Genetic Analysis of the Dominant White-Spotting (W) Region on Mouse Chromosome 5: Identification of Cloned DNA Markers Near W

Proc. Natl. Acad. Sci. USA Vol. 85, pp. 9635-9639, December 1988 Genetics

Genetic analysis of the dominant white-spotting (W) region on mouse chromosome 5: Identification of cloned DNA markers near W (mouse genetics/developmental genes/pigmentation genes/deletion/Mus spretus backcross) EDWIN N. GEISSLER*t, SHIRLEY V. CHENGt, JAMES F. GUSELLAt, AND DAVID E. HOUSMAN* *Center for Cancer Research, E17-536, Massachusetts Institute of Technology, 40 Ames Street, Cambridge, MA 02139; and tNeurogenetics Laboratory, Massachusetts General Hospital and Department of Genetics, Harvard University, Boston, MA 02114 Communicated by Elizabeth S. Russell, August 25, 1988 (receivedfor review June 1, 1988)

ABSTRACT We have assigned several mouse cDNA and but does complement the mutation rump-white (Rw) (4), genomic clones to the W region of mouse chromosome 5, which is also closely linked to both W and Ph (5, 6). Ph, Rw, established their position with respect to various marker loci in and a third tightly linked locus, recessive-spotting (rs) (10), the region, and provided molecular verification that the W19I affect pigmentation. Ph/+ mice have a belt of white fur mutation is a deletion. Meiotic recombination analysis of an usually extending from the shoulder to the rump (5). White interspecific mouse backcross indicated the following gene spotting in Rw/+ mice is limited to the dorsal sacrolumbar order and distances [in centimorgans (cM)]: centromere- region (6). rs/rs animals display a small amount of white Emv-1-(13 cM)-D4S76-(17 cM)-D5SC25-(5 cM)-a-casein-(l spotting (10). While pigmentation is affected in all, effects on cM)-fi-casein-(6 cM)-afetoprotein-(18 cM)-p-glucuroni- blood and germ cells appear to be minimal or nonexistent in dase. DSSC25, an anonymous locus defined by a mouse brain patch, rump-white, and recessive-spotting mice. Ph/Ph em- cDNA, maps near the map position of W and within the bryos die in utero with a fetal hydrops (5). Rw/Rw homozy- breakpoints of the presumed genetic deletion that causes the gotes also die in midgestation of unknown causes (6). W19H phenotype. Southern analysis of DNAs of W19H/+ Although a great deal is known about the developmental interspecific F1 hybrid mice and somatic cell hybrid lines biology of the individual tissue defects, the nature of the carrying the W19H deletion chromosome showed the deletion of pleiotropism and the molecular mechanism of W gene action DSSC25. In fact, analysis of other mutations at or near the W remain obscure (1-3). The isolation of DNA sequences from locus, which had been transferred from the strain of origin for the W locus could be ofgreat assistance in elucidating Wgene many backcross generations, revealed the retention of donor action in these tissues. Since the primary gene product of the restriction fragment length polymorphisms at the D5SC25 W locus is unknown, we could not use current techniques to locus. Such evidence confirms close linkage between D5SC25 clone this gene directly. Therefore, we attempted to identify and W (within 1 cM) and indicates that the D5SC25 cDNA clone cloned sequences that map near enough to be useful in could serve as a starting point in a chromosome "walk" to W isolating W by chromosome "walking" or other indirect and other closely linked loci that affect development. cloning techniques. We have used cDNA and random genomic clones, some of which map to the region of human A large number of mutations that affect pigmentation in the chromosome 4 known to be homologous to mouse chromo- laboratory mouse map in the region of the dominant white- some 5. We have established their positions with respect to spotting (W) locus on mouse chromosome 5. Most of these markers that flank W on mouse chromosome 5. Finally, we are allelic with W, but a few are mutations at one of three have mapped these clones with respect to mutations at or nearby spotting genes (1-6). Most mutations at the W locus very near the W locus, including two spontaneous mutations have pleiotropic effects upon hematopoiesis, gametogenesis, to W, W39 and W44; a putative radiation-induced deletion, and melanoblast development. The two best-studied muta- WI9H; and two of the very closely linked spotting mutations, tions at this locus, W and W', exhibit effects that are typical Rw and rs (2-6, 10, 11). This approach has identified several of the majority of W mutations. The W/+ mouse is viable, cDNA clones in the W region, one of which maps very near nonanemic, and fertile and has a white belly spot, white feet, to W in the W19H deletion. and white tail tip (7). W/W mice usually die late in gestation from a severe macrocytic anemia, but on selected back- grounds a small fraction survives (8). The surviving W/W MATERIALS AND METHODS mice are severely anemic, sterile, and devoid ofpigmentation Analysis of Genomic DNA. The methods involved in pre- in their skin and coat hairs but have fully pigmented eyes paring high molecular weight DNA, restriction endonuclease (black-eyed white). The W'', or "viable-W", allele was so digestion, electrophoresis, Southern transfer, hybridization, named because some WV/WV homozygotes have normal and washing of genomic DNA filters have been described life-spans. WV/Wv mice are black-eyed white, less anemic (12). than W/W animals, and almost always sterile. In addition to Mice. Mice carrying the W19H mutation were obtained from the white-spotting pattern characteristic of W/+ mice, WV/+ Lynn Lamoreux (Texas A & M University). The mutant heterozygotes exhibit diluted pigmentation in the colored stock was originally maintained on a C3H/101 hybrid back- areas of their coats and have a mild macrocytic anemia (9). ground at Harwell (4) but had been backcrossed to JU for W19H, a more recent, radiation-induced mutation, is a three generations and to C3H/HeJ for a single generation at putative deletion (4). Homozygotes die prior to implantation Texas A & M. We obtained a litter from the backcross to but the heterozygotes survive with a phenotype very similar C3H/HeJ. Several WJ9H/+ JU/C3H females were mated to that of W/+ mice. W19H does not complement mutations with partially inbred Mus spretus males obtained from The at W or at patch (Ph) (4), a closely linked spotting gene (5, 6), Jackson Laboratory. WI9H/+ (M. musculus domesticus x

The publication costs of this article were defrayed in part by page charge Abbreviations: RFLP, restriction fragment length polymorphism; payment. This article must therefore be hereby marked "advertisement" HAT, hypoxanthine/aminopterin/thymidine; cM, centimorgan(s). in accordance with 18 U.S.C. §1734 solely to indicate this fact. tTo whom reprint requests should be addressed. 9635 Downloaded by guest on September 28, 2021 9636 Genetics: Geissler et al. Proc. Natl. Acad. Sci. USA 85 (1988) M. spretus)Fl females were backcrossed to C57BL/6J males 5 (see Fig. 1A). By observing the frequencies of meiotic (The Jackson Laboratory) to produce progeny segregating for recombination with these marker loci, we placed three wl9H. previously unlocalized mouse cDNA clones and one random +/+ F1 females from M. musculus x M. spretus matings genomic sequence in the W region on chromosome 5. These and their N1 backcross progeny were provided by Lianne are, listed by order of increasing distance from the cen- Russell (Oak Ridge National Laboratories). Strain 101 mice tromere, D4S76, DSSC25, and the casein gene family (Fig. were purchased from Cumberland View Farms (Clinton, 1B). All these loci map between Emv-J and Afp. The number TN). Animals carrying W, Rw, and rs mutations were of recombinants out of the total number of progeny analyzed provided by Jane Barker and David Harrison (The Jackson for each interval in this multipoint cross is indicated in Fig. Laboratory). 1B. The order of the clones was determined from a simple Somatic Cell Hybrids. Somatic cell hybrid cell lines were multipoint analysis selecting for the order requiring the generated by fusing spleen cells from a WI9H/+ JU/C3H minimum number of multiple crossovers per gamete. The female with a hypoxanthine phosphoribosyltransferase- occurrence of a single recombination between a- and 83- deficient (HPRT-) hamster fibroblast line (CHTG49) by PEG casein genes established that a-casein is the more proximal fusion and hypoxanthine/aminopterin/thymidine (HAT) se- member of this gene family. The amino acid homology and lection as described (13). To isolate subclones that retained tight linkage of the bovine a- and 3-casein genes suggest that the WI9H deletion chromosome but had segregated the they derive from a common ancestral gene (21). Therefore, it wild-type homologue, we selected for clones that had segre- is ofinterest that we observed a recombination between these gated the active mouse X chromosome to enrich for sub- loci in our analysis of a relatively small number of potential clones that had segregated several mouse chromosomes. recombinants. Unfortunately we were unable to detect a HAT selection was removed and cells were grown for 3-4 RFLP for y-casein in this backcross and, therefore, could not days to allow segregation of mouse chromosomes. HPRT- order this member of the family. clones were isolated by growth in normal medium supple- The most likely gene order is as follows: centromere- mented with thioguanine (5 pug/ml). Emv-J-D4S76-DSSC25-a-casein-/3-casein-Afp-Gus. Com- DNA Probes. p7D. John Horowitz provided a subcloned parison of our data with the published genetic map of mouse 4.2-kilobase (kb) Xba I fragment of p7D, a clone of the chromosome 5 demonstrates that D5SC25 and the casein BALB/c Emv-J ecotropic provirus and flanking cellular genes map in the W region (Fig. 1 A and B). DNA (14). A 2-kb Sac I insert fragment comprising flanking DSSC25 Maps in the W19H Deletion. The proximity of these cellular DNA detected a Taq I restriction fragment length genes to the W locus was established by determining whether polymorphism (RFLP) between M. spretus and M. musculus the structure of these loci was altered by the W19" mutation, DNAs that segregated in the backcross offspring. a putative deletion involving 2-7 cM about the W locus (4). p8MD4. This is an anonymous human genomic clone A WJ9H/+ M. musculus domesticus female was crossed to a defining locus D4576 from the short arm of human chromo- +/+ Mus spretus male to produce +/+ and W19H/+ F1 some 4 (24). The 2-kb HindIII insert was used to detect a Taq offspring. The DNAs of +/+ and WI9H/+ littermate off- I RFLP between the two mouse species. Hybridizations with spring were digested with endonucleases that resolved spre- this human probe were performed in 30% (vol/vol) formam- tus and domesticus alleles at loci of interest on chromosome ide at 42°C. 5. Hybridizing restriction fragment(s) derived from the W19H pSC25. This is an anonymous cDNA isolated from a mouse (domesticus) chromosome of the WI9H/+ mouse were com- brain cDNA library defining the locus D5SC25 (24). A 0.4-kb pared with that of the normal (domesticus) homologue of the EcoRI-HincII insert fragment of this clone detected a Taq I +/+ littermate. Any deviation in the size of the WJ9H/+ RFLP segregating in the backcross. domesticus-specific band(s) could be attributed to the influ- Casein cDNAs. Jeffrey Rosen provided cDNA clones for ence of the W9I mutation. As expected, we failed to detect mouse a-, 13-, and y-caseins cloned into the Pst I site of any alteration associated with W19H at loci that map proximal plasmid pBR322 (pCMall, pCM/313, and pCM-yl9, respec- to Pgm-1 or distal to bl-namely, Emv-J (p7D), D4S76, Alb-J, tively) (15). The full-length inserts were used as probes. The Afp, and Gus (data not shown). Similarly, the genomic a- and /3-casein cDNA clones detected Taq I RFLPs segre- sequences corresponding to P-casein (data not shown) and gating in the M. spretus backcross offspring. We were unable a-casein (Fig. 2 Upper), which map 5 and 6 cM proximal to to identify a RFLP for y-casein in this cross. Afp, respectively, are unaffected by the W19H mutation. a-Fetoprotein and albumin cDNA. Shirley Tilghman pro- On the other hand, DSSC25 genomic DNA sequences, vided cDNA clones for mouse a-fetoprotein (Afp) and albu- which map 5 cM proximal to the a-casein gene, are absent min (Alb-1), pmAfp-1 and pmalb-2, respectively (16, 17). from the domesticus chromosome carrying the putative HindIII insert fragments of pmAfp-1 and pmalb-2 were deletion W19H (Fig. 2 Lower). W19H/+ heterozygotes failed employed as the probes in this backcross analysis. pmAfp-1 to manifest novel pSC25-hybridizing fragments, indicating revealed a BamHI RFLP segregating in the backcross ani- that D5SC25 genomic sequences are confined to the deleted mals. region. The identification of this deleted cDNA clone pro- ,8-Glucuronidase cDNA. Roger Ganschow provided the vides molecular evidence confirming that the W19H mutation mouse 83-glucuronidase cDNA clone pGus-1 (18). A 1.5-kb is a genetic deletion. Pst I-HindIII insert fragment was used to detect an EcoRI As further confirmation that D5SC25 maps within the RFLP in the backcross analyses. deletion, a series of somatic cell hybrids, some carrying the deletion chromosome, were derived by fusing CHTG49 RESULTS hamster fibroblasts with spleen cells from a WJ9H/+ C3H/JU female. The W5OA2T series are of particular interest because Linkage Map of Mouse Chromosome 5 from Meiotic Re- they lack DSSC25 sequence (Fig. 3 Lower) while retaining combination Frequency. Following a recent innovation in markers that flank W-namely, alb-l (Fig. 3 Upper) and gene mapping in the mouse (19), we mated (M. musculus Emv-l (p7D), a-, ,B-, and y-casein genes, Afp, and Gus (data domesticus x M. spretus)F1 +/+ females to M. musculus not shown). Since our meiotic recombination analysis and domesticus + / + males and characterized 85 N1 backcross WJ9H/+ interspecific hybrid data clearly demonstrate that offspring for the presence of spretus-specific polymorphic D5SC25 maps between Emv-J and the a-casein locus and restriction fragments segregating at three marker loci (Emv-J, within the WI9H deletion, the W5OA2T hybrids must have Afp, and Gus) that flank the W region of mouse chromosome retained the WI9H deletion chromosome 5 but segregated the Downloaded by guest on September 28, 2021 Genetics: Geissler et al. Proc. Natl. Acad. Sci. USA 85 (1988) 9637 w Ph rs A Afp alb-1 Gus

24 3 2 16

w 19H

B Emv- 1 D4S76 D5SC25 acas fcas Afp Gus KJ ,I 13 _3 17 +3 5 +1 1 +1 6 +1 18 +3 (11/83) (14/84) (4/87) (1/85) (5/84) (15/85) /~~~~~~~~~ /~~~~~~~~~ ./X / C Rw acas fcas l lIl /\ l 3 <1 l <1 ? I1 / ~~19H

W 2 L 1 Ph D5SC25

FIG. 1. Linkage maps of mouse chromosome 5. (A) Distances between previously mapped markers. Map distances and gene orders were determined in a number of laboratories; the map is updated annually by M. Davisson and T. Roderick (The Jackson Laboratory) (20). The mapping data for the W9I putative deletion shown here were published by Lyon et al. (4). The circle represents the centromere, and the numbers indicate the approximate distances [in centimorgans (cM)] between the loci (the map distances are not drawn to scale). (B) Distances established in the current paper for the seven loci tested by RFLP analysis of 85 N1 backcross offspring of normal (M. musculus domesticus x M. spretus)Fl hybrid females mated to +/+ C57BL/6J males. The estimated distances between loci are indicated in cM below the chromosome map with the standard error (and the number of recombinants expressed as a fraction of the total number of progeny tested). (C) Composite of the meiotic recombination data shown in A and B and the deletion mapping data, depicting the relationship between markers in the W region and the putative boundaries of the WI9H deletion. We estimate the maximum size of the W19H deletion to be -3 cM. The lack of recombination between WI9H and the a-casein gene in 50 potential recombinants suggests that the distance between the distal breakpoint and the a-casein gene is on the order of 1 cM.

normal homologue. Therefore, these data confirm that However, the presence of the WI9H deletion did not signif- DSSC25 maps within the WI9H deletion and demonstrate the icantly reduce the recombination between the P-casein gene usefulness of these cell lines for mapping nonpolymorphic and Afp (data not shown). Therefore, it is unlikely that the loci with respect to the WI9H deletion. WI9H deletion maps in this interval. The most likely location Mapping the Distal W19H Deletion Breakpoint. To confirm for the deletion is proximal to the casein gene family, the position of WI9H proximal to the a-casein gene and consistent with the map position of the deleted DSSC2S estimate the distance between the a-casein gene and the locus. The most likely gene order then is Emv-1-D4S76- distal deletion breakpoint, an interspecific F1 backcross was W'9H-a-casein-g-casein-AfpGus. The absence of a cross- made between WJ9H/+ (musculus x spretus) females and over between W19H and the a-casein gene in 50 potential C57BL/6J males to produce progeny segregating for WJ9H. recombinants indicates that the a-casein gene lies near the The map orders of DNA markers in the two types -of distal deletion breakpoint. interspecific crosses are consistent (data not shown). Since Lack of Recombination Between D5SC25 and Alleles at W, no recombination occurred between either the a- or the rs, and Rw. To obtain direct evidence concerning the prox- ,B-casein gene and W'9H, the gene order of these loci with imity ofDSSC25 and W, we tested several W-region-congenic respect to Afp was not directly verified in this experiment. mouse lines for the occurrence of recombination between Downloaded by guest on September 28, 2021 9638 Genetics: Geissler et al. Proc. Natl. Acad. Sci. USA 85 (1988)

Spretus Fl's rarM-)

- ~~~~~+ < m + (-)(- +4- + ms c1 a) + cs (D + <-) ~ -I :35I V. + 4 3 + + 0) It - = I 23 - m) I -,,I C *w-_ * a casein 6 - 08- a_, pSC25 0.6- -t l

SpretusSpretus Fl1'sF I7FIG. 4. Southern analysis of congenic mouse lines. Genomic + < m DNAs from congenic lines carrying mutations at or very near to W I--+ were digested with Alu I and hybridized with pSC25. Genomic DNAs <, 31 3 + were from C3H.B6-W39/+ (backcross generation N50), B6,C3H- 4.4 - W44/W44 (N4o), B6.C3H-rs/rs (N30), B6-+/+, and C3H-+/+ mice. * * -M In all three lines the DSSC25 RFLP allele of the strain of origin is _1 -s retained. alleles at the D5SC25 locus (Fig. 4). Therefore, the DSSC25 pSC25 locus has been cotransferred with selected mutations at Wfor 2.3 - .9 90 backcross generations. The mean length(s) of the donor chromosomal segment that accompanies a selected locus during the generation of congenic strains has been estimated by Johnson (22) as 1/n, where n = the number of generations of backcrossing beyond F1. The 95% confidence interval is FIG. 2. Southern analysis of interspecific F1 hybrids. Genomic (-Hn 0.975)/n < 1/n < (In 0.025)/n. Therefore, the cosegre- DNAs of two WI9H/+ (M. muscUlUs domesticus X M. spretus)Fl gation of D5SC25 with the selected W mutations for 90 mice, a +/+ F1 littermate, and +/+ C3H/HeJ and +/+ M. spretus generations indicates that D55C25 is within 1 cM of the W controls were digested with Taq I and hybridized with a-casein locus. and (Lower) cDNAs. Deletion of DSSC25 but not of (Upper) pSC25 Furthermore, the B6.C3H-rs line, produced by =30 gen- the a-casein gene is seen in the W19H/+ genomic DNA digests. S and M indicate spretus- and musculus-specific restriction fragments. Size erations of backcrossing, retained the donor allele at D5SC25 markers (kb) are at left. (Fig. 4). DSSC25 has also cosegregated with Rw for three backcross generations involved in the generation of a novel these loci during transfer of the selected W mutation onto a congenic line carrying this mutation (data not shown). These different genetic background. The strain of origin was either observations further validate our assertion that W and C57BL/6J (B6) or C3H/HeJ (C3H). The congenic lines are DSSC25 are very close. designated by the strain onto which the locus has been transferred followed by the strain of origin. Both W congenic DISCUSSION lines tested, B6.C3H-W44 and C3H.B6-W39 (produced at The Jackson Laboratory by 40 and 50 generations of backcross- We have identified cloned cDNAs that are closely linked to gene on mouse chromosome 5, ing, respectively), had segregated donor alleles at Pgm-1, 3 the white-spotting complex established the map positions of these loci with respect to to W not shown), but retained the donor cM proximal (data known markers, and confirmed that WJ9H is a deletion. One of the new markers, pSC25, an anonymous mouse brain W5OA2TGreS cDNA, maps within 1 cM of W and within the WI9H deletion. 3 n a |2 3 4 5 6 7 8 9 The a-casein gene maps just distal to the W19H deletion. These clones should be extremely useful for investigating the 23 - relationship among the several tightly linked developmental 6 m 40 9- 004qwq 1100 loci in this region, as well as the molecular basis of their . f a,s Albumin 6- 4m.0obam A_____~~~~~O W_ control of embryonic development and differentiation. 00 Due to its close proximity to the map position of the W 4- locus, it is possible that the pSC25 cDNA represents the transcript of one of the white-spotting genes such as W, Ph, or rs. All three are distinct with respect to their phenotypic effects but have been shown to be tightly linked by recom- I pSC25 bination analyses. The rs locus maps within 1 cM of W with no recombination events so far observed (10, 11). Recombi- nation frequencies between Ph and W alleles vary from 0.07 FIG. 3. Southern analysis of somatic cell hybrids. Genomic to 0.35, depending on the W allele tested (3). It is unlikely that DNAs of hybrid cell lines produced by fusion of CHTG49 hamster the deleted pSC25 cDNA represents the Rw transcript since fibroblasts with WI9H/+ mouse spleen cells were digested with Msp complementation data of Lyon et al. (4) demonstrate that the I and hybridized with the albumin cDNA pmalb-2 (Upper) and pSC25 W19" deletion does not extend to Rw. (Lower). Included are the following control DNAs: CHTG49 hamster pSC25 was obtained from a mouse brain cDNA library, and (CHTG), C3H/HeJ +/+ (C3H), and JU +/+ (Ju). Eight out of nine although mice carrying mutations at Ph, W, and rs have no of the HAT-resistant paren- thioguanine-resistant (TG'CS) subclones obvious neuropathology, we cannot preclude the possibility tal hybrid clone W5OA2 (data not shown) carry only the WI9' that these genes are expressed in the brain. Deletions deletion chromosome. The other subclone, no. 7, has segregated both mouse chromosomes 5. Two independent HAT-resistant hybrid confined to individual genes would be very useful for gener- clones, W5OC1 and W42A1, retain the normal homologue (the ating a long-range restriction map of the DSSC25 and casein- W5OC1 lane shows products of incomplete endonuclease digestion). gene regions and identifying genomic rearrangements asso- Downloaded by guest on September 28, 2021 Genetics: Geissler et al. Proc. Natl. Acad. Sci. USA 85 (1988) 9639

ciated with mutations at one or more of these white-spotting 1. Russell, E. S. (1979) Adv. Genet. 20, 357-459. loci. Some W alleles may represent deletions confined to W. 2. Silvers, W. K. (1979) Coat Colors ofMice: A Modelfor Gene Mutations of this type would be expected to exhibit the null Action and Interaction (Springer, New York), pp. 206-241. 3. Geissler, E. N., McFarland, E. C. & Russell, E. S. (1981) phenotype represented by the classic W allele. Unfortu- Genetics 97, 337-361. nately, the only known deletion in this region, W'9H, spans 4. Lyon, M. F., Glenister, P. H., Loutit, J. F., Evans, E. P. & W, Ph, and 1(5-') (4), a recessive embryonic lethal mutation Peters, J. (1984) Genet. Res. 44, 161-168. (23). The original complementation data indicated that this 5. Gruneberg, H. & Truslove, G. M. (1960) Genet. Res. 1, 69-90. deletion encompasses between 2 and 7 cM distal to Rw (4). 6. Searle, A. G. & Truslove, G. M. (1970) Genet. Res. 15, 227- The data presented in the current paper allow us to further 235. define the extent of the WJ9H deletion. DSSC25 maps within 7. Fekete, E., Little, C. C. & Cloudman, A. M. (1941) Proc. Natl. Acad. Sci. USA 27, 114-116. the W`9H deletion, and the distal deletion breakpoint maps 8. Coulombre, J. & Russell, E. S. (1954) J. Exp. Zool. 126, 277- between D5SC25 and the a-casein gene, a distance of =5 cM. 296. Since the published genetic map indicates that W is 9 cM 9. Russell, E. S. (1949) Genetics 34, 708-728. proximal to Afp (20) and we have shown that the a-casein 10. Dickie, M. (1966) Mouse News Lett. 35, 31. gene maps 6 cM proximal to Afp and distal to the W19H 11. Southard, A. & Green, M. (1971) Mouse News Lett. 45, 29. deletion, the maximum size of the deletion is -3 cM. Since 12. Kent, R. B., Fallows, D. F., Geissler, E. G., Glaser, T., DSSC25 maps 5 cM proximal to a-casein, DSSC25 should Emanuel, J. R., Lalley, P. A., Levenson, R. L. & Housman, map very near the proximal deletion breakpoint. As this D. E. (1987) Proc. Natl. Acad. Sci. USA 84, 5369-5373. 13. Glaser, T., Lewis, W. H., Bruns, G. A. P., Watkins, P. C., breakpoint separates Rw from the closely linked markers W Rogler, C. E., Shows, T. B., Powers, V. E., Willard, H. F., and Ph, D5SC25 should be near all three spotting loci. Goguen, J. M., Simola, K. 0. J. & Housman, D. E. (1986) Cloned genomic sequences from the W region could be Nature (London) 321, 882-887. used to identify additional expressed sequences that map (i) 14. Horowitz, J. M. & Risser, R. (1985) J. Virol. 56, 798-806. within the W"9H deletion and (ii) within the region of the 15. Gupta, P., Rosen, J., D'Eustachio, P. & Ruddle, F. (1982) J. donor chromosome retained by the W-congenic lines. Can- Cell. Biol. 93, 199-204. didate W sequences could be identified by their differential 16. Tilghman, S., Kioussis, D., Gorin, M. B., Garcia Ruiz, J. P. & expression in developing neural crest, germ cells, and hem- Ingram, R. S. (1979) J. Biol. Chem. 254, 7393-7399. atopoietic tissues or from transcripts of altered size or 17. Kioussis, D., Eiferman, F., van de Rijn, P., Gorin, M. B., Ingram, R. S. & Tilghman, S. (1981) J. Biol. Chem. 256, 1960- concentration in the tissues of W mutants. Ultimately, the 1967. verification of a W-locus clone will rely on its ability to 18. Palmer, R., Gallagher, P. M., Boyko, W. L. & Ganschow, ameliorate one or more of the W-locus tissue defects when R. E. (1983) Proc. Natl. Acad. Sci. USA 80, 7597-7600. introduced into developing W-mutant embryos. 19. Avner, P., Amar, L., Dandalo, L. & Guenet, J. L. (1988) Trends Genet. 4, 18-23. We thank Dr. Elizabeth Russell and Dr. Jane Barker for many 20. Roderick, T. & Davisson, M. (1986) Mouse News Lett. 75, 11- helpful discussions, for the congenic lines used in these experiments, 12. for bringing our attention to the W19H mutation, and for their careful 21. Jones, W. K., Yu-Lee, L.-Y., Clift, S. M., Brown, T. L. & review ofthis manuscript. This work was supported by Public Health Rosen, J. M. (1985) J. Biol. Chem. 260, 7042-7050. Service Grants GM27882, CA17575, NS16367, and NS20012 and by 22. Johnson, L. J. (1981) J. Hered. 72, 27-31. the Hereditary Disease Foundation. E.N.G. acknowledges postdoc- 23. Papaioannou, V. E. (1987) Dev. Genet. 8, 27-34, toral support of Public Health Service Grant CA07826. S.V.C. was 24. Cheng, S. V., Martin, G. R., Nadeau, J. H., Haines, J. L., funded by fellowships from the National Huntington's Disease Bucan, M., Kozak, C. A., MacDonald, M. E., Lockyer, J. L., Association and the Hereditary Disease Foundation. J.F.G. is a Ledley, F. D., Woo, S. L. C., Lehrach, H., Gilliam, T. C. & Searle Scholar of the Chicago Community Trust. Gusella, J. F., Genomics, in press. Downloaded by guest on September 28, 2021