Bad Hair Day: Testosterone and Wnts

Abstract

Androgens have an important role in normal skin physiology, as well as in the pathogenesis of many skin conditions, such as acne vulgaris, hirsutism, and androgenic alopecia. Kretzchumar et al. (2015) investigate the relationship between androgen receptor (AR) signaling and β-catenin/Wnt signaling pathways in murine hair follicles.

The paradoxical role of androgens in hair follicle biology is not completely understood: androgens trigger hair development at puberty but androgenic alopecia (AGA) in later life. In this issue, Kretzchumar et al. (2015) explore the inhibitory role of androgens by defining a reciprocal relationship between activated β-catenin/Wnt and AR signaling within the hair follicle. Their work identifies AR as a negative regulator of B-catenin signaling, a key signaling pathway in hair cycling and development.

The AR belongs to the superfamily of nuclear hormone receptors. Upon binding to the AR, testosterone or its more potent product, 5α-dihydrotestosterone (5α-DHT), undergoes a conformational change, and the ligand/AR complex translocates from the cytosol to the nucleus where it controls transcription of AR target genes. The activity of AR is controlled by coregulatory proteins that influence ligand specificity and DNA-binding capacity (Heinlein and Chang, 2002a). However, the AR may also trigger rapid, non-genomic effects when present in the cytoplasm, including activation of the mitogen-activated protein kinase cascade and regulation of intracellular calcium levels (Heinlein and Chang, 2002b).

The AR is widely expressed in the skin and, in particular, within the androgen-responsive skin appendages: sweat glands, sebaceous glands, and hair follicles. Numerous studies have characterized AR expression in the skin, although a true consensus on AR expression within the skin and its specific cell types is lacking, as the results vary, depending on the models, reagents, and methods used to detect its expression (i.e., qPCR or immunohistochemistry, human vs. mouse). As pointed out by Kretzchumar et al. (2015), differences likely also exist between AR activity in mouse and human skin.

Cutaneous androgen metabolism occurs within sweat glands, sebaceous glands, and hair follicles where androgens are synthesized de novo from cholesterol or through the conversion of dehydroepiandrosterone sulfate (DHEA-S) and dehydroepiandrosterone (DHEA)—weaker circulating androgens produced by the adrenal gland. All isoforms of the enzymes required to produce and degrade testosterone and 5α-DHT are present within the pilosebaceous unit. These include steroid sulfatase, 3β-hydroxysterioid dehydrogenase (HSD), 17β-HSD, 5α-reductase, 3α-HSD, and aromatase. The expression and activity of these enzymes varies between males and females, body location (scalp versus face), and even anatomical structure (hair follicle versus sebaceous gland). For example, type 1 5α-reductase is expressed predominantly within the sebaceous gland, whereas type 2 5α-reductase is located in hair follicles. Differences in expression and activity of these enzymes indicate a tight regulatory process for androgen metabolism within the skin. For a detailed review of androgen metabolism in the skin, see Chen, et al., 2002.

The β-catenin/Wnt signaling pathway is critical to the development of both hair follicles and sebaceous glands. Epidermal stem cells reside within the bulge region of the pilosebaceous unit and can give rise to progeny that differentiate along multiple cell lineages, including epidermal and follicular keratinocytes as well as sebaceous glands. As daughter cells migrate from the bulge region, the Wnt/wingless (Wnt) and Sonic Hedgehog (Shh) signaling pathways are intricately involved in these cell fate decisions. Cells destined to become sebocytes have increased Shh and Myc signaling and decreased Wnt signaling, whereas cells destined to become hair follicles have increased β-catenin/Wnt signaling. In transgenic mouse models, intact Wnt signaling promotes hair follicle differentiation, whereas inhibition of Wnt signaling through the prevention of Lef1/β-catenin interaction leads to sebocyte differentiation. Similarly, inactivating mutations in LEF1 are commonly found in sebaceous tumors. (Takeda et al., 2006) Regulation of the β-catenin/Wnt signaling pathway is key.

In their manuscript, Kretzchumar et al. (2015), investigate the relationship between AR signaling and β-catenin/wnt signaling in mouse hair follicle bulb cells. The authors characterize the expression of β- catenin and AR within the epidermis and the dermis, concentrating on the hair follicle. They observe that AR and β-catenin display almost reciprocal patterns of expression––such that, when β-catenin is expressed within the nucleus, AR is localized to the cytoplasm and vice versa. These findings are especially notable during the anagen phase of the cell cycle, during which nuclear expression of β-catenin was localized to the upper bulb cells of the hair follicle, whereas AR expression was limited to the dermal papillae cells. During the telogen and catagen phases, β-catenin was absent from the nucleus in hair follicle bulb cells, but AR was detected. This shift in subcellular location highlights the potential importance of each signaling pathway controlling the phases of the hair cycle.

To study the role of AR modulation of β-catenin expression and activity, the authors used a combination of in vitro and in vivo methods. In cell culture, they transfect immortalized sebocytes (Seb-E6E7) with the TOPFLASH Wnt reporter system, as sebaceous glands are highly responsive to androgens. They also treat transgenic mouse lines with conditional activated β-catenin (ΔK5ΔNβ-cateninER, ΔK14ΔNβ-cateninER, and ΔK15ΔNβ-cateninER) with a combination of testosterone, an AR activator, or bicalutamide, a potent AR antagonist. Altogether, their data strongly implicate AR as a negative regulator of β-catenin/ Wnt-dependent transcription.

These findings bring up some interesting points. First, what is the balance of AR regulation in hair follicle and sebaceous gland development? The roles that androgens have in the development and growth of sebaceous glands, sebum production, and acne vulgaris are well established (Pochi and Strauss, 1969). Wnt signaling is decreased within progenitor cells that develop into sebocytes (Takeda et al., 2006). As the data shown by Kretzchumar et al. (2015, this issue) demonstrate one mechanism by which androgens may contribute to sebaceous gland hyperplasia is through inhibition of the Wnt signaling pathway. However, sebaceous gland hyperplasia was a more subtle finding in the present study. This is likely because the authors concentrated on androgen treatment effects on the hair follicle and during continuous β-catenin signaling. Future studies focused on the sebaceous gland response, which employ the strategy of combining genetic models (like loss-offunction β-catenin) with pharmacologic treatments (such as androgen agonists and antagonists) may be interesting.

What are the clinical implications of this work? Androgens stimulate terminal beard hair, axillary hair, and pubic hair growth after puberty, yet trigger follicle miniaturization in AGA in later life—the androgen paradox. This paradox can be extended to the present work: is regulation of β-catenin/Wnt pathway by AR in these two polar opposite conditions different? With the high levels of androgens detected within the hair follicle bulb in AGA, it is likely that β-catenin/Wnt signaling is inhibited, consistent with the enlarged sebaceous glands observed in AGA-affected scalp. In contrast, during puberty in the axillae, e.g., modified androgen metabolism and signaling likely allow for preferential development of hair follicles over sebaceous glands. Additional work is still needed to fully understand the androgen paradox.

Besides the above, the mechanistic insights provided by Kretzchumar et al. (2015, this issue), open new biologic questions and therapeutic avenues for the treatment of hair and sebaceous gland disorders. What is the detailed mechanism by which testosterone antagonizes the β-catenin pathway here? As suggested by Kretzchumar et al., it is most likely an indirect mechanism. Prostaglandins are possibly one underexplored area for this indirect regulation. For example, the Ptgs/PGE₂ pathway is implicated gastric tumorigenesis by increasing activation of the Wnt signaling pathway. Prostaglandins also influence the hair cycle, with prostaglandin F2a (PGF_2α) triggering eyelash growth, whereas prostaglandin D2 decreases hair growth and is also significantly elevated in bald scalp of AGA (Garza et al., 2012). In several contexts an enzyme that synthesizes PGD₂ (Ptgds) has been shown to be induced by testosterone. As different prostaglandins have opposing biological effects, PGD₂ may mediate testosterone inhibition of Wnt signaling in AGA. Future studies are needed to test this hypothesis.

Androgens influence many physiological processes including the development of the immune, nervous, skeletal, and muscle systems (Chang et al., 2013). In addition to all the traditional effects on the skin (i.e., development of acne and AGA), androgens also have a key role in wound healing. In mouse models in which AR signaling is inhibited by castration, 5α-reductase inhibition, or AR-null mice, wound healing was accelerated (Ashcroft and Mills, 2002). The findings by Kretzchumar et al., (2015) may provide the explanation for this accelerated wound healing in the absence of AR, in that β-catenin/Wnt signaling is critical for wound repair (Bielefeld et al., 2013). It is interesting to speculate that AR roles in other physiological processes may be tied to its role as a β-catenin/Wnt inhibitor.

All in all, using a combination of in vitro and in vivo models, Kretzchumar et al. (2015), provided us with mechanistic insight into the roles of androgens and β-catenin/Wnt signaling in mouse hair follicles. With follow-up and confirmatory studies in humans, we may be one step closer to understanding the full pathology associated with androgen-mediated skin diseases.

Clinical Implications.

• Testosterone antagonizes the Wnt pathway, which is important for hair function.

• Inhibition of testosterone can enhance Wnt pathway activity.

• While not universally applicable, these results suggest possible mechanisms of androgenetic alopecia pathogenesis.