CASPASE-1 LEVEL IS HIGHER IN THE SCALP IN ANDROGENETIC ALOPECIA
– Journal of Dermatologic Surgery-2012 VOLUME 1, Number 7JUAN PABLO DE RIVERO VACCARI, PHD,* MARTY E. SAWAYA, MD, PHD,† FRANK BRAND, III, BS,‡ BERNARD P. NUSBAUM, MD,§ ALAN J. BAUMAN, MD, HELEN M. BRAMLETT, PHD,* W. DALTON DIETRICH, PHD,* AND ROBERT W. KEANE, PHD‡
BACKGROUND AND OBJECTIVES
Inflammasomes that activate caspase-1 govern the innate immune inflammatory response. Whether hair loss associated with androgenetic alopecia (AGA) involves caspase-1 activation is not known.
Immunohistochemical staining for caspase-1 was performed on scalp tissue sections, and protein lysates were analyzed from individuals with AGA (no treatment), and individuals with AGA taking finasteride with apparent hair growth, individuals with AGA taking finasteride without noted hair growth, and normal controls. In vitro studies of human keratinocytes were conducted to establish effects of finasteride, dihydrotestosterone (DHT), and testosterone on caspase-1 levels using immunoblot analysis.
Caspase-1 is expressed in normal human adult epidermal keratinocytes. Caspase-1 expression is greater in men with AGA. In contrast, in men taking finasteride, caspase-1 levels were lower and were similar to those in normal controls. In vitro studies showed that keratinocytes treated with finasteride in combination with testosterone or DHT resulted in a significant decrease in caspase-1 expression.
In vivo and in vitro finasteride treatment resulted in lower caspase-1 expression, supporting the idea that androgens influence innate immunity involved in the hair cycle in AGA. These findings may provide a basis for development of novel treatments for inflammatory skin and hair diseases.
The authors have indicated no significant interest with commercial supporters.
Androgenetic alopecia (AGA) is a common condition affecting up to half of the male population in the United States and has a significant psychosocial effect on patients. The development of AGA depends on several factors, including genetics and androgens, but the role of androgens and how they affect the scalp hair cycle is not well understood. Scalp biopsies from men with AGA reveal an inflammatory component,1 but whether androgens influence inflammatory processes remains unclear. In men with AGA, testosterone is irreversibly converted to dihydrotestosterone (DHT), which is the target tissue active androgen that shortens the scalp hair cycle.2 Two isoenzymes for 5a-reductase type I (5aR1) and type II (5aR2) mediate the conversion of testosterone to DHT. Levels of 5aR2 are higher in scalp hair follicles of individuals with AGA than in normal scalp.2 5aR1 has been localized to sebaceous glands and implicated in exacerbation of acne vulgaris.2 Thus, regional differences in the two isoenzymes may affect local DHT levels in skin.2
Finasteride is one of the approved treatments that the Food and Drug Administration has approved for men with AGA. It specifically inhibits 5aR2 in hair follicles of men with AGA. Men undergoing oral finasteride therapy may see cosmetically acceptable hair regrowth response, especially in the first 2 years of therapy, but for the majority of patients, these results are not long lasting and some individuals become resistant to therapy, and others show no hair growth response at all. Oral finasteride must be continuously administered and does not appear to provide significant improvement in hair regrowth for the majority of patients. More- over, deleterious side effects have been reported in men taking finasteride. Thus, understanding the role androgens in the hair cycle and whether androgens influence inflammation and innate immunity is crucial for development of new thera- pies with fewer side effects and to manage hair loss in AGA and other types of hair diseases.
An essential part of the innate immune response in humans is the assembly of inflammasomes—cyto-solic complexes of proteins that activate caspase-1 to process proinflammatory cytokines such as inter- leukin (IL)-1b. Inflammasome activation plays an essential role in pathologic conditions affecting the skin, including vitiligo,3,4 contact hypersensitivity,5 ultraviolet irradiation,6 periodic fever syndromes with skin involvement,7 and psoriatic lesions.8 In inflamed skin, keratinocytes are a main source of IL-1b, and IL-1b released from keratinocytes induces T-cell development, thereby exacerbating cutaneous inflammation. Our previous work demonstrates that caspase-1 is primarily expressed in keratinocytes and influences hair follicle homeo- stasis,9 but whether steroid hormones and antister- oid hormone therapy influence caspase-1 activation by inflammasomes remains unknown.
In this study, we analyzed the expression pattern of a critical inflammasome-signaling protein (cas- pase-1) in individuals with AGA and normal controls. By using anti-caspase-1 antibodies in AGA- and finasteride-treated individuals, we show that AGA involves activation of the caspase-1 in discrete areas of the scalp. Finasteride decreased the expression of caspase-1. Moreover, in vitro studies revealed that finasteride in combination with testosterone or DHT significantly decreases levels of caspase-1 in human keratinocytes. Thus, our findings may be useful in treatment strategies targeting hair and skin diseases.
Subjects and Methods
The clinical phenotypes of the 18 patients in this study are summarized in Table 1. Five patients were men with AGA receiving no treatment (untreated); seven patients had been on finasteride treatment with positive, apparent hair growth (+ hair growth response); six patients had been on finasteride treatment with no apparent hair growth ( hair growth, nonresponders); and five patients did not have AGA. The hair growth of “nonresponders” was characterized by progression of AGA while subjects received finasteride therapy. Small pieces of human scalp tissue were obtained during routine hair transplant surgery for an elective cosmetic procedure to aesthetically correct for hair loss. Adult men aged 25 to 80 with mild to moderate hair loss on the front and vertex of the scalp signed informed consent to undergo hair transplant surgery. Patients were men diagnosed with AGA between aged 26 to 80 for nontreated patients and 32 to 70 for patients treated with finasteride. The length of treatment ranged from 2 months to 10 years. Individuals with AGA had a Norwood classification of male pattern baldness of III or VI. A large scalp donor strip (1.5 cm wide 9 15–22 cm long) was routinely excised from the occipital area of the scalp, where the hairs are thick and unaffected by the balding miniaturization process. Hairs from the donor strip were microdissected and replanted into the balding frontal and vertex areas of scalp. Small residual pieces of tis- sue, usually 2 to 3 mm2, containing a scant number of intact and fragmented terminal hair follicles were used for analysis. Small excised pieces were obtained from the balding frontal or vertex areas of scalp as new hair follicles were implanted from the donor strip. These small pieces of tissue (1–2 mm) were also used for research analysis and kept frozen at 40°C. The Institutional Review Board at the University of Miami approved the procedures for obtaining scalp tissue. All tissue donors were kept anonymous, without patient identifier information, which was known only to the cosmetic surgeons (BPN and AJB) who performed the procedures.
Frozen scalp specimens were thawed, and a portion was then placed in fixative containing a mixture of 40% formaldehyde, glacial acid, and methanol (1:1:8 by volume) for 48 h. The other portion of the specimen was used for immunoblot analysis. The samples were embedded in paraffin and cut into 10-lm sections. Sets of 10 serial sections were placed onto consecutive slides. Sections were stained with anti-caspase-1 (Imgenex) followed by biotinylated horse antirabbit immunoglobulin (1:1000, Vector Elite ABC kit; Vector Laboratories, Burlingame, CA) and streptavidin-horseradish peroxidase followed by 3,3′-diaminobenzidine until a brown reaction product was observed. Negative controls without primary antibody and controls using an irrelevant antibody of the same isotype were included in each experiment.
Scalp tissues were extracted in lysis buffer and immunoblotted for caspase-1 protein expression using monoclonal anti-caspase-1 (IMG, 5028, Imgenex) as described previously.10
Steroid Treatment of Human Keratinocytes
Primary adult normal human epidermal keratinocytes were purchased from Lonza (Basel, Switzerland) and grown according to the manufacturer’s instructions. Approximately 5 9 105 cells were plated in six-well tissue culture plates and grown for 72 h. Testosterone (4-androsten-17b-OL-3- ONE) and DHT (5a-dihydrotestosterone, 5a-AN- DROSTAN-17b-OL-3-ONE were purchased from Steraloids (Newport, RI), and finasteride was purchased from Sigma-Aldrich. Cultures were treated with 100 nm testosterone, 50 nm DHT, or 5 lm finasteride for 2 h alone. Other cultures were pre- treated with testosterone or DHT for 1 h followed by 5 lm finasteride for 2 h. Cells were washed in Dulbecco’s phosphate buffered saline and lysed, and proteins were analyzed using immunoblotting.
Data are expressed as means ± standard errors of the mean. Statistical comparisons between uninjured and injured groups were made using the one- tailed Student t-test. p < .05 indicated statistically significant differences.
Figure 1 shows the pattern of expression of caspase-1 staining within human hair and the scalp of balding men and in balding men who responded well (+ hair growth) or did not respond ( hair growth) to finasteride treatment. In normal controls, caspase-1 was expressed weakly throughout all levels of the epidermis (Figure 1A), and more-intense staining was consistently observed in basal keratinocytes. In men with AGA (Figure 1B) and men with AGA who were nonresponders to finasteride (Figure 1C), caspase-1 was highly expressed throughout all levels of the epidermis, including the basal keratinocytes. Men with AGA who responded well (+ hair growth) to the treatment (Figure 1D) demonstrated a similar pattern of immunostaining as observed in normal controls (Figure 1A). Weak immunoreactivity was consistently observed in basal keratinocytes, whereas little expression was detected within the superficial layers of the epidermis. In addition to the positive staining of the epidermis, caspase-1 expression was found within cells of the outer root sheath immediately above the hair bulb (Figure 1E, F), but little or no positive staining was observed in hair structures in normal scalp (Figure 1E) of men with AGA who responded well (+ hair growth) to finasteride treatment (Figure 1G).
Figure 1. Immunohistochemical staining of caspase-1 expression in human scalp and hair. Normal scalp (A, E), balding men with androgenetic alopecia (AGA) (B, F); men with AGA responding poorly to finasteride (C, G), and men with AGA responding well to finasteride (D, H). Caspase-1 expression in the epidermis (A–D) and inner and outer root sheath (E–H). Negative controls without primary antibody and controls using an irrelevant antibody of the same isotype were included in each experiment.
Immunoblot analysis indicated that caspase-1 expression was significantly greater in nontreated men with AGA and men with AGA not responding ( hair growth) to finasteride (Figure 2), whereas densitometric analysis of immunoblots indicated that men who responded with hair growth to finasteride demonstrated significantly lower caspase-1 levels. Therefore, it appears that finasteride has effects on caspase-1 activation or processing.
Figure 2. Representative immunoblots of hair and scalp tissue blotted for caspase-1. Patients 1 to 5, with androgenetic alopecia (AGA), untreated; patients 1 to 7, with AGA treated orally with finasteride but responding poorly (non responders); patients 1 to 6, with AGA treated orally with finasteride and responding well (responders). Densitometric analysis revealed that patients who respond well to finasteride treatment have significant lower caspase-1 levels, whereas patients who respond poorly to finasteride treatment have levels similar to those expressed in untreated subjects with AGA. b-actin was used as a standard and protein-loading control. Data are shown as means ± standard errors of the mean. *p < .05.
￼Figure 3. Human epidermal keratinocytes were treated with 100 nm testosterone (T), 50 nm dihydrotestosterone (D), or 5 lm finasteride (F) alone for 2 h or pretreated with 100 nm testosterone or 50 nm dihydrotestosterone for 1 h followed by treatment with 5 lm finasteride for 2 h. Finasteride in combination with steroid hormones decreased caspase-1 activation significantly more than in controls and steroid treatment alone. b-actin was used as a standard and protein loading control. Data are shown as means ± standard errors of the mean. *p < .05.
To determine whether finasteride treatment influenced caspase-1 levels in human keratinocytes, we cultured primary adult human epidermal keratino-cytes and treated cultures with testosterone, DHT, and finasteride or various combinations of these steroids and immunoblotted for caspase-1. As shown in Figure 3, treatment with 100 nm testosterone, 50 nm DHT, or 5 lm finasteride alone for 2 h did not significantly alter levels of caspase-1 in human keratinocytes, but pretreatment of keratinocytes with 100 nm testosterone or 50 nm DHT for 1 h followed by treatment with 5 lm finasteride for 2 h significantly decreased levels of caspase-1 in keratinocytes, suggesting that finasteride in combination with these steroid hormones influences caspase-1 activation by inflammasomes. Thus, the measurement of levels of expression of caspase-1 in hair and scalp tissue maybe used for determination of drug efficacy in treatment of AGA.
This study showed that caspase-1 of the innate inflammasome pathway in human adult epidermal keratinocytes is expressed in balding areas of scalp tissue of patients with AGA and that finasteride treatment results in less expression of this inflammasome protein. Moreover, in vitro studies revealed that finasteride in combination with testosterone or DHT significantly decreased levels of caspase-1 in human keratinocytes. Because there is a perifollicular lymphohistiocytic inflammatory pattern in the infundibular-isthmus area of the hair follicle in 40% of scalp biopsies with AGA,9 it appears that caspase-1 may regulate an upstream innate inflammatory mechanism regulating the hair cycle and AGA.
Dihydrotestosterone (DHT) is the primary androgen influencing miniaturization of hair follicles on the scalp in men with AGA. With successive hair cycles, DHT triggers genetically susceptible scalp hair follicles to produce smaller, thinner, finer hairs, eventually initiating apoptosis. Early in the AGA process, perifollicular lymphohistiocytic inflammatory cells are found surrounding the infundibular-isthmus area of the hair follicle in scalp tissue of men with AGA, indicating that inflammation plays an important role in this disease process.11 Our finding that caspase-1 expression is greater in hair follicles is in agreement with our earlier work9 and suggests that inflammatory caspases may be involved in AGA. Men taking finasteride (5a-reductase type II inhibitor that blocks production of DHT) with a noted hair growth positive response indicates that suppression of DHT reduces caspase-1, suggesting that steroids, especially androgens, influence caspase-1 levels, which may facilitate the miniaturizing pro- cess, eventually leading to apoptosis. Biologically, caspase-1 is regulated at the transcriptional and inflammasome level.7 On the inflammasome level, caspase-1 is activated upon recognition of certain molecular patterns, including Toll-like receptors (TLRs). Several pathogen-, danger-, and disease- associated molecular patterns that activate inflammasomes have been described,7 but the disease-associated molecular patterns in lesional skin in AGA are not known. Our findings suggest that, in epidermal AGA keratinocytes, disease-associated molecular patterns activate caspase-1 in inflammasomes, leading to IL-1b production. In support of this idea is a recent report that testosterone may serve as a trigger for inflammation and innate immunity by reducing expression of TLR4 on a macrophage cell line and TLR4 expression on monocytes and macro- phages in vivo.12 Moreover, androgens such as DHT influence sebocytes and infundibular keratinocytes and play an important role as active participants in innate immunity in acne pathogenesis.13 Our results indicate that androgens such as DHT influence innate immune inflammatory mechanisms involving caspase-1 activation. These findings suggest that endocrine–immune interactions play a crucial role in scalp tissue and that steroid interaction with the innate immune response may be an event in the inflammatory process.
Recently, it has been shown that hyperactivation of caspase-1 mediates an inflammatory cell death program known as pyroptosis,14 suggesting that this inflammatory cell death process may be involved with the etiology of AGA. Our findings demonstrate that AGA involves inflammasome activation, supporting the idea that inflammasomes participate in the etiology of a variety of diseases affecting the skin and hair. Such information not only sheds light on the mechanisms underlying innate immunity in hair loss, but also provides the basis for development of novel treatments for inflammatory skin and hair diseases.
We thank all patients who participated in this study.
1. Jaworsky C, Kligman AM, Murphy GF. Characterization of inflammatory infiltrates in male pattern alopecia: implications for pathogenesis. Br J Dermatol 1992;127: 239–46.
2. Alsantali A, Shapiro J. Androgens and hair loss. Curr Opin Endocrinol Diabetes Obes 2009;16:246–53.
3. Jin Y, Birlea SA, Fain PR, Spritz RA, et al. Genetic variations in NALP1 are associated with generalized vitiligo in a Romanian population. J Invest Dermatol 2007;127:2558–62.
4. Jin Y, Mailloux CM, Gowan K, Riccardi SL, et al. NALP1 in vitiligo-associated multiple autoimmune disease. N Engl J Med 2007;356:1216–25.
5. Watanabe H, Gaide O, Petrilli V, Martinon F, et al. Activation of the IL-1beta-processing inflammasome is involved in contact hypersensitivity. J Invest Dermatol 2007;127:1956–63.
6. Feldmeyer L, Keller M, Niklaus G, Hohl D, et al. The inflammasome mediates UVB-induced activation and secretion of interleukin-1beta by keratinocytes. Curr Biol 2007;17:1140–5.
7. Martinon F, Mayor A, Tschopp J. The inflammasomes: guardians of the body. Annu Rev Immunol 2009;27: 229–65.
8. Kuhn JP, Evert M, Friedrich N, Kannengiesser S, et al. Noninvasive quantification of hepatic fat content using three- echo dixon magnetic resonance imaging with correction for T2* relaxation effects. Invest Radiol 2011;46:783–9.
9. Sawaya ME, Blume-Peytavi U, Mullins DL, Nusbaum BP, et al. Effects of finasteride on apoptosis and regulation of the human hair cycle. J Cutan Med Surg 2002;6:1–9.
10. de Rivero Vaccari JP, Lotocki G, Marcillo AE, Dietrich WD, et al. A molecular platform in neurons regulates inflammation after spinal cord injury. J Neurosci 2008;28:
11. Whiting DA HF. Color atlas of differential diagnosis of hair loss. Cedar Grove, NJ: Canfield Publishing; 1996.