Arch Dermatol Res (1998) 290 : 574–578
© Springer-Verlag 1998
S H O RT C O M M U N I C AT I O N
Marcus Maurer · Eva M. J. Peters · Vladimir A. Botchkarev · Ralf Paus
Intact hair follicle innervation is not essential for anagen induction and development
Received: 4 March 1998
Abstract Neuropeptides produced, stored and secreted by the unusually dense sensory and autonomic innervation of hair follicles (HFs) can induce hair growth (anagen) and may be involved in hair growth control. To test the role of follicle innervation of HF cycling in vivo, we generated innervation-deficient HFs by unilateral surgical denervation of a defined region of back skin in C57BL/6 mice and assessed its effect on spontaneous and induced anagen development. Successful denervation was demonstrated by the absence of PGP 9.5+ or tyrosine hydroxylase+ nerves and nerve-associated neuropeptides (substance P, CGRP). By quantitative histomorphometry, no significant difference in spontaneous or cyclosporin A-induced anagen development could be detected between sham-operated control skin and denervated skin. Only after hair growth induction by depilation, a discrete, marginally significant retardation of anagen development was apparent in denervated HFs. Thus, even though cutaneous nerves may exert a minor modulatory role in depilation-induced hair growth, they are not essential for normal murine anagen development. Key words Hair growth · C57BL/6 mice · Denervation Increasing insight into the role of nerves in the control of skin immunity, wound healing, and keratinocyte growth [1, 2] has renewed interest in the question of whether or not nerves and the neuropeptides and neurotransmitters
M. Maurer1 (Y) · E. M. J. Peters · V. A. Botchkarev · R. Paus Department of Dermatology, Charité, Humboldt-Universität zu Berlin, D-10117 Berlin, Germany Present address: of Pathology, Division of Experimental Pathology, Research North, Beth Israel Deaconess Medical Center School, P. O. Box 15707, Boston, MA 02215, USA Tel. +1-617-6670638; Fax +1-617-6673616, e-mail:
[email protected]
1 Department
released by them, play any role in hair growth control [3–6]. Skin transplantation and skin organ culture experiments have already suggested that hair follicle (HF) growth (anagen) can proceed even after the severence or ablation of follicle innervation [7, 8]. However, it remains unclear whether standardized, complete nerve ablation of fully mature HFs affects their cycling pattern, specifically the time course of anagen development. Interestingly, surgical or chemical denervation in newborn mouse or opossum skin can affect HF morphogenesis [9, 10]. The fact that anagen development during each new hair cycle recapitulates in part aspects of HF morphogenesis [11], encourages the investigation of whether the selective surgical denervation of a defined skin territory affects spontaneous and/or induced HF cycling in this skin region. We therefore employed the C57BL/6 mouse model for hair research [12, 13] to study whether unilateral surgical denervation of dorsal cutaneous nerves (DCNs) in the T3–T12 dermatomes had any effect on the spontaneous, pharmacologically induced or depilation-induced murine anagen development of back skin HFs. Female C57BL/6 mice in the telogen stage of the hair cycle (6–9 weeks old) were purchased from Charles River, Germany, and were housed in communal cages with 12-h light periods. A denervation technique previously described for rat back skin [14] was employed, and a 2.5– 3 cm midline incision was made in the dorsal skin under anaesthesia. The DCNs were exposed under a dissection microscope and DCNs T3–12 on the right side were removed from close to their exit point from the body wall to their entry into the skin. The skin was closed with 9-mm steel wound clips and the completeness of denervation was verified by testing the appropriate skin region for pinch sensitivity [14]. Three standard hair growth assays were employed: depilation-induced (1), pharmacologically induced (2), and spontaneous initiation (3) of anagen. (1) By depilation: anagen was induced in the back skin of mice in the telogen stage of the hair cycle by rosin/wax depilation as previously described [13]. This procedure results in the de-
575 Fig. 1 A–H Surgical denervation results in the complete loss of hair follicle innervation. A, C, D, E: Detection of PGP 9.5-IR, SP-IR, CGRP-IR, and TH-IR nerve fibres by immunofluorescence, respectively, in the back skin of mice 14 days after surgical denervation (B, D, F, H = sham operated control skin). Abbreviations: s = sebaceaous gland, e = epidermis, hs = hair shaft, ors = outer root sheath, d = dermis, c = nerve fibre cable. Arrows B: small = epidermal fibres, large = longitudinal perifollicular nerve fibres, arrowheads = circular perifollicular nerve fibres; D: small = free nerve endings in dermis, arrowhead = circular perifollicular nerve fibres; F: small = free nerve endings in dermis; H: small = free nerve endings in dermis, large = cable
velopment of mature anagen VI HFs within 9 days and the appearance of uniformly black hair shafts within 11–12 days after anagen induction. (2) By cyclosporin: anagen was induced pharmacologically by three intraperitoneal injections of cyclosporin A (CsA) [12]. Mice were treated with CsA (250 mg/kg; Novartis, Basel, Switzerland) on days 0, 1, and 3, which led to an anagen-associated switch in skin color within 5–10 days. (3) Sponta-
neous: mice were subjected to denervation on day 22 after birth (immediately before the spontaneous onset of the first genuine postnatal hair cycle) [18]. All animals exhibited macroscopic signs of spontaneous anagen development within 2 weeks, as is to be expected for mice of this age. Animals were sacrificed during various stages of anagen development and back skin from the paravertebral denervated and sham-operated region was harvested par-
576
A
A
B
B
Fig. 2 A, B Spontaneous development of anagen in denervated and control skin. 2A: x axis = days after denervation of 22-days old mice immediately before the spontaneous onset of the second postnatal hair cycle. y axis = percentage of mice showing anagen associated change of skin color in the denervated and control area of back skin. Data derived from 30 mice tested in three independent experiments. 2B: x axis = days after denervation of 22-days old mice immediately before the spontaneous onset of the second postnatal hair cycle. y axis = progress of hair follicles in the hair cycle as assessed by histomorphometry: left = hair cycle stages: telo = telogen, ana I–VI = anagen I to VI; right = corresponding hair cycle score values: telogen = 1, anagen I to VI = 2 to 7. Values are mean ± SEM (n = 10/timepoint). Statistical analysis was performed by the two-tailed, independent Student’s t-test for paired samples
Fig. 3 A, B Anagen development induced by depilation in denervated and control skin. 3A: x axis = days after induction of a new hair cycle by depilation. y axis = percentage of mice showing anagen associated change of skin color in the denervated and control area of back skin. Data derived from 30 mice tested in three independent experiments. 3B: x axis = days after induction of anagen by depilation. y axis = progress of hair follicles in the hair cycle as assessed by histomorphometry: left = hair cycle stages: telo = telogen, ana I–VI = anagen I to VI; right = corresponding hair cycle score values: telogen = 1, anagen I to VI = 2 to 7. Values are mean ± SEM (n = 10/timepoint). Statistical analysis was performed by the two-tailed, independent Student’s t-test for paired samples (* = P < 0.05)
allel to the vertebral line to obtain longitudinal sections through the HFs. Cryostat sections (15 µm) of perfusionfixed skin biopsies (4% paraformaldehyde and 14% saturated picric acid) were stained for nerves or neuropeptides as previously described [4], using polyclonal rabbit primary antisera against the pan-neuronal marker PGP 9.5, (Paesel and Lorei, Frankfurt, Germany), the adrenergic marker tyrosine hydroxylase (TH), or the neuropeptides substance P or calcitonin gene-related peptide (CGRP) (all from Biogenesis, Poole, UK) as markers for sensory nerves. This was followed by incubation with TRITCconjugated goat antirabbit IgG (Jackson ImmunoResearch, West Grove, Pa.) Sections were examined at 400 × magnification under a Zeiss fluorescent microscope and were photodocumented using a digital image analysis system as previously described [5].
To assess the progress of HFs in the hair cycle by quantitative histomorphometry, skin biopsies were fixed for routine histology (5% buffered formalin, pH 7.4), paraffin-embedded, and processed for routine hematoxylin-eosin staining. Individual HFs were confined to the hair cycle stages telogen or anagen I–VI following the classification of Chase [15]. The percentage of HFs in each defined cycle stage was calculated. Hair cycle score values for the denervated and control skin regions of individual mice were obtained by multiplication of the percentages of HFs in these defined hair cycle stages according to the following score: telogen = 1, anagen I = 2, anagen II = 3, anagen III = 4, anagen IV = 5, anagen V = 6, and anagen VI = 7. The resulting values were then added and divided by the number of HFs staged [16, 17]. In order to demonstrate a loss of functional sensory innervation, the denervated paravertebral skin regions and
577
their sham-operated contralateral counterparts were subjected to daily pinch sensitivity testing for 14 days following neurectomy. This indicated successful sensory denervation immediately after DCN dissection and throughout the observation period which was confirmed histologically by showing that, in sham-operated skin, PGP 9.5+ (pan-neuronal marker), TH+ (adrenergic marker), substance P+ or CGRP+ (sensory neuropeptides) nerves were visible (Fig. 1 A, C, E, G) in their expected, well-defined perifollicular locations [4, 5], yet were completely absent in denervated skin 14 days after the microdissection of DCNs T3–12 (Fig. 1 B, D, F, H). To assess whether intact skin and HF innervation is essential to the normal development of a spontaneous wave of anagen HFs, the back skin of 22-day-old telogen mice, which spontaneously enter their first genuine postnatal hair cycle within about a week [18], was unilaterally denervated by microsurgical DCN dissection. Denervated and control skin regions exhibited essentially identical progress of HFs in the hair cycle as judged by the strictly anagen-coupled change in skin pigmentation [19] from pink to black (Fig. 2 A) and by missing differences in the anagen-associated increase in skin thickness (not shown, see references 16, 20). To detect slight differences in the velocity of anagen development, we then employed quantitative histomorphometry of denervated and control skin sections harvested before as well as 5, 10, and 15 days after denervation. This revealed that the absence of HF innervation did not significantly alter the spontaneous, synchronized transformation of back skin HFs from telogen to anagen VI (Fig. 2 B). Likewise, pharmacologically induced anagen development by intraperitoneal injections of high-dose CsA to adolescent telogen mice [12] was normal in denervated skin, compared to sham-operated control skin (animals with anagen-associated change in pigmentation of control and denervated skin area: 12 of 20 vs 13 of 20 mice on day 6 and 20 of 20 vs 20 of 20 animals on day 12, respectively). Since, however, neither spontaneous nor pharmacologically induced anagen development achieve the degree of hair cycle synchronization that is obtainable with anagen induction by depilation [20], we reasoned that very small hair cycling differences between denervated and normal skin might have been missed in the previous assays, but might still be apparent in depilation-induced hair growth. When a new hair cycle was induced by depilation of telogen follicles in adolescent mice immediately after unilateral denervation had been performed, macroscopically, the development of anagen HFs in denervated skin was virtually indistinguishable from that in control skin (P > 0.05 for all time-points) (Fig. 3 A). However, quantitative histomorphometry revealed slight, marginally significant differences in the velocity of depilation-induced anagen development between denervated and normal HFs (Fig. 3 B). Denervated HFs showed retarded entry into early anagen (anagen I–II) (P < 0.05) and reached the fi-
nal stage of anagen development (anagen VI, day 12 after depilation) slightly later than HFs with intact innervation. It is important to recognize that this study cannot rule out the possiblility of neural mechanisms of hair growth control [6]. Potential negative effects of nerve ablation on follicle growth might be compensated for by reparatory mechanisms, suice surgical deneravation induces a complex cascade of reparatory responses, including for example the upregulation of multiple hair growth-modulatory factors such as nerve growth factor [21] and the induction of sympathetic hyperinnervation [22]). Moreover, nervederived signals might still be exploited for hair growth modulation, even if such signals are not essential for hair growth control. Finally, under pathological conditions, nerve-derived signals could well exert a damaging influence on hair growth (cf. induction of dystrophic catagen in mice by intracutaneous injection of substance P [17]). Our observation that only depilation-induced anagen development showed a slight retardation after denervation may be related to the discrete epidermal and follicular wounding response and the corresponding changes in the cutaneous cytokine milieu which occur immediately after the trauma of anagen induction by wax/rosin plucking [20, 23] (the reparatory cascade induced by denervation and the wounding response induced by depilation might have a slight and transitory growth-inhibitory effect on anagen development during its initial, most rapid and most vulnerable phase of HF transformation). Nevertheless, our study suggests that neural stimuli are not critical elements in hair cycle control, and proves that intact HF innervation is not required for normal anagen induction and development, neither in infantile nor in adolescent mouse skin. Acknowledgements We thank Dr. G. Lewin for most helpful advice with the skin denervation technique, and R. Pliet, E. Hagen, and R. Böhmer for excellent technical assistance. This work was supported in part by a research fellowship (MA 1909/1-1) from Deutsche Forschungsgemeinschaft (DFG) to MM, and by grant Pa 345/6-1 from DFG to RP.
References 1. Ansel JC, Kaynard AH, Armstrong CA, Olerud J, Bunnett NW, Payan D (1996) Skin-nervous system interactions. J Invest Dermatol 106 : 198–204 2. Ansel JC, Armstrong CA, Song I, Quinlan KL, Olerud JE, Caughman SW, Bunnett NW (1997) Interactions of the skin and nervous system. J Invest Dermatol Symp Proc 2 : 23–26 3. Paus R, Heinzelmann T, Schultz KD, Furkert J, Fechner K, Czarnetzki BM (1994) Hair growth induction by substance P. Lab Invest 71 : 134–140 4. Botchkarev VA, Eichmüller S, Pietsch P, Johansson O, Maurer M, Paus R (1997) A simple immunofluorescence-technique for simultaneous visualization of mast cells and reveals selectivity and hair cycle – dependent changes in mast cel. – nerve fibre contacts in murine skin. Arch Dermatol Res 289 : 292–302 5. Botchkarev VA, Eichmüller S, Johansson O, Paus R (1997) Hair cycle-dependent plasticity of skin and hair follicle innervation in normal murine skin. J Comp Neurol 386 : 379–395 6. Paus R, Peters EMJ, Eichmüller S, Botchkarev VA (1997) Neural mechanisms of hair growth control. J Invest Dermatol Symp Proc 2 : 61–68
578 7. Kanno Y, Takeda K, Daikoku S (1991) Development of the hair in the rat: in vivo and in transplanted tissue. J Dermatol 18 : 262–270 8. Kashiwagi M, Kuroki T, Huh N (1997) Specific inhibition of hair follicle formation by epidermal growth factor in an organ culture of developing mouse skin. Dev Biol 189 : 22–32 9. Jones TE, Munger BL (1987) Neural modulation of cutaneous differentiation: epidermal hyperplasia and precocious hair development following partial neuralectomy in opossum pups. Neurosci Lett 79 : 6–10 10. Asada-Kubota M (1995) Inhibition of hair growth by subcutaneous injection of a sympathetic neurotoxin, 6-hypdroxydopamine in neonatal mice. Anat Embryol (Berl) 191 : 407–414 11. Paus R (1996) Control of the hair cycle and hair diseases as cycling disorders. Curr Opin Dermatol 3 : 248–258 12. Paus R, Stenn KS, Link RE (1989) The induction of anagen hair growth in telogen mouse skin by cyclosporin A administration. Lab Invest 60 : 365–369 13. Paus R, Maurer M, Slominski A, Czarnetzki BM (1994) Mast cell involvement in murine hair growth. Dev Biol 163 : 230– 240 14. Diamond J, Coughlin M, Macintyre L, Holmes M, Visheau B (1987) Evidence that endogenous beta nerve growth factor is responsible for the collateral sprouting, but not the regeneration, of nociceptive axons in adult rats. Proc Natl Acad Sci USA 84 : 6596–6600 15. Chase HB (1954) Growth of hair. Physiol Rev 34 : 113–126
16. Maurer M, Handjiski B, Paus R (1997) Hair growth modulation by topical immunophilin ligands. Induction of anagen, inhibition of massive catagen development, and relative protection from chemotherapy-induced alopecia. Am J Pathol 150 : 1433– 1441 17. Maurer M, Fischer E, Handjiski B, Stebut EV, Algermissen B, Bavandi A, Paus R (1997) Activated skin mast cells are involved in murine hair follicle regression (catagen). Lab Invest 77 : 319–332 18. Paus R, Foitzik K, Welker P, Bulfone-Paus S, Eichmueller S (1997) Transforming growth factor-beta type I and type II expression during murine hair follicle development and cycling. J Invest Dermatol 109 : 518–526 19. Slominski A, Paus R (1993) Melanogenesis is coupled to murine anagen: Toward new concepts for the role of melanocytes and the regulation of melanogenesis in hair growth. J Invest Dermatol 101 : 90S–97S 20. Paus R, Stenn KS, Link RE (1990) Telogen skin contains an inhibitor of hair growth. Br J Dermatol 122 : 777–784 21. Mearow KM, Kril Y, Diamond J (1993) Increased NGF mRNA expression in denervated skin. Neuroreport 4 : 351–354 22. Crowe R, Mitsou J, McGrouther DA, Burnstock G (1993) An increase in the growth of hair associated with hyperinnervation of the underlying vessels in rabbit skin. Neurosci Lett 161 : 105–108 23. Argyris TS (1967) Hair growth induction by damage. Adv Biol Skin 9 : 339–345
