The molecular basis of hair growth

ARTICLE

The hair follicle, long the exclusive playground of microscopists, has recently come under the scrutiny of molecular geneticists. This comes as no surprise, as many aspects of cell life and death are easily accessible in the hair follicle and congenital disorders of hair growth and differentiation ­ often poignantly ­ illustrate the patho-biological consequences of defects in the intricate mechanisms regulating hair growth.

All this attention has resulted in many exciting new insights into the functioning of the hair follicle. The elucidation of the genetic basis of hair disorders has been and will be of tremendous help in this process. Basic research is beginning to help us understand how the newly discovered genes interact. Both lines of research are beginning to sketch the contours of a molecular map of follicle development.

Here, we will discuss our view of this map and attempt to position a number of key players in hair follicle development. Figure 1 summarises the following discussion.

Initiation of hair growth

All ectodermal appendages, including specialised ones such as ears and eyes, start as an ectodermal placode. This is a circumscript area containing cells that have been assigned a specialised role. The hair follicle is no exception. Early in embryogenesis the skin forms ectodermal placodes. These are areas where ectodermal cells accumulate and ultimately differentiate into the hair follicle and its associated structures [1]. The underlying mesoderm is thought to initiate placode formation. There is evidence from studies in other organisms such as Drosophila that the Delta-Notch pathway, better known for its role in neurogenesis, is involved. By a process known as lateral inhibition [2], cells destined to form the ectodermal part of the hair follicle are differentiated from the rest of the primitive skin or peridermis. Shortly after the placode is separated from its surroundings, further differentiation is initiated. This initiation is partly dependent upon the presence of the Bone Morphogenetic Protein (BMP)-inhibitor Noggin [3] but also on a recently identified TNF-beta family member known as ectodysplasin. Ectodysplasin (or EDA) has two slightly different conformations that bind to related but distinct receptors. The most important type, EDA-A1, interacts with EDAR, a tumour necrosis factor receptor family member [4-6]. Failure to bind in cell culture stops the ectodermal placode from initiating its next stage of development, evagination into the mesoderm [5]. Ectodysplasin seems to be required for the initiation of differentia-tion of epidermal appendages. This is vividly illustrated by the effect of naturally occurring mutations. Humans with defects in EDA-A1 or its receptor EDAR suffer from X-linked hypohidrotic ectodermal dysplasia (Christ-Siemens-Tourraine syndrome) or autosomal dominant/ recessive hypohidrotic ectodermal dysplasia, respectively [7, 8]. Patients suffering from these disorders lack hair and sweat glands and have defective teeth. In addition, they also have dysmorphic features of the mid-face. A mouse mutant called Tabby has mutations in the mouse Ectodysplasin gene [9]. Another mouse mutant called Downless (dl) lacks a functional Ectodysplasin receptor [6]. These mice are good animal models for the human disorders.

The human phenotype partly overlaps that of mice lacking the Egf-receptor and recent work has demonstrated that the Ectodysplasin-EDAR complex probably interacts with the Egf-receptor pathway since high doses of Egf can partly rescue the Tabby phenotype [10]. How this interaction takes place is currently unknown. It is possible that the interaction is on the level of p38/MAP-kinase since both pathways can regulate this protein kinase although in an opposite manner [11, 12]. The Delta-Notch route probably initiates the Ectodysplasin route, but it is not known how this happens.

Direction of growth and why we do not have hairy palms

Upon initiation of hair growth, the axes of the future hair follicle need to be established. This is called polarisation and is necessary to determine the direction of hair growth. The hair follicle achieves this by having the hair bulb grow preferentially on one side of the horizontal axes. The BMPs play a role in the establishment of up versus down [3]. Proximo-distal polarity is probably established by a molecule that goes by the whimsical name Sonic Hedgehog or SHH. It is related to the drosophila gene Hedgehog (Hh), a signalling molecule responsible for the establishment of segment polarity, regulation of limb growth and initiation of neural tube formation in fruit flies [13]. The mouse knockout for SHH has a severe, lethal phenotype with limb and neural tube defects. It is less well known that this mouse also has abnormal hair growth [14, 15]. The hairs are disoriented and do not grow well, showing that sonic hedgehog is required for polarisation of the hair follicle and proper differentiation of various cell groups. Recent data suggest that the dermal papilla is one of these [16]. SHH regulates the expression of other molecules such as Engrailed-1 [17]. If the latter is over-expressed, normal tissue polarity is disrupted resulting in ectopic hair [18]. This particular pathway probably ensures that some areas of our bodies do not grow hair, such as the palms of the hands.

Human disorders characterised by defective sonic hedgehog signalling illustrate that this pathway is operative in human hair growth. The Conradi-Hünermann-Happle syndrome is caused by a defect in (8-7) sterol-isomerase emopamil binding protein, a crucial enzyme in cholesterol biosynthesis [19, 20]. Sonic hedgehog needs to be conjugated to cholesterol for full biological activity [21, 22]. Consequently, in Conradi-Hünermann-Happle syndrome, SHH signalling is defective and patients have, among other symptoms, linear hypotrichosis probably caused by defective differentiation of the inner root sheath.

From orientation to growth and differentiation ­ hair and cancer

SHH binds to a receptor known as patched or PTC [23]. This transmembrane protein is coupled with a 7-transmembrane signal transducer (probably G-protein coupled) called smoothened (SMO) [24, 25]. The signalling process is complex. In the unbound state, PTC inhibits SMO function. Upon binding of SHH to PTC, SMO is released from inhibition and signals to downstream molecules such as engrailed-1 (EN-1, see above) and GLI. Mutations in PTC cause basal cell nevus syndrome [26]. Abnormal hair has not been described as part of this phenotype although in our experience patients have rather coarse and matted hair. The most prominent features of basal cell nevus syndrome are developmental disturbances and basal cell carcinomas, linking this part of the cascade to cell proliferation and differentiation. Upon binding of SHH to the PTC-SMO complex, the latter activates several downstream genes, such as EN-1. This gene regulates hair growth by influencing the expression of WNT (an acronym for Wingless type mouse mammary tumor virus integration site family) genes and vice versa [27]. The WNT genes are related to drosophila wingless, another segment polarity gene. WNT genes bind to a receptor called frizzled that in turn stabilises the beta-catenin-E-cadherin protein complex [28]. This complex binds to the transcription factor lymphoid enhancer-binding factor 1 (TCF/LEF 1) [29, 30]. Mice lacting this transcription factor have a phenotype reminiscent of the human disorder scalp-ear-nipple syndrome [31]. They lack body hair and whiskers and have underdeveloped mammary glands. If beta-catenin expression goes unchecked, as demonstrated by Gat et al. [32], hair continues to grow in extant hair follicles and new follicles arise where they were absent before. Their growth is disoriented and SHH expression is abnormal, suggesting that WNT genes can modulate SHH expression. In addition, pilomatricomas and other hair follicle tumours arise. Interestingly, these tumours can also occur in Gardner syndrome. This disease is caused by mutations in the APC gene that negatively regulates beta-catenin expression [33].

This is another example of cross talk with the epidermal growth factor-receptor (EGFR). Through Src-kinases, EGFR may influence expression of the cytoskeleton protein E-cadherin [34]. The latter is capable of down-regulating beta-catenin expression [35]. In this way, the EGF receptor can fine-tune the activity of the WNT pathway, thereby determining the extent and nature of hair follicle growth and differentiation. The TCF/Lef1 mouse knockout phenotype is a sub-set of the EGFR mouse knockout phenotype. Like the communication between EDAR and the EGF receptor at the level of p38/MAP kinase, the interaction between EGFR and beta-catenin shows that differentiation of the hair follicle is finely regulated and intimately linked to the same pathways that initiate its development. Other genes regulated by WNT genes include the BMP's and the already mentioned SHH, illustrating the continuous cross talk and feedback taking place during development.

Of keratins, curly hair and blocky teeth

The WNT cascade can activate a number of homeobox genes [36]. We speculate that the human distal-less homeobox gene 3 (DLX3) is one of them. Mutations in this gene cause the tricho-dento-osseous syndrome (TDO), characterised by kinky, curly hair, blocky teeth (taurodontism) and abnormal cranial bones [37]. The drosophila distal-less gene was originally identified as a homeodomain gene involved in leg patterning [38]. Several mutant alleles have been described leading to abnormalities of the distal legs and antennae. It is interesting that a gene that is involved in shaping the terminal appendages of a fly is still shaping appendages in humans.

In TDO, the hair is not disoriented (not unruly), but kinky and curly. The nail abnormalities that are associated with the syndrome (splitting of nails) suggest a defective organisation of keratins and/or other intermediate filament molecules. The teeth also have enamel hypoplasia, supporting this hypothesis. Expression of keratins is thus linked to basic steps in hair follicle development through an intermediate gene. Many more such links probably exist, but most have not yet been discovered. Even the one mentioned here is still speculative.

Keratin abnormalities are definitely the cause of the hair disorder monilethrix. Mutations in the so-called "hard" keratins hHb1 and hHb6 were found by Korge et al. and independently by Winter et al. [39-43]. These sulphur-rich keratins form the bulk of the hair cuticle and the fragility of the hair in monilethrix is the obvious consequence of the disturbance of the keratin network. Less obvious is that the scarring alopecia that is part of the phenotype may indicate that the hard keratins also play a role in the structural integrity of the hair follicle itself.

Life and death in the cycling hair follicle

Once the hair follicle is fully developed, it must initiate its cycle of continuous growth and differentiation. In contrast to many other mammals, humans only molt once (shortly after birth). Little is known about the regulation of hair follicle cycling in humans. It is probably influenced (as it is for example in sheep) by sex hormones, light and circadian genes. It is now known that programmed cell death or apoptosis is essential for proper cycling. Recent research has shown that the zinc-finger containing transcription factor hairless (HR) plays a central role in this process. In 1998, Ahmad et al. demonstrated that mutations in the HR gene cause the disease atrichia universalis with papular lesions [44]. In an inbred Pakistani family, newborn children underwent ritual shaving at one week of age. Affected children, though born with hair, never grew any after the shaving. In other words, the first molt did not take place. Affected children were incapable of growing terminal hair or vellus hairs. On the scalp, they developed papular lesions (recent research shows that these are not limited to the scalp). Examination of similar lesions in the homologous mouse Hairless mutant has demonstrated that these lesions are hair follicles that have degenerated to cysts filled with keratinous material. Panteleyev et al. have convincingly demonstrated that this is caused by a failure of the hair follicle to properly localise apoptosis during the telogen-catagen transition [45]. Surviving parts of the hair follicle form the cysts while the remnants disappear leaving apoptotic ghosts behind. More recent work by Soma et al. has shown that apoptosis occurs as early as the anagen-telogen transition [46]. Apparently, it is instrumental in reorganising the hair follicle such that the telogen hair can be ejected. This is consistent with the results of experimental stimulation of hair follicles with the apoptosis-inducing molecule TNF-alpha. Here, apoptosis is de-localised throughout the follicle and the hair is lost [46]. Interestingly, ectodysplasin is a molecule that belongs to the TNF superfamily. Instead of inducing apoptosis, it induces growth. Here again is a tantalising link between events that occur early in hair follicle morphogenesis and those that define its fate later in life. A lack of ectodysplasin signalling as in CST may lead to the phenotype not only through a failure of growth but also through unchecked apoptosis.

In androgenetic alopecia, involuted hair follicles are disposed of by apoptosis. The hairless transcription factor was an excellent candidate for a causative role in this common ailment. Unfortunately, recent data suggest that this is not the case [47].

Long hair, long cycles

Hair follicle cycling is clocked by factors, such as sex hormones, day length and clock genes. We know very little about this process in humans. More data exist for economically important mammals such as sheep, where practical applications have emerged in daily sheep handling-practice [48].

The Angora mutation in cats, mice and hamsters has been known since ancient times and has been thoroughly bred into these animals for the beautiful lustrous coats it generates. In 1994, Hebert et al. demonstrated that the phenotype in mice is caused by mutations that either delete or render dysfunctional the fibroblast growth factor (FGF) 5 gene [49]. This member of the fibroblast growth factor family apparently limits the length of the anagen phase. Despite what their name might suggest, most FGF's are in the business of restraining growth. FGF3 for instance, inhibits bone growth. Mutations that render the receptor FGFR3 constitutionally active cause the disorders achondroplasia and thanatophoric dwarfism [50]. In these skeletal dysplasias, growth has been severely restrained. It is likely that FGF5 works in much the same manner in regulating the length of the anagen phase of the cycle. Whether the molecule has a similar role in regulating the length of human hair is not known. It would be of interest to seek out FGF5 polymorphisms related to hair length in a human population.

Holding everything together

Hair is usually firmly anchored to the skin. A number of adhesion molecules are involved in the anchorage. The plakophilin-1 (PKP1) gene codes for a component of the desmosome. It is mutated in the ectodermal dysplasia-skin fragility syndrome [51]. One of the features of this syndrome is a scarring alopecia. PKP-1 has been shown to interact with keratins [51]. It is thus possible that it serves as a link between the intermediate filament network and the intercellular connective tissue. A close relative, plakophilin-3, has a desmosomal and a nuclear localisation and contains, like PKP-1, several armadillo repeats marking it as a member of the cadherin family [52, 53]. Cadherins are involved in the regulation of the WNT pathway. Therefore, plakophilin-1 may link the cytoskeleton to gene transcription, implying that ectodermal dysplasia-skin fragility syndrome may also be a disorder of cell-cell communication. This notion is supported by the finding that Clouston syndrome is caused by mutations in connexin-30 [54]. This gene is also known as GJB6 and is a component of gap junctions, intercellular transport channels necessary for cell-cell communication. How disturbances in this process are capable of producing ectodermal dysplasias is not understood.

A complex system

It is obvious that no process functions in isolation and cross-links abound. We showed that there exist links between EDA and EGFR signalling, between EGFR and WNT, between the desmosome and the cytoskeleton and between the cytoskeleton and the transcription machinery. Any linear model such as the one that we outlined here is by its very nature incomplete at best. Nevertheless, it can generate hypotheses that can form the basis for further investigations into the regulation of one of the most fascinating skin appendages, the hair follicle.

CONCLUSION

Acknowledgements This work is supported by grants from Dutch Organisation for Scientific Research (NWO grant number 920-03-085) and Rebirth SA, Luxembourg.

Abbreviations

EGF: Epidermal growth factor
EGFR: Epidermal growth factor receptor
EDA: Ectodysplasin
EDAR: Ectodysplasin receptor
APC: Adenomatous polyposis coli gene product
TCF/LEF 1: Transcription factor/ lymphoid enhancer-binding factor 1
SHH: Sonic hedgehog
PTC: Patched
SMO: Smoothened
GLI: Glioma associated gene
EN-1: Engrailed-1
WNT: Wingless associated mouse mammary tumor family integration site
FRZ: Frizzled
BMP: Bone morphogenetic protein
BMPR: Bone morphogenetic protein receptor
Hr: Hairless

REFERENCES

1. Viallet JP, Dhouailly D. Retinoic acid and mouse skin morphogenesis. I. Expression pattern of retinoic acid receptor genes during hair vibrissa follicle, plantar, and nasal gland development. J Invest Dermatol 1994; 103: 116-21.

2. Pourquie O. Skin development: delta laid bare. Curr Biol 2000; 10: R425-8.

3. Botchkarev VA, Botchkareva NV, Roth W, et al. Noggin is a mesenchymally derived stimulator of hair-follicle induction. Nat Cell Biol 1999; 1: 158-64.

4. Mikkola ML, Pispa J, Pekkanen M, et al. Ectodysplasin, a protein required for epithelial morphogenesis, is a novel TNF homologue and promotes cell-matrix adhesion. Mech Dev 1999; 88: 133-46.

5. Yan M, Wang LC, Hymowitz SG, et al. Two-amino acid molecular switch in an epithelial morphogen that regulates binding to two distinct receptors. Science 2000; 290: 523-7.

6. Headon DJ, Overbeek PA. Involvement of a novel Tnf receptor homologue in hair follicle induction. Nat Genet 1999; 22: 370-4.

7. Kere J, Srivastava AK, Montonen O, et al. X-linked anhidrotic (hypohidrotic) ectodermal dysplasia is caused by mutation in a novel transmembrane protein. Nat Genet 1996; 13: 409-16.

8. Monreal AW, Ferguson BM, Headon DJ, Street SL, Overbeek PA, Zonana J. Mutations in the human homologue of mouse dl cause autosomal recessive and dominant hypohidrotic ectodermal dysplasia. Nat Genet 1999; 22: 366-9.

9. Ferguson BM, Brockdorff N, Formstone E, Ngyuen T, Kronmiller JE, Zonana J. Cloning of Tabby, the murine homolog of the human EDA gene: evidence for a membrane-associated protein with a short collagenous domain. Hum Mol Genet 1997; 6: 1589-94.

10. Isaacs K, Brown G, Moore GP. Interactions between epidermal growth factor and the Tabby mutation in skin. Exp Dermatol 1998; 7: 273-80.

11. Yoon YM, Oh CD, Kim DY, et al. Epidermal growth factor negatively regulates chondrogenesis of mesenchymal cells by modulating the protein kinase C-alpha, Erk-1, and p38 MAPK signaling pathways. J Biol Chem 2000; 275: 12353-9.

12. Read MA, Whitley MZ, Gupta S, et al. Tumor necrosis factor alpha-induced E-selectin expression is activated by the nuclear factor-kappaB and c-JUN N-terminal kinase/p38 mitogen-activated protein kinase pathways. J Biol Chem 1997; 272: 2753-61.

13. Vervoort M. Hedgehog and wing development in Drosophila: a morphogen at work? Bioessays 2000; 22: 460-8.

14. St-Jacques B, Dassule HR, Karavanova I, et al. Sonic hedgehog signaling is essential for hair development. Curr Biol 1998; 8: 1058-68.

15. Chiang C, Swan RZ, Grachtchouk M, et al. Essential role for Sonic hedgehog during hair follicle morphogenesis. Dev Biol 1999; 205: 1-9.

16. Karlsson L, Bondjers C, Betsholtz C. Roles for PDGF-A and sonic hedgehog in development of mesenchymal components of the hair follicle. Development 1999; 126: 2611-21.

17. Zhang XM, Lin E, Yang XJ. Sonic hedgehog-mediated ventralization disrupts formation of the midbrain-hindbrain junction in the chick embryo. Dev Neurosci 2000; 22: 207-16.

18. Loomis CA, Harris E, Michaud J, Wurst W, Hanks M, Joyner AL. The mouse Engrailed-1 gene and ventral limb patterning. Nature 1996; 382: 360-3.

19. Braverman N, Lin P, Moebius FF, et al. Mutations in the gene encoding 3 beta-hydroxysteroid-delta 8, delta 7-isomerase cause X-linked dominant Conradi-Hunermann syndrome. Nat Genet 1999; 22: 291-4.

20. Derry JM, Gormally E, Means GD, et al. Mutations in a delta 8-delta 7 sterol isomerase in the tattered mouse and X-linked dominant chondrodysplasia punctata. Nat Genet 1999; 22: 286-90.

21. Kelley RL, Roessler E, Hennekam RC, et al. Holoprosencephaly in RSH/Smith-Lemli-Opitz syndrome: does abnormal cholesterol metabolism affect the function of Sonic Hedgehog? Am J Med Genet 1996; 66: 478-84.

22. Roessler E, Belloni E, Gaudenz K, et al. Mutations in the C-terminal domain of Sonic Hedgehog cause holoprosencephaly. Hum Mol Genet 1997; 6: 1847-53.

23. Marigo V, Davey RA, Zuo Y, Cunningham JM, Tabin CJ. Biochemical evidence that patched is the Hedgehog receptor. Nature 1996; 384: 176-9.

24. Alcedo J, Ayzenzon M, Von Ohlen T, Noll M, Hooper JE. The Drosophila smoothened gene encodes a seven-pass membrane protein, a putative receptor for the hedgehog signal. Cell 1996; 86: 221-32.

25. Stone DM, Hynes M, Armanini M, et al. The tumour-suppressor gene patched encodes a candidate receptor for Sonic hedgehog. Nature 1996; 384: 129-34.

26. Hahn H, Wicking C, Zaphiropoulous PG, et al. Mutations of the human homolog of Drosophila patched in the nevoid basal cell carcinoma syndrome. Cell 1996; 85: 841-51.

27. van den Heuvel M, Klingensmith J, Perrimon N, Nusse R. Cell patterning in the Drosophila segment: engrailed and wingless antigen distributions in segment polarity mutant embryos. Dev Suppl 1993: 105-14.

28. Toyofuku T, Hong Z, Kuzuya T, Tada M, Hori M. Wnt/frizzled-2 signaling induces aggregation and adhesion among cardiac myocytes by increased cadherin-beta-catenin complex. J Cell Biol 2000; 150: 225-41.

29. Huber O, Korn R, McLaughlin J, Ohsugi M, Herrmann BG, Kemler R. Nuclear localization of beta-catenin by interaction with transcription factor LEF-1. Mech Dev 1996; 59: 3-10.

30. Eastman Q, Grosschedl R. Regulation of LEF-1/TCF transcription factors by Wnt and other signals. Curr Opin Cell Biol 1999; 11: 233-40.

31. Edwards MJ, McDonald D, Moore P, Rae J. Scalp-ear-nipple syndrome: additional manifestations. Am J Med Genet 1994; 50: 247-50.

32. Gat U, DasGupta R, Degenstein L, Fuchs E. De Novo hair follicle morphogenesis and hair tumors in mice expressing a truncated beta-catenin in skin. Cell 1998; 95: 605-14.

33. Munemitsu S, Albert I, Souza B, Rubinfeld B, Polakis P. Regulation of intracellular beta-catenin levels by the adenomatous polyposis coli (APC) tumor-suppressor protein. Proc Natl Acad Sci USA 1995; 92: 3046-50.

34. Calautti E, Cabodi S, Stein PL, Hatzfeld M, Kedersha N, Paolo Dotto G. Tyrosine phosphorylation and src family kinases control keratinocyte cell-cell adhesion. J Cell Biol 1998; 141: 1449-65.

35. Orsulic S, Huber O, Aberle H, Arnold S, Kemler R. E-cadherin binding prevents beta-catenin nuclear localization and beta-catenin/LEF-1-mediated transactivation. J Cell Sci 1999; 112: 1237-45.

36. Thuringer F, Bienz M. Indirect autoregulation of a homeotic Drosophila gene mediated by extracellular signaling. Proc Natl Acad Sci USA 1993; 90: 3899-903.

37. Price JA, Bowden DW, Wright JT, Pettenati MJ, Hart TC. Identification of a mutation in DLX3 associated with tricho-dento-osseous (TDO) syndrome. Hum Mol Genet 1998; 7: 563-9.

38. Cohen SM, Bronner G, Kuttner F, Jurgens G, Jackle H. Distal-less encodes a homoeodomain protein required for limb development in Drosophila. Nature 1989; 338: 432-4.

39. Winter H, Rogers MA, Langbein L, et al. Mutations in the hair cortex keratin hHb6 cause the inherited hair disease monilethrix. Nat Genet 1997; 16: 372-4.

40. Korge BP, Hamm H, Jury CS, et al. Identification of novel mutations in basic hair keratins hHb1 and hHb6 in monilethrix: implications for protein structure and clinical phenotype. J Invest Dermatol 1999; 113: 607-12.

41. Korge BP, Healy E, Munro CS, et al. A mutational hotspot in the 2B domain of human hair basic keratin 6 (hHb6) in monilethrix patients. J Invest Dermatol 1998; 111: 896-9.

42. Korge BP, Healy E, Traupe H, et al. Point mutation in the helix termination peptide (HTP) of human type II hair keratin hHb6 causes monilethrix in five families. Exp Dermatol 1999; 8: 310-2.

43. Zlotogorski A, Horev L, Glaser B. Monilethrix: a keratin hHb6 mutation is co-dominant with variable expression. Exp Dermatol 1998; 7: 268-72.

44. Ahmad W, Faiyaz ul Haque M, Brancolini V, et al. Alopecia universalis associated with a mutation in the human hairless gene. Science 1998; 279: 720-4.

45. Panteleyev AA, Botchkareva NV, Sundberg JP, Christiano AM, Paus R. The role of the hairless (hr) gene in the regulation of hair follicle catagen transformation. Am J Pathol 1999; 155: 159-71.

46. Soma T, Ogo M, Suzuki J, Takahashi T, Hibino T. Analysis of apoptotic cell death in human hair follicles in vivo and in vitro. J Invest Dermatol 1998; 111: 948-54.

47. Sprecher E, Shalata A, Dabhah K, et al. Androgenetic alopecia in heterozygous carriers of a mutation in the human hairless gene. J Am Acad Dermatol 2000; 42: 978-82.

48. Hollis DE, Chapman RE, Panaretto BA, Moore GP. Morphological changes in the skin and wool fibres of Merino sheep infused with mouse epidermal growth factor. Aust J Biol Sci 1983; 36: 419-34.

49. Hebert JM, Rosenquist T, Gotz J, Martin GR. FGF5 as a regulator of the hair growth cycle: evidence from targeted and spontaneous mutations. Cell 1994; 78: 1017-25.

50. Webster MK, Donoghue DJ. Constitutive activation of fibroblast growth factor receptor 3 by the transmembrane domain point mutation found in achondroplasia. Embo J 1996; 15: 520-7.

51. McGrath JA, McMillan JR, Shemanko CS, et al. Mutations in the plakophilin 1 gene result in ectodermal dysplasia/skin fragility syndrome. Nat Genet 1997; 17: 240-4.

52. Bonne S, van Hengel J, Nollet F, Kools P, van Roy F. Plakophilin-3, a novel armadillo-like protein present in nuclei and desmosomes of epithelial cells. J Cell Sci 1999; 112: 2265-76.

53. Schmidt A, Langbein L, Pratzel S, Rode M, Rackwitz HR, Franke WW. Plakophilin 3-a novel cell-type-specific desmosomal plaque protein. Differentiation 1999; 64: 291-306.

54. Lamartine J, Munhoz Essenfelder G, Kibar Z, et al. Mutations in GJB6 cause hidrotic ectodermal dysplasia. Nat Genet 2000; 26: 142-4.