SHORT COMMUNICATION

Molecular Basis for the rhino (hrrh-8J) Phenotype: A NonsenseMutation in the Mouse hairless Gene

Wasim Ahmad, Andrei A. Panteleyev, John P. Sundberg,* and Angela M. Christiano†,1

Department of Dermatology and †Department of Genetics and Development, Columbia University, College of Physicians and Surgeons,New York, New York 10032; and *The Jackson Laboratory, Bar Harbor, Maine 04609

Received May 5, 1998; accepted July 28, 1998

The hairless (hr) and rhino (hrrh) mutations are au-tosomal recessive allelic mutations that map to mouseChromosome 14. Both hairless and rhino mice have anumber of skin and nail abnormalities and develop astriking form of total alopecia at approximately 3–4weeks of age. The molecular basis of the hairlessmouse phenotype was previously found to be the re-sult of a murine leukemia proviral insertion in intron6 of the hr gene that resulted in aberrant splicing. Inthis study, we report a 2-bp substitution in exon 4 ofthe hr gene in a second allele of hr, rhino 8J (hrrh-8J),leading to a nonsense mutation. These findings docu-ment the molecular basis of the rhino phenotype forthe first time and suggest that rhino is a functionalknock-out of the hr gene. © 1998 Academic Press

Formation of the hair follicle involves a complexseries of reciprocal interactions between the dermisand the epidermis, resulting in the formation of anepidermal placode, the hair plug, and the dermalpapilla, and finally, resulting in the differentiation ofhair matrix cells to form the inner root sheath andhair shaft of the follicle (6). In humans, the hairfollicles cycle in a mosaic pattern such that folliclesare present in all stages from anagen to telogen at alltimes, and in principle, each follicle cycles in anasynchronous manner with respect to its neighbor-ing follicles. In contrast, mice shed hair in a progres-sive, synchronized pattern, beginning at the headand progressing caudally (15). A number of muta-tions causing generalized hair loss occur in the lab-oratory mouse. The most widely studied include thehairless (hr), rhino (hrrh) and nude (Hfh11nu) muta-tions. The hairless and rhino mutations are autoso-mal recessive allelic mutations that map to mouseChromosome 14 (4, 10, 14).

The rhino phenotype is a more severe manifestationof the hairless mutation. Both hairless and rhino mice

have normal skin and hair follicles at birth. At about14 days of age, homozygous mutant mice can be recog-nized by loss of periorbital hair, and during the thirdweek of life, they become essentially naked except for afew scattered hairs (Fig. 1a). The cellular defect be-comes apparent at the end of the first catagen stage ofthe hair cycle in rhino mutant mice. The dermal papillafails to follow the contracting follicle and instead be-comes stranded deep in the reticular dermis. As aconsequence, the papillae do not become reassociatedwith the overlying follicles to initiate the second haircycle, resulting in the formation of utricles and deepdermal cysts (Fig. 1b), which are generally lined bystratified squamous epithelium and interspersed indi-vidual sebaceous cells (9).

The phenotype of the hairless mutant resemblesthe human diseases known as congenital alopeciauniversalis (MIM 203655) and papular atrichia(MIM 209500), which are likely to represent thesame disorder (14), and which are currently desig-nated generalized atrichia. Recently, the gene for thehuman disease was linked to chromosome 8p21 (1,11), the human homologue of hr was cloned andmapped to the same locus, and a mutation was iden-tified in a large Pakistani family with this phenotype(1). The hairless gene product encodes a putativesingle zinc finger transcription factor with restrictedexpression in the brain and skin of mouse, rat, andhuman (1, 2, 17). The function of the hairless proteinin the brain remains unknown, however, the geneappears to be directly regulated by thyroid hormonein the brain, where it is postulated to function in anautoregulatory fashion (18).

The first of the hairless alleles, carried in the strainHRS/J, was found to be the result of insertion of amurine leukemia provirus in intron 6 of the hairlessgene, which is thought to interfere with mRNA splicing(2, 13). Here, we report the presence of a homozygousnonsense mutation in the coding region of hr in asecond allele, the rhino mouse (B10.D2/nSnJ-hrrh-8J).

Genomic DNA from rhino and wildtype mice wasobtained from The Jackson Laboratory (Bar Harbor,

1 To whom correspondence should be addressed at Department ofDermatology, Columbia University, 630 West 168th Street VC-1526,New York, NY 10032. Fax: (212) 305-7391. E-mail: amc65@columbia.edu.

GENOMICS 53, 383–386 (1998)ARTICLE NO. GE985495

383

0888-7543/98 $25.00Copyright © 1998 by Academic Press

All rights of reproduction in any form reserved.

384 SHORT COMMUNICATION

ME). PCR amplification from genomic DNA was per-formed using primers based on the available mousehr gene sequence (GenBank Accession No. Z32675).The primers used to amplify a 2.5-kb PCR productcontaining exons 4 – 6 were as follows: sense,59CGAGATGGCAGGATTAGGCT39; and antisense,59GCGACCACAGGCTACACACA39. PCR was car-ried out under standard conditions: denaturation at94°C for 1 min, annealing at 60°C for 1 min, exten-sion at 72°C for 1 min, in an OmniGene ThermalCycler (Marsh Scientific, Rochester, NY). The PCRproducts were electrophoresed in a 1–2% agarose gelin 13 TBE buffer. The PCR products were elutedfrom the agarose gel with the QIAquick gel extrac-tion kit (Qiagen Inc., Santa Clarita, CA) and se-quenced on both strands using the ABI Prism Rho-damine Terminator Cycle Sequencing ReadyReaction Sequencing Kit and the ABI Model 310DNA Sequencer (PE Applied Biosystems, FosterCity, CA). The analysis of the resulting sequencesrevealed a 2-bp substitution at nucleotide positions1910 and 1911 of exon 4, resulting in a glutamine tohistidine amino acid substition at position 511(Q511H) and a lysine to a nonsense codon at the

adjacent amino acid 512 (K512X) (Fig. 2a), creating anew restriction enzyme site for the endonucleaseMseI. The PCR product was digested with MseI (NewEngland Biolabs) at 37°C, and products were sepa-rated on a 1.5% agarose gel (Fig. 2b), confirming thepresence of the mutation.

The rhino mouse displays markedly redundantskin folds, presumably due to distention of the follic-ular infundibulum with laminated cornified mate-rial, resulting in the formation of utriculi or comedo-like dilations of the follicles (7). A number ofbiochemical studies point to gross abnormalities inlipid metabolism in the rhino mouse (3, 8), includinga unique sensitivity to carcinogens, UV-induced skinneoplasms, dioxin toxicity (5, 12, 16), and respon-siveness to retinoic acid analogues (7). The identifi-cation of a nonsense mutation in the murine hairlessgene in association with the severe rhino phenotypefurther supports the important role of the hr gene inhair follicle physiology and suggests that the rhinomouse may provide a useful experimental model sys-tem for some genetically controlled forms of alopeciain humans.

FIG. 1. (A) Rhino 8J mutant mouse (top) adjacent to a littermate control (bottom) illustrates the generalized alopecia and skin foldingcharacteristic of this mutant. (B) The dermal papilla (DP, arrow) is attached to the base of a normal telogen follicle in a control mouse forcomparison to the rhino histology (H&E, 253 magnification). (C) Microscopically, skin from a 3-month-old male B10.D2/nSnJ-hrrh-8J/hrrh-8J

illustrates the massively dilated infundibular segment of the hair follicle (utricle, U, arrow) as well as smaller, deep dermal cysts (DC, arrow)and the absence of the dermal papilla (H&E, 103 magnification).

FIG. 2. Mutation analysis in the rhino 8J mouse. (A) Direct sequencing of PCR products in wildtype and homozygous mice is shown todemonstrate the segregation of the mutation with the disease phenotype. (B). The 2.5-kb PCR product encompassing exons 4, 5, and 6 of thehairless gene is digested with the restriction enzyme MseI for confirmation of the mutation in the rhino 8J mouse. Lanes 2 and 3 containPCR-amplified genomic DNA from wildtype and rhino 8J mice. Lanes 4 and 5 demonstrate MseI cleaved PCR products from lanes 2 and 3.Molecular weight markers (100-bp fragments) are shown in lanes 1 and 6 for reference.

385SHORT COMMUNICATION

ACKNOWLEDGMENTS

We appreciate insightful discussions with Dr. Catherine Thomp-son, The Johns Hopkins University. This work was supported in partby the National Alopecia Areata Foundation (J.P.S. and A.M.C.) andthe NIH NIAMS Skin Disease Research Center in the Department ofDermatology, Columbia University (NIH P30-AR44535).

REFERENCES

1. Ahmad, W., Ul Haque, M. F., Brancolini, V., Tsou, H. C., UlHaque, S., Lam, H., Aita, V. M., Owen, J., deBlaquiere, M.,Frank, J., Cserhalmi-Friedman, P. B., Leask, A., McGrath,J. A., Peacocke, M., Ahmad, M., Ott, J., and Christiano, A. M.(1998). Alopecia universalis associated with a mutation in thehuman hairless gene. Science 279: 720–724.

2. Cachon-Gonzalez, M. B., Fenner, S., Coffin, J. M., Moran, C.,Best S., and Stoye, J. P. (1994). Structure and expression of thehairless gene of mice. Proc. Natl. Acad. Sci. USA 91: 7717–7721.

3. Davies, R. E., Austin, W. A., and Logani, M. K. (1971). Therhino mutant mouse as an experimental tool. Trans. N. Y. Acad.Sci. 33: 680.

4. Doolittle, D. P., Davisson, M. T., Guidi, J. N., and Green, M. C.(1996). Catalog of mutant genes and polymorphic loci. In “Ge-netic Variants and Strains of the Laboratory Mouse” (M. F.Lyon, S. Rastan, and S. D. M. Brown, Eds.), 3rd ed., Vol. 1, pp.17–854, Oxford Univ. Press, Oxford.

5. Forbes, P. D. (1981). Hairless mice for carcinogenesis studies.In “Proceedings of the 2nd NCI/EPA/NIOSH CollaborativeWorkshop: Progress on Joint Environmental and OccupationalCancer Studies” (H. F. Kraybill, I. C. Blackwood, and N. B.Freas, Eds.), pp. 671–684, U.S. Gov. Printing Office, Washing-ton, DC.

6. Hardy, M. (1992). The secret life of the hair follicle. TrendsGenet. 8: 55–61.

7. Kligman, L. H. and Kligman, A. M. (1989). The skin of the rhinomouse. In “Monographs on Pathology of Laboratory Animals: In-tegument and Mammary Glands” (T. C. Jones, U. Mohar, andR. D. Hunt, Eds.), p.187, Springer-Verlag, Heidelberg, Germany.

8. Logani, M. K., Nhari, D. B., and Davies, R. E. (1977). Compo-sition of novel triesters from the skin of the rhino mutantmouse. Lipids 12: 283.

9. Mann, S. J. (1971). Hair loss and cyst formation in hairless andrhino mutant mice. Anat. Rec. 170: 485–499.

10. Mouse Genome Database (MGD) (1998). Maintained at TheJackson Laboratory. URL: http://www.informatics.jax.org.

11. Nothen, M. M., Cichon, S., Vogt, I. R., Hemmer, S., Kruse, R.,Knapp, M., Holler, T., Ul Haque, M. F., Ul Haque, S., Propping,P., Ahmad, M., and Rietschel, M. (1998). A gene for universalcongenital alopecia maps to chromosome 8p21–22. Am. J. Hum.Genet. 62: 386–390.

12. Poland, A., Knutson, J. C., and Glover, E. (1984). Histologicalchanges produced by 2,3,7, 8-tetrachlorodibenzo-p-dioxin in theskin of mice carrying mutations that affect the integument.J. Invest. Dermatol. 83: 454–459.

13. Stoye, J. P., Fenner, S., Greenoak, G. E., Moran, C., and Coffin,J. M. (1988). Role of endogenous retoviruses as mutagens: Thehairless mutation of mice. Cell 54: 383–391.

14. Sundberg, J. P. (1994). The hairless (hr) and rhino (hrrh) mu-tations, chromosome 14. In “Handbook of Mouse Mutationswith Skin and Hair Abnormalities: Animal Models and Biomed-ical Tools” (J. P. Sundberg, Ed.), pp. 291–312, CRC Press, BocaRaton, FL.

15. Sundberg, J. P., and King, L. E. (1996). Mouse mutations withskin and hair abnormalities as animal models for dermatolog-ical research. J. Invest. Dermatol. 106: 368–376.

16. Sundberg, J. P., Sundberg, B. A., and Beamer, W. G. (1997).Comparison of chemical carcinogen skin tumor induction effi-ciency in inbred, mutant, and hybrid strains of mice: Morpho-logic variations of induced tumors and absence of a papilloma-virus co-carcinogen. Mol. Carcinogen. 20: 19–32.

17. Thompson, C. C. (1996). Thyroid hormone-responsive genes indeveloping cerebellum include a novel synaptogamin and ahairless homolog. J. Neurosci. 16: 7832–7840.

18. Thompson, C. C., and Bottcher, M. C. (1997). The product of athyroid hormone-responsive gene interacts with thyroid hor-mone receptors. Proc. Natl. Acad. Sci. USA 94: 8527–8532.

386 SHORT COMMUNICATION