also facilitates meiotic analysis because the position of each centromere can be located; this is particularly useful in analyzing reciprocal translocations ( 23). Figure 8 shows that gross differences in the amount of C-band material on homologous chromosomes in a musculus x molossinus F1 did not interfere with meiotic pairing ( 14).
The C-band regions of musculus can also be stained using Hoechst 33258 ( 24). Hsu et al. ( 15) have reported that Hoechst 33258 produces bright fluorescence in the C-band regions of musculus and molossinus chromosomes but not in those of other species of Mus. Changes in satellite DNA could be responsible for this, since Hoechst 33258 shows enhanced fluorescence when bound to some AT-rich DNA sequences, such as those known to be present in musculus satellite DNA.
Greater specificity in detecting classes of repetitive DNA can be obtained by using antibodies to specific DNA bases, e.g., anti-5-mythycytidine. These antibodies react only with single-stranded DNA so that, by using selective methods of denaturation of DNA in conjunction with specific antibodies, one can visualize specific classes of repetitive DNA sequences n mouse chromosomes ( 14, 25, 26). In musculus, satellite DNA contains much more 5-methylcytosine than does main band DNA ( 27). When chromosomes are UV-irradiated, the major effect is the production of thymine dimers, which creates single-stranded regions primarily in AT-rich regions such as satellite DNA. One observes intense binding of anti-5-methylcytidine tot he satellite DNA because it is both methylated and single stranded under these conditions ( Figure 9). However, little binding of the antibody occurs in main band DNA because in this case the 5-methylcytosine is located in GC-rich sequences which remain double stranded ( 26). We have not yet applied this method to mouse species other than musculus; it might detect changes in nucleotide sequences in satellite DNA and permit distinctions other banding techniques do not allow.
Another method that has been used to distinguish between strains of musculus is silver staining of nucleolus organizer regions (NORs) ( 28) which identifies sites of active 18S and 28S ribosomal RNA (rRNA) genes. For example, Ag-stained NORs are present on numbers 12, 15, and 18 in both BALB/cJ and C57BL/6J, but, of the two strains, only BALB/cJ has NOR staining on chromosome 16 and only C57BL/6J has an NOR on number 19 ( 29, 30). Figure 10 shows that NORs are present on most of the same chromosomes in Mus musculus molossinus. In this animal there are NORs on chromosomes 12, 16, 17, 18, and 19 and it is clear that the presence of an NOR is unrelated to the presence or absence of C-band material ( 30). In other strains of mice in situ hybridization with radioactively labeled rRNA has been used to show that rDNA is present on five of the same pairs of chromosomes (all but number 17) ( 31, 32). The finding that number 17 can be included in the list is of interest because the rDNA cluster must be located relatively close to the H2 major histocompatibility locus.
Mouse-human somatic cell hybrids are important tools in current genetic studies. Mouse-human hybrids have been used to demonstrate that silver stain detects active, but not inactive, rRNA gene clusters. Hybrids that have lost human chromosomes have only mouse NORs silver stained and produce only mouse rRNA, even though the human NOR chromosomes are still present ( 33, 34). Conversely, hybrids that have selectively lost mouse chromosomes have only human NORs silver stained and only human rRNA is synthesized, even though the mouse NOR chromosomes are still present ( 29, 34). Mouse-human hybrids have also provided a powerful tool for mapping the human genome. In just eight years the number of genes assigned to human autosomes has grown from one to nearly 150 ( 35), most of them assigned using somatic cell genetic analysis of interspecific hybrids. The same methods that have proved so useful for mapping the human genome are now being applied to the mouse, and the human is on the way to becoming an excellent model system for those interested in mouse genetics. Chinese hamster-mouse hybrids usually lose mouse chromosomes ( Figure 11), and by correlating the presence of a mouse isozyme, antigen or other genetic marker with the presence of a particular mouse chromosome one can assign the relevant gene to its chromosome ( 36). More than a dozen biochemical markers have been assigned to mouse chromosomes in this way, and hundreds more can be. a method has also been worked out by which one can determine the linear order of linked genes using somatic cell hybrids ( 37). Prior to hybridization, the cells of the parent species to be mapped are heavily irradiated to produce chromosome breaks. The closer together two linked genes are, the higher the dose of radiation required to produce a break between them with loss of the fragment that lacks a centromere. Thus, by correlating the dose of radiation with the frequency of separation of linked loci, one can determine both linear order and relative distances between genes on each chromosome. Finally, using the ability to distinguish homologous chromosomes, mouse-mouse somatic cell hybrids have been used to map both structural and regulatory genes to mouse chromosomes ( 38, 39).
A special banding method, Giemsa staining at pH 11 (G11), has proved particularly useful in the analysis of interspecific hybrid cells because mouse chromosomes stain strongly, except at the centromere, whereas primate chromosomes generally stain faintly ( 40). This has made it possible to identify translocations between entire chromosomes of two species, as well as of much shorter segments of chromosome. Furthermore, feeding isolated chromosomes to cells in culture leads to incorporation of very short segments of genetic material, sometimes only long enough to contain two closely linked genes, e.g., those coding for thymidine kinase and galactokinase in both human ( 41) and mouse ( 42, 43) genomes. The ability to produce viable cells in culture whose chromosomes can contain DNA of two species may be relevant to geneticists who work with whole animals. The work of Mintz and her collaborators ( 44) indicates that it is possible to make chimeric mice using, as one parent, in vitro cultured teratocarcinoma cells. When the teratocarcinoma cells were mutagenized and a mutant (HPRT-) subline selected in vitro, mice containing some HPRT- cells were produced ( 41). If such HPRT- teratocarcinoma cells were converted to wild type (HPRT+) with DNA of a different species, it might be possible to produce mouse chimeras that contained blocks of genes from a species other than Mus: chimeras in the real sense! One can even imagine, for example, putting the human HLA major histocompatibility complex on a mouse background, or carrying out similar manipulations affecting the mouse genome in a profound way by artificial introgression of alien genetic material.
1. Miller, O.J., Miller, D.A., Kouri, R.E., Allderdice, P.W., Dev, V.G., Grewal, M.S., and Hutton, J.J. (1971). Proc. Natl. Acad. Sci. USA 68: 1530.
See also
MGI.
2. Committee on Standardized Genetic Nomenclature for Mice. (1972). J. Hered. 63: 69.
See also
MGI.
3. Caspersson, T., Farber, S., Foley, G.E., Kudynowski, J., Modest, E.J., Simonssen, E., Wagh, U., and Zech, L. (1968). Exp. Cell Res. 49: 219.
See also
PubMed.
4. Nesbitt, M.N., and Francke, U. (1973). Chromosoma 41: 145.
See also
MGI.
5. Miller, D.A., Kouri, R.E., Dev, V.G., Grewal, M.S., Hutton, J.J., and Miller, O.J. (1971). Proc Natl. Acad. Sci USA 68: 2699.
6. Miller, O.J., and Miller, D.A. (1975). Ann. Rev. Genet. 9: 285.
See also
PubMed.
7. Miller, D.A., Dev, V.G., Tantravahi, R., Miller, O.J., Schiffman, M.B., Yates, R.A., And Gluecksohn-Waelsch, S. (1974). Genetics 78: 905.
See also
MGI.
8. Yunis, J.J. (1976) Science 191: 1268.
See also
PubMed.
9. Dofuku, R., Biedler, J.L., Spengler, B.A., and Old, L.J. (1975). Proc. Natl. Acad. Sci. USA 72: 1515.
See also
PubMed.
10. Dev, V.G., Miller, D.A., Miller, O.J., Marshall, J.T., Jr., and Hsu, T.C. (1973). Exp. Cell Res. 79: 475.
11. Sutton, W.D., and McCallum, M. (1972). J. Mol. Biol. 71: 633.
12. Rice, N.R., and Straus, N.A. (1973). Proc. Natl. Acad. Sci. USA 70: 3546.
13. Arnheim, N., and Southern, E.M. (1977) Cell 11: 363.
14. Dev, V.G., Miller, D.A., Tantravahi, R., Schreck, R.R., Roderick, T.H., Erlanger, B.F., and Miller, O.J. (1975). Chromosoma 53: 335.
See also
PubMed.
15. Hsu, T.C., Markvong, A., and Marshall, J.T. (1978). Cytogenet. Cell Genet. 20: 304.
See also
PubMed.
16. Markvong, A., Marshall, J.T., Pathak, S., and Hsu, T.C. (1975). Cytogenet. Cell Genet. 14: 116.
See also
PubMed.
17. Gropp, A., Winkling, H., Zech, L., and Muller, H. (1972). Chromosoma 39: 265.
See also
MGI.
18. Capanna, E., Gropp, A., Winkling, H., Noack, G., and Civitelli, M.-V. (1976). Chromosoma 58: 341.
See also
MGI.
19. Lehmann, E.V., and Radbruch, A. (1977). Experientia 33: 1025.
See also
MGI.
20. Forejt, J. (1973). Chromosoma 43: 187.
See also
PubMed.
21. Dev, V.G., Miller, D.A., and Miller, O.J. (1973). Genetics 75: 663.
See also
PubMed.
22. Miller, D.A., Tantravahi, R., Dev, V.G., and Miller, O.J. (1976). Genetics 84: 67.
See also
PubMed.
23. Allderdice, P.W., Dev, V.G., Miller, D.A., Jagiello, G.M., and Miller, O.J. (1973). Chromosoma 41: 103.
See also
PubMed.
24. Natarajan, A.T., and Gropp, A. (1972). Exp. Cell Res. 74: 245.
See also
PubMed.
25. Miller, O.J., Schnedl, W., Allen, J., and Erlanger, B.F. (1974). Nature 251: 636.
See also
PubMed.
26. Schreck, R.R., Dev, V.G., Erlanger, B.F., and Miller, O.J. (1977). Chromosoma 62: 337.
See also
PubMed.
27. Salomon, R., and Kaye, A.M. (1970). Biochim. Biophys. Acta 204: 340.
See also
PubMed.
28. Goodpasture, C., and Bloom, S.E. (1975). Chromosoma 53: 37.
See also
PubMed.
29. Miller, O.J., Miller, D.A., Dev, V.G., Tantravahi, R., and Croce, C.M. (1976). Proc. Natl. Acad. Sci. USA 73: 4531.
See also
PubMed.
30. Dev, V.G., Tantravahi, R., Miller, D.A., and Miller, O.J. (1977). Genetics 86: 389.
See also
MGI.
31. Henderson, A.S., Eicher, E.M., Yu, M.T., and Atwood, K.C. (1974). Chromosoma 49: 155.
See also
MGI.
32. Elsevier, S.M., and Ruddle, F.H. (1975). Chromosoma 52: 219.
See also
MGI.
33. Miller, D.A., Dev, V.G., Tantravahi, R., and Miller, O.J. (1976). Exp. Cell Res. 101: 235.
See also
PubMed.
34. Croce, C.M., Talavera, A., Basilico, C., and Miller, O.J. (1977). Proc. Natl. Acad. Sci. USA 74: 694.
See also
PubMed.
35. McKusick, V.A., and Ruddle, F.H. (1977). Science 196: 390.
See also
PubMed.
36. Franke, U., Lalley, P.A., Moss, W., Ivy, J., and Minna, J.D. (1977). Cytogenet. Cell Genet. 19: 57.
37. Goss, S., and Harris, H. (1977). J. Cell Sci. 25: 17.
See also
PubMed.
38. Hashmi, S., and Miller, O.J. (1976). Cytogenet. Cell Genet. 17: 35.
39. Benoff, S., and Skoultchi, A.I. (1977). Cell 12: 263.
See also
PubMed.
40. Friend, K.K., Dorman, B.P., Kucherlapati, R.S., and Ruddle, F.H. (1976). Exp. Cell Res. 99: 31.
See also
PubMed.
41. Willecke, K., Lange, R., Kruger, A., and Reber, T. (1976). Proc. Natl. Acad. Sci. USA 73: 1274.
See also
PubMed.
42. McBreen, P., Orkwiszewski, K.G., Chern, C.J., Mellman, W.J., and Croce, C.M. (1977). Cytogenet. Cell Genet. 19: 7.
See also
MGI.
43. Kozak, C.A., and Ruddle, F.H. (1977). Somatic Cell Genet. 3: 121.
See also
MGI.
44. Dewey, M.J., Martin, D.W., Jr., Martin, G.R., and Mintz, B. (1977). Proc. Natl. Acad. Sci. USA 74: 5564.
See also
PubMed.