philin ligand (IPL) cyclosporing A (CsA) is a potent hair growth stimulator in humans and mice. It induces re- versible hypertrichosis in a dose-dependent manner ...
Arch Dermatol Res (1996) 288 : 408–410
© Springer-Verlag 1996
S H O RT C O M M U N I C AT I O N
Ralf Paus · Jörg-Andreas Böttge · Beate M. Henz · Marcus Maurer
Hair growth control by immunosuppression
Received: 19 June 1995
Key words Hair growth · Alopecia · Cyclosporin · Immunosuppressants · FK506 · Rapamycin
Hair loss (alopecia) most often results from disturbances in the cyclic growth and regression activity of hair follicles, including premature termination of active hair growth (anagen) and initiation of hair follicle regression (catagen) [1–3]. Therefore, more reliable and effective hair growth-modulatory drugs – an urgent clinical need as well as a multibillion dollar market – should be well-targeted manipulators of hair follicle cycling. In particular, the development of “hair drugs” that act as catagen blockers is essential to treating the vast majority of patients with alopecia seen in clinical practice [1, 3]. However, this awaits more systematic exploration, and it has not yet been demonstrated that spontaneous catagen can be inhibited at all. Interestingly, the immunosuppressive immunophilin ligand (IPL) cyclosporing A (CsA) is a potent hair growth stimulator in humans and mice. It induces reversible hypertrichosis in a dose-dependent manner, e.g. in transplant recipients, and can stimulate hair regrowth in immunologically mediated hair loss (e.g. alopecia areata) [2, 3]. Previously we have shown that CsA at high doses induces anagen in resting (i.e. telogen) follicles [4], and inhibits pharmacologically triggered massive catagen development [1] and chemotherapy-induced alopecia [5] in mice. We have now studied the blocking of spontaneous catagen by CsA, and the anagen-inducing and/or catageninhibiting properties of two other immunosuppressive IPL, FK506 and rapamycin, and of the non-immunophilin-binding cyclosporin analog CsH. The test drugs were studied in the C57BL/6 mouse model for hair re-
R. Paus (Y) · J.-A. Böttge · B. M. Henz · M. Maurer Department of Dermatology, Virchow Hospital, Humboldt-Universität zu Berlin, D-13353 Berlin, Germany Tel. + 49-30-45065043; Fax + 49-30-45065900
search by assessing defined, hair cycle-dependent cutaneous colour changes that reliably indicate hair cycling in the back skin of mice after repetitive intraperitoneal (i.p.) drug injection, according to a previously described method [1, 4–6]. Anagen induction was tested in mice with all back skin hair follicles in telogen (pink skin), while catagen inhibition was studied in mice that had been induced to enter anagen by depilation (black skin) (for details, see legend to Fig. 1). High doses of CsA (3 × 50–250 mg/kg) or FK506 (3 × 0.25 – 5 mg/kg) significantly induced anagen and retarded spontaneous catagen development. Seven days after the first injection of 125 mg/kg CsA to mice with all back skin follicles in telogen, 88% of the mice (14/16) showed macroscopic signs of anagen induction, as opposed to 25% of vehicle-injected control mice (5/20). Nine days after the first injection of 2.5 mg/kg FK506, 50% of the mice (8/16) displayed anagen induction (controls 25%). With regard to catagen induction, 19 days after depilation, 85% (17/20) of all vehicle-treated control mice had spontaneously re-entered telogen via catagen in a defined reference area of their back skin, as opposed to 6% (1/16) of mice treated with 3 × 125 mg/kg CsA and 31% (5/16) of mice treated with 3 × 1.25 mg/kg FK506. In contrast, neither CsH (3 × 50–250 mg/kg) nor rapamycin (3 × 0.25 – 5 mg/kg) injected i.p. significantly induced anagen or inhibited catagen. Figure 1 a shows catagen inhibition by CsA, Figure 1 b shows anagen induction, and Figure 1 c catagen inhibition by FK506, compared with the appropriate vehicle controls. All macroscopic changes were controlled and verified by histology. This is the first time that the inhibition of spontaneous catagen development by any drug has been demonstrated. Anagen induction and catagen inhibition by CsA in mice correspond well with the hypertrichosis seen with clinical CsA administration, since hypertrichosis results from converting vellus into terminal hair follicles (most likely by anagen induction and prolongation in these follicles) and by recruiting a maximal percentage of terminal hair follicles in telogen into the anagen phase [2, 3]. These convincing parallels with CsA modulation of human hair
409
a
b
c
Fig. 1 a–c Hair cycle modulation by CsA and FK506. a Catagen inhibition following CsA injection to anagen-induced mice (days 8, 10 and 12 after anagen induction in telogen mice by depilation; i.e. drug injection was performed when all follicles were in of early to middle anagen VI); b anagen induction following FK506 injection to telogen mice (days 0, 1 and 3); c catagen inhibition following FK506 injection to anagen-induced mice (as in a). Anagen and catagen development in CsH- or rapamycin-treated mice at any of the study time points was not significantly different from control mice (not shown). The drug doses indicated are the doses injected three times per mouse i. p. (for corresponding mg/kg, see text). For each experiment, the test and control groups comprised four or five animals. All experiments were repeated three times and showed very similar trends (i.e. a total of 16–20 mice in four separate experiments were assayed per test drug dose or vehicle control). The percentage of mice in anagen (indicating anagen induction) or telogen (indicating prior catagen development) in a defined reference region of the back skin was assessed for each group separately, and the mean percentage values +/– SEM from all four experiments were calculated (shown in the figure. P-values were calculated using the independent one-tailed Student’s t-test (*P < 0.05, **P < 0.01). Based on the strict coupling of hair follicle pigment production to anagen [1–6], “anagen” was recorded once pink back skin had assumed a greyish-black colour, and “catagen” was recorded once black back skin had reverted to a greyish-pink colour in a selected reference region of the back [1,4–6]. All experiments were initiated with 7–9-week-old female C57BL/6 mice (~ 20 g, Charles River, Sulzfeld, Germany). CsA was used in the form of Sandimmun® (Sandoz), crystalline CsH and rapamycin were a gift from Prof. H. Schreier, Sandoz, Basle, and crystalline FK506 was a gift from Dr. K. Murato, Fujisawa, Munich. Test drugs were dissolved and diluted with olive oil and ethanol (vehicle control). The macroscopic and microscopic appearance of anagen induction/ catagen inhibition was highly comparable to that which we had previously photodocumented [1, 4–9]
growth argue in favour of the concept that CsA may well serve as an effective catagen-blocking agent for human alopecia [1, 3]. However, the serious side effects associated with systemic CsA administration do not justify its general use for trichological purposes. Rather, the signal transduction pathways and the key target genes that underlie the hair growth effects of CsA should be dissected so as to define molecular targets for unrelated, less toxic, but similarly effective hair growth-modulatory drugs. With regard to FK506, analogies between the murine and human systems have to be drawn with greater caution. Here, we report the first hair growth-modulatory effects of systemic FK506, while anagen induction by topical, but not by oral, FK506 has already been described in mice and hamsters [7–9]. In contrast, recent clinical trials have not yet revealed hypertrichosis induction by FK506 [10] (more subtle FK506 effects on human hair follicle growth and cycling in vivo may easily have escaped notice, and await careful analysis). The previously reported failure to obtain anagen induction by systemic FK506 may be related to the oral route, vehicle, dose and time course of FK506 administration chosen, as well as to mouse strain differences [7]. The clinical explorations of FK506 seems justified since, in contrast to CsA, it is also an effective immunosuppressant on topical administration to human skin [11]. Thus, it is reasonable to ask whether topical FK506 with its lower risk of toxicity than systemic CsA or FK506 could be developed into an effective and relatively safe drug for treating severe forms of alopecia
410 Table 1 Hair cycle-modulatory effects of systemically applied test drugs, compared with their immunophilin-binding and calcineurin inhibiting activity [12, 13]. We had previously demonstrated anagen induction by i.p. or topical CsA in C57BL/6 mice [4], while others have recently reported that topical rapamycin or ascomycin, an IPL comparable to rapamycin, have no hair growth-stimulating effects in mice [9], as seen here with systemic rapamycin
Signal transduction
Hair effects
ing of anagen-stimulatory or catagen-inhibitory properties of candidate “hair drugs”. Acknowledgements The excellent technical assistence of R. Pliet is gratefully acknowledged. This study was supported in part by grants to RP from Deutsche Forschungsgemeinschaft (Pa 354/3-2), and from Sandoz Stiftung für Therapeutische Forschung, Nürnberg.
CsA
CsH
FK506 Rapamycin
Immunophilin binding Calcineurin inhibition
+
–
+
+
References
+
–
+
–
Anagen induction Catagen inhibition
+
–
+
–
+
–
+
–
1. Paus R, Handjiski B, Czarnetzki BM, Eichmüller SA (1994) A murine model for inducing and manipulating hair follicle regression (catagen): the effects of dexamethasone and cyclosporin A. J Invest Dermatol 103 : 143–147 2. Rook A, Dawber RP (eds) (1991) Diseases of the hair and scalp. Blackwell, Oxford 3. Paus R (1996) Control of the hair cycle and hair disease as cycling disorders. Curr Opin Dermatol 3 : 248–258 4. Paus R, Stenn KS, Link RE (1989) The induction of anagen hair growth in telogen mouse skin by cyclosporine A administration. Lab Invest 60 : 365–369 5. Paus R, Handjiski BM, Eichmüller S, Czarnetzki BM (1994) Chemotherapy-induced alopecia in mice: inhibition by cyclosporin A, and modulation by dexamethasone. Am J Pathol 144 : 719–734 6. Paus R, Maurer M, Slominski A, Czarnetzki BM (1994) Mast cell involvement in murine hair growth. Dev Biol 163 : 230– 240 7. Yamamoto S, Jiang, H, Kato R (1994) Stimulation of hair growth by topical application of FK506, a potent immunosuppressive agent. J Invest Dermatol 102 : 160–164 8. Jiang H, Yamamoto S, Kato S (1995) Induction of anagen in telogen mouse skin by topical application of FK 506, a potent immunosuppressant. J Invest Dermatol 104 : 523–525 9. Iwabuchi T, Maruyama T, Sei Y, Adachi I (1995) Effects of immunosuppressive peptidylprolyl cis-trans isomerase (PPIase) inhibitors, cyclosporin A, FK 506, ascomycin and rapamycin, on hair growth initiation in mouse; immunosuppression is not required for new hair growth. J Dermatol Sci 9 : 64–69 10. European FK506 Multicentre Liver Study Group (1994) Randomised trial comparing tacrolimus (FK506) and cyclosporin in prevention of liver allograft rejection. Lancet 344 : 423–428 11. Lauerma AI, Maibach HI (1994) Topical FK506 – clinical potential or laboratory curiosity? Dermatology 188 : 173–176 12. Schreiber SL, Crabtree GR (1992) The mechanisms of action of cyclosporin A and FK 506. Immunol Today 13 : 136–142 13. Rao A (1994) NF-ATp: a transcription factor required for the co-ordinate induction of cytokine genes. Immunol Today 15 : 274–281
caused by an autoimmune attack on the follicle, such as alopecia areata, by blocking premature catagen induction and by recruiting telogen follicles back into anagen, while simultanously suppressing the T cell-dominated inflammatory infiltrate. Fully aware of possible differences between murine and human hair follicles with respect to their response to IPL, our study provides new leads to developing more effective “hair drugs” that are less toxic than FK506 or CsA. The current data suggest that the immunophilinbinding of IPL does not suffice to alter hair cycling by IPL, and that calcineurin inhibition is critical (Table 1). Ater binding to their respective immunophilins, CsA and FK 506, but not rapamycin, are thought to act as immunosuppressants by inhibiting calcineurin activity, and thus the phosphorylation status and nuclear translocation of NF-AT [12, 13]. This transcription factor is in turn required for the coordinated induction of apparently rather few, selected cytokine genes such as IL-2, IL-3, Il-4, GMCSF and TNF-alpha [13]. Calcineurin and/or NF-AT activity as well as selected NF-AT-sensitive cytokine genes, therefore, should be promising molecular targets for hair cycle modulation. Before turning to the human system, the C57BL/6 mouse model [1, 3–6], which requires relatively few animals to yield reproducible results (Fig. 1), offers a simple and sensitive in vivo assay for the screen-
