Androgen responsive genes as they affect hair growth

ARTICLE

The hair follicle is unique in that it continuously cycles, undergoing stages of growth, involution, and rest throughout life. It is well known that the target tissue active androgen, dihydrotestosterone (DHT) plays a key role in signaling anagen hair follicles into a miniaturized state, and with successive hair cycles, produces smaller, finer, thinner, indeterminate hairs, hence signaling programmed cell death, apoptosis. Finasteride is a specific 5alpha-reductase type II enzyme blocking agent which inhibits the formation of DHT, and has been shown to be effective in treating men aged 18-41 years, with AGA. Patients taking finasteride show stabilization and improvement in hair growth on the top of the scalp producing thicker, longer hair follicles in 86% of men [1]. The effects of finasteride on the regulation of the hair cycle and apoptosis has not been determined. Since finasteride selectively inhibits DHT production, this study may indicate specific androgen responsive genes, such as those that regulate programmed cell death as they affect hair growth.

Recent experimental evidence suggests that apoptosis is an important event in regulation of the hair cycle as anagen hairs normally grow for 4-7 years, then cycle into resting phases of catagen, telogen. For this cycle to occur, specific apoptotic events take place, which may involve caspases, which are cysteine proteases that play an important role in the effector phase of programmed cell death or apoptosis [2]. The mammalian caspase family currently comprises over 10 known members. These include: caspase-1 (ICE); caspase 2 (ICH-1); caspase 3 (CPP-32, Yama, apopain); caspase 4 (TX, ICH-2, ICE-rel-II); caspase 5 (ICE rel-III, TY); caspase 6 (Mch-2); caspase 7 (Mch3, ICE-LAP3, CMH-1); caspase 8 (MACH, FLICE, Mch5); caspase 9 (ICE-LAP6, Mch6); caspase 10 (Mch-4) [3].

Apoptosis has been frequently observed in a variety of pathological states in these tissues, including alopecia areata, lichen planus, fixed drug eruptions, graft versus host reactions, warts, and neoplasias. We hypothesize that regulation of hair follicles undergoing miniaturization due to the influence of DHT may involve the caspase apoptotic death pathway.

Finasteride's effect in blocking or inhibiting the synthesis of DHT may affect specific caspases involved in the signaling of cell death, which improves hair growth by allowing the hair to return to its normal cycle. We therefore propose to examine whether finasteride interferes with apoptosis in hair follicles via a caspase dependent pathway. Our results are important in that they provide further information about the molecular mechanisms and key regulators of apoptosis in the epidermal and dermal tissues as well as helping to define critical periods for effective therapeutic treatment. A more precise understanding of the mechanisms governing the action of finasteride may help form the basis for further developments of new strategies for successful treatment of hair diseases.

Materials and methods

Ten men, ages 18-35 years, with normal scalp hair (Norwood Hamilton Stage 1), were recruited for donation of 1 (4 mm) punch scalp biopsy. In the next group 10 patients, males aged 18-35 years, had AGA with stages III-V on the Norwood Hamilton Scale. These individuals were biopsied at the screening visit where a 4 mm punch scalp biopsy was obtained from the thinning vertex affected area. These patients were then placed on finasteride 1 mg/day treatment for 6 months, after which a 2nd punch biopsy was obtained for comparison studies. All patients had history and physical exams prior to enrollment in the study, which was approved by the local IRB, and each patient signed informed consent to the study.

Histological analysis

Scalp biopsies from patients were frozen at ­ 40° C and delivered to University of Miami Dept Physiology labs for further processing. Biopsy pieces were cut in half and fixed in formalin, embedded in polyester wax and 10 muM sections cut. Sets of 10 serial sections were placed onto consecutive slides. Eight slides were double stained for caspases or usurpin (flip) and cell type markers for hair and skin cells; the remaining 2 slides were kept in case of technical difficulties. Anti-caspase antibodies against all the known caspases, usurpin/flip were provided by Dr. Donald Nicholson, MerckFrosst, Kirkland, Canada. Cells were double-stained with anti-caspase antibodies followed by biotinylated horse anti-rabbit immunoglobulin (1:100 Vector elite ABC kit) and streptavidin-horseradish peroxidase followed by 3-3'diaminobenzidine (DAB) until a brown reaction product was observed [4]. Sections were immunohistochemically stained with antibodies that identify keratinocytes, fibroblasts, melanocytes, dermal papilla cells of hair follicles, followed by TrueBlue. Double labeled profiles stain deep purple or black when co-localization occurs because the TrueBlue reaction product is blue and DAB is brown. Negative controls were performed in parallel without primary antibody. Specificity of binding was evaluated in control labelings using irrelevant antibodies of the same class, and controls using secondary antibodies alone. For accurate quantification, means of antibody positive cells and differences among experimental groups were analyzed by one-factor ANOVA and Bonferroni's multiple comparison test.

Results

Immunohistochemical findings of caspase expression

Table I summarizes the results of immunohistochemical staining of activated caspases 1-10, usurpin and XIAP in normal human hair from the scalp. Tables II and III summarize the results of positive immunoreactivity for men with AGA affected scalp and the same men 6 months after finasteride treatment, respectively. Histological analysis reveals that caspases 1, 3, 8 and 9 are all detected predominately within the isthmic and upper portion of the hair shaft in both normal and AGA patients. This is the area where the inner and outer root sheaths still surround the hair shaft. Staining is detected in sections of lower isthmus, near the sebaceous duct where the infundibulum commences. Within the catagen phase of the hair follicle cycle, more robust staining for activated caspase-3 was observed than in any other phases of the cycle, suggesting an effector role for caspase-3 in the distal portion of the hair apoptotic pathway. A very different pattern of immunostaining was evident in the hair bulb where weaker immunoreactivity was observed for activated caspase-1, -3, and -9 only.

Overall, in the epidermis, the sebaceous gland and eccrine glands of normal scalp demonstrated positive immunoreactivity for caspases 1, 3, 8 and XIAP. All specimens, whether normal or AGA affected, as well as treated specimens, exhibit the same caspases, however, the levels of expression differ between the groups. In AGA affected tissues, the expression of caspase 1, 3, 8 and 9 is greater then normal, and after 6 months of finasteride treatment, the expression of caspases decreases, similar to the level found in normal scalp.

Discussion

The hair follicle has been described as having 3 stages of growth: anagen, catagen and telogen. Based on morphological evidence, this unique dermal structure undergoes involution, which to a large extent reflects coordinated keratinocyte apoptosis in the regressing proximal hair bulb. In mice, the steady state mRNA levels for some of the gene products implicated in the control of apoptosis, i.e., Fas, transforming growth factor (TGF-beta) and tumor necrosis factor (TNF-beta), rise when a follicle enters into anagen-catagen-telogen transformation of the hair cycle [5]. In sheep, the infusion of epidermal growth factor (EGF) induces synchronized catagen by triggering massive keratinocyte apoptosis in the proximal hair bulb [6, 7], and multiple additional signaling molecules have been implicated in the control of catagen, i.e., fibroblast growth factor (FGF)-5, TGF-beta, insulin-like growth factor (IGF)-1, parathyroid hormone related peptide (PTHrp) [8, 9]. However, the direct role that these factors play in the control of follicle keratinocyte apoptosis in situ has not been clarified.

Our results demonstrate that caspase 1 and 3 are expressed in discrete areas of the hair follicle. It appears that all the caspase family members are expressed in normal cells in an enzymatically inactive pro-form, and that upon the onset of apoptosis they are converted to an enzymatically active processed form via a poorly understood mechanism involving either oligomerization, self-proteolysis or cleavage by another family member or both [10, 11]. To date, only a few different cellular substrates for the caspase family have been identified, suggesting an unusually high substrate specificity for these proteases [10, 12]. All members of the protease family show an extremely high prefe-rence for cleavage of their substrates after an aspartyl residue at the P1 position [13-16]. Once activated, caspases are sensitive to inhibition by the viral products CrmA [17, 18] and p35 [19, 20], and perhaps more significantly from the point of normal regulation, by endogenous inhibitors, X-linked inhibitor of apoptosis (XIAP) [21]. It is unclear why cells that have committed apoptosis would need an inhibitor, so presumably XIAP, which seems to be targeted against caspases 3 and 7 [21], serves to regulate adventitious proteolysis before it has reached the catastrophic threshold [22].

When defining the differences in caspase expression in normal men compared to men with AGA then treated for 6 months with finasteride, it was found that caspase 1, 3, 8 and 9 are present, but that levels of caspases differed in the groups. It may be that DHT affects the balance of cytokines and growth factors that affect specific caspases in the hair follicle, giving more value to systemic inhibition of DHT affecting the hair cycle. Use of finasteride for treating AGA may be to inhibit cellular formation of DHT which may affect caspase levels that induce apoptosis and miniaturize the hair follicle.

The immunohistochemical findings (Table I-III) demonstrate the importance of caspase 3 in normal hair homeostatsis and that different areas of the hair follicle selectively regulate the caspase system which may play an important role in signaling various stages of the hair cycle. Also, finasteride treatment altered caspase expression in the hair follicle. Similarly, caspase 3 plays a key role in apoptosis in neurons and astrocytes [23].

Recent reports have shown that apoptosis contributes to different phases of the hair cycle. It is now firmly established that apoptosis is regulated by an intracellular proteolytic cascade, primarily mediated by members of the caspase family of cysteine proteases, which cleave one another and various key intracellular target proteins to destroy the cell. Two prototypical signalling pathways for the induction of apoptosis have been described [24]. One pathway involves ligation of death receptors that activate procaspase 8 and possibly other initiator caspases. The other pathway is controlled by the mitochondrion and involves the apoptosis protease inducing factor-1 (Apaf-1). Once activated by cytochrome c, Apaf-1, together with cofactor nucleotide triphosphates (dATP, or ATP), they then bind and activate procaspase 9, which in turn cleaves and activates caspase 3 and other downstream caspases. Therefore, caspase 8 and 9 represent the pinnical caspases in the death receptors and cytochrome c/Apap-1 pathways respectively.

Our studies demonstrate that the death of hair follicles involves distinct patterns of expression of active caspases. Active caspase 8, an initiator of the death receptor pathway, was predominately found in the isthmic and upper lower portion of the shaft. This pattern of expression suggests that the death receptor pathway is activated during hair renewal and is initiated by toxic substances that bind to death receptors, i.e., TNF-alpha. Interestingly, activated caspase 3, a downstream effector caspase, was higher in catagen hair then in other phases of the hair cycle, indicating a role in the terminal stage of the apoptotic pathway. Activated caspase 1 was also found in the hair bulb and hair shaft. We did find appreciable changes in the levels of expression of caspase 1 before and after finasteride treatment in AGA, suggesting that caspase 1 may play an important role in inflammation by activating cytokines, as well as mediating apoptosis in normal hair regulation. Previous to this study, the dermal papilla area of the hair follicle was thought to be the main control focus of cell growth and inhibition [25]. The findings from this study also suggest an important role of the upper-lower portion, infundibular area of the hair shaft where inner and outer root sheath are abruptly changing and that this area may play a role in the regulation of normal hair apoptosis. Caspase 3 seems to be playing the key role in the apoptotic pathway during the catagen phase of the hair cycle in these areas. Finasteride may exhibit its influence by selectively inhibiting DHT, which affects a multitude of "androgen responsive genes", such as the caspase pathway, which affects programmed cell death in the hair cycle

CONCLUSION

Acknowledgements

This work was supported by an educational grant from Merck & Co, whitehouse station, NJ, USA.

REFERENCES

1. Kaufman K, Olsen E, Whiting D, et al. Finasteride in the treatment of men with androgenetic alopecia. J Amer Acad Derm 1998; 39: 578-89.

2. Henkart PA. ICE family proteases: mediators of all apoptotic cell death? Immunity 1996; 4: 195-201.

3. Alnemeri ES, Livingston DJ, Nicholson DW, Salvesen G, Thornberry NA, Wong WW, Yuan J. Human ICE/CED-3 protease nomenclature. Cell 1996; 87: 171.

4. Bethea JR, Castro M, Keane RW, Lee TT, Dietrich WD, Yezierski. Traumatic spinal cord injury induces nuclear factor-k/B activation. J Neurosci 1998; 18: 3251-60.

5. Seiberg M, Marthinuss J, Stenn KS. Changes in expression of apoptosis-associated genes in skin mark early catagen. J Invest Dermatol 1995; 104: 78-82.

6. Moore GP, Panaretto BA, Carter NB. Epidermal hyperplasia and wool follicle regression in sheep infused with epidermal growth factor. J Invest Dermatol 1985; 84: 172-5.

7. Hollis DE, Chapman RE. Apoptosis in wool follicles during mouse epidermal growth factor (mECF) induced catagen regression. J Invest Dermatol 1987; 88: 455-8.

8. Stenn KS, Combates NJ, Ellersten KJ, et al. Hair follicle growth controls. Dermatol Clin 1996; 14: 543-8.

9. Raff MC, Barres BA, Burne JF, et al. Programmed cell death and the control of cell survival: lessons from the nervous system. Science 1993; 262: 695-700.

10. Martin SJ, Green DR. Protease activation during apoptosis: death by a thousand cuts. Cell 1995; 82: 1-4.

11. Yang X, Chang HY, Baltimore D. Autoproteolysis activation of pro-caspases by oligomerization. Mol Cell 1998; 1: 319-25.

12. Kayalar C, Ord T, Testa PM, et al. Cleavage of actin by interleukin 1b-converting enzyme to reverse DNase I inhibition. Proc Natl Acad Sci USA 1995; 93: 2234-8.

13. Howard AD, Kostura MJ, Thornberry N, Ding GJF, et al. IL-1 converting enzyme requires aspartic acid residues for processing the IL-1b precursor at two distinct sites and does not cleave 31 kDa IL-1a. J Immunology 1991; 147: 2964-9.

14. Nicholson DW, Ali A, Thornberry NA, et al. Identification and inhibition of the ICE/Ced-3 protease necessary for mammalian apoptosis. Nature 1995; 376: 37-43.

15. Talanian RV, Quinlan C, Trautz S, et al. Substrate specificities of caspase family proteases. J Biol Chem 1997; 272: 9677-82.

16. Thornberry NA, Rano TA, Peterson EP, et al. A combinational approach defines specificities of members of the caspase family and granzyme B. J Biol Chem 1997; 272: 17907-11.

17. Ray CA, Black RA, Kronheim SR, et al. Viral inhibition of inflammation; Cowpox virus encodes an inhibitor of the interleukin-1b-converting enzyme. Cell 1992; 69: 592-604.

18. Gagliardini V, Fernandez P, Lee RKK, et al. Prevention of vertebrate neuronal death by the crm A gene. Science 1994; 263: 826-8.

19. Xue D, Horvitz HR. Inhibition of the Caenorhabditis elegans cell death protease CED-3 by CED-3 cleavage site in baculovirus p35 protein. Nature 1995; 377: 248-51.

20. Bump NJ, Hackett M, Hugunin M, Seshagiri S, Brady K, Chen P, Ferenz C, et al. Inhibition of ICE family proteases by baculovirus anti-apoptotic protein p35. Science 1995; 269: 1885-8.

21. Deveraux QL, Takahashi R, Salvesen GS, Reed JC. X-linked IAP is a direct inhibitor of cell death proteases. Nature 1997; 388: 300-4.

22. Salvesen GS, Dixit VM. Caspases: Intracellular signalling by proteolysis. Cell 1997; 91: 443-6.

23. Keane RW, Srinivasan A, Foster LM, et al. Activation of CPP32 during apoptosis of neurons and astrocytes. J Neurol Res 1997; 48: 168-80.

24. Los M, Wesselborg S, Schulze-Osthoff K. The role of caspases in development, immunity and apoptotic signal transduction. Lessons from knockout mice. Immunity 1999; 10: 629-39.

25. Randall V. The use of dermal papilla cells in normal and abnormal hair follicle biology. Dermatol Clinics 1996; 4: 585-94.