Advanced Inhibition of Undesired Human Hair Growth by PPAR|[gamma]| Modulation|[quest]|

TO THE EDITOR The inhibition of unwanted hair growth (hirsutism, hypertrichosis) by safe and effective agents that do not destroy the hair follicle (HF) remains an important challenge in clinical dermatology. Being interested in identifying alternative therapies to depilation, laser therapy, anti-androgens, and eflornithine (Blume-Peytavi, 2013), we explored whether modulating peroxisome proliferator–activated receptor-γ (PPARγ) may inhibit human hair growth. PPAR activation controls the expression of genes that regulate inflammation and lipid metabolism and exerts strong anti-inflammatory effects (Choi and Bothwell, 2012, Supplementary References S7 and S12 online). Moreover, PPARγ activation stimulates keratinocyte and sebocyte differentiation and also epidermal lipid synthesis and regulates the epidermal barrier homeostasis (Supplementary Figures S3 and S10 online). PPARγ signaling is also required for the maintenance of HF epithelial stem cells in mice, although insufficient PPARγ signaling may predispose patients to develop the inflammatory cicatricial alopecia, lichen planopilaris (LPP; Karnik et al., 2009; Harries et al., 2013). Recently, interest has grown in developing selective PPARγ modulators that retain the desirable therapeutic activities without the unwanted effects of classical agonists (Higgins and Depaoli, 2010). Therefore, we tested a newly designed anti-inflammatory PPARγ modulator, GMG-43AC, a chemical entity with structural similarity to propionic acid, with a favorable toxicological profile (Pirat et al., 2012). GMG-43AC functions as a selective PPARγ modulator (see Supplementary Text and Supplementary Figure S1a–c online). GMG-43AC was administered to microdissected, organ-cultured HFs from a female patient’s scalp, obtained after written informed consent and with the University of Luebeck ethics approval (Kloepper et al., 2010). Its potential anti-inflammatory effects were assessed in normal human keratinocytes (NHKs) exposed to defined inflammatory stimuli. Lactate dehydrogenase (LDH) levels in the HF organ culture supernatants showed only a slight increase by day 6 (Supplementary Figure S2a online), suggesting only low-level GMG-43AC-associated general tissue toxicity. Although the administration of GMG-43AC for 6 days did not alter hair shaft elongation (Supplementary Figure S2b online), HF cycle analyses revealed that all tested GMG-43AC doses strongly decreased the percentage of HFs in anagen, and induced premature HF regression (catagen; Figure 1a): although only 55% of the HFs treated with vehicle transformed into the catagen phase after 6 days, almost all (85%) of the 0.01 mM GMG-43AC-treated HFs were in this phase. In the two higher doses of GMG-43AC, 70% of HFs were in the catagen phase. These results were independently confirmed by measuring the hair cycle score (s2), which reached significance in the 0.01 mM dose (Figure 1b), supporting the hypothesis that GMG-43AC promotes catagen. Although GMG-43AC did not significantly alter the percentage of proliferating hair matrix keratinocytes in anagen VI HFs, it tended to stimulate keratinocyte apoptosis in the hair matrix (significance reached at 0.1 mM; Supplementary Figure S2c online). Next, we explored whether GMG-43AC modulates the expression of two signature keratins expressed by human HF progenitor cells, namely, keratins K15 and K19 (Lyle et al., 1998; Kloepper et al., 2008, Supplementary References S6 and S8 online). Interestingly, GMG-43AC significantly upregulated K15 immunoreactivity in all tested doses in situ (Figure 1c) but did not significantly alter the number of K15-positive cells (Figure 1c). A positive impact on human HF epithelial progenitor cells was independently corroborated by demonstrating that K19 immunoreactivity was also significantly upregulated by GMG-43AC; moreover, the number of K19-positive cells in the lower outer root sheath was increased (Figure 1d). Thus, GMG-43AC may effectively inhibit unwanted human hair growth by inducing catagen while simultaneously preserving (or possibly even promoting) the HF’s epithelial progenitor cell pools. As a first attempt toward identifying early-response candidate target genes of GMG-43AC-mediated stimulation, two independent sets of organ-cultured HFs from a female patient’s scalp were compared by genome-wide DNA microarray analysis (see Supplementary Text online). This identified 39 candidate early-response genes as being substantially upregulated (i.e., >3 fold, P<0.05) in the HFs of both patients investigated by stimulation with 0.01 mM GMG-43AC for 6 hours (Supplementary Table S1 online). Among these upregulated genes, none was a gene recognized to be specific for epithelial HF stem cells. However, COL6A1, which encodes the alpha 1 chain of collagen VI, is known to be expressed in very close proximity to the HF (Watson et al., 2001) and may be involved in HF development (Smith, 1994). Quantitative reverse transcriptase–PCR independently confirmed that GMG-43AC significantly increased COL6A1 mRNA expression (Supplementary Figure S2d online; for extended discussion, see legend of Supplementary Table S1 online). As PPARγ activation elicits anti-inflammatory responses (Choi and Bothwell, 2012), we finally investigated the effect of GMG-43AC on mRNA expression and protein secretion of IL-6 in NHKs treated with IFN-γ or tumor necrosis factor (TNF)-α. As expected, treatment of NHKs with IFN-γ (30 ng ml−1) or TNF-α (10 ng ml−1) for 6 hours markedly increased both IL-6 transcription and protein secretion. Co-administration of GMG-43AC prevented the above effects at the mRNA (Figure 2a and b) and protein levels (Figure 2c and d). Notably, no significant change in the mRNA expression or the secreted amount of IL-6 was seen following treatment with GMG-43AC (Figure 2). Although the anti-inflammatory mechanisms of GMG-43AC remain to be identified, its parent compound, GED-0507-34L, mediates its anti-inflammatory effects, at least in part, by trans-repression of NF-κB signaling (Mastrofrancesco et al., 2013). The current pilot study shows that PPARγ activation via GMG-43AC (Supplementary Figure S1 online) represents an interesting hair growth–inhibitory strategy that constitutes a substantial advancement over currently practiced therapies for managing undesired hair growth: we showed that stimulating PPARγ signaling can effectively abrogate hair growth by premature catagen induction while protecting human HF epithelial stem cells. The latter appears important to maintain the HF and its stem cells as a regeneration resource for optimal skin wound healing (Plikus et al., 2012, Supplementary Reference S14 online). Therefore, if applied together with standard depilatory therapies, PPARγ modulators such as GMG-43AC may allow the optimization of human hair growth inhibition by combining catagen induction with stem cell preservation and anti-inflammatory properties, thus reducing the pro-inflammatory sequelae of depilation. Together with its favorable toxicity profile (Pirat et al., 2012), this renders topically applied GMG-43AC an interesting anti-hirsutism candidate. PPARγ modulators such as GMG-43AC may also be therapeutically useful in cicatricial alopecia such as LPP. Here, LPP progression based on irreversible, inflammation-induced HF stem cell damage (Karnik et al., 2009; Harries et al., 2013) may be inhibited by this anti-inflammatory and stem cell–protective PPARγ modulator. RP and MP have received basic research grants from Giuliani S.p.A., Milano, Italy, and have served in a consultancy function for this company. YR has received travel support from Giuliani S.p.A. The excellent technical assistance of Astrid Becker and Heike Krauth is gratefully acknowledged. This study was supported by a basic research grant from Giuliani S.p.A., Milano, Italy, to RP and MP. SUPPLEMENTARY MATERIAL Supplementary material is linked to the online version of the paper