Towards optimization of an organotypic assay system that imitates human hair follicle-like epithelial-mesenchymal interactions

BACKGROUND

Human hair growth can currently be studied in vitro by the use of organ-cultured scalp hair follicles (HFs). However, simplified organotypic systems are needed for dissecting the underlying epithelial-mesenchymal interactions and as screening tools for candidate hair growth-modulatory agents.

OBJECTIVES

To optimize the design and culture conditions of previously published organotypic systems that imitate epithelial-mesenchymal interactions in the human HF as closely as possible.

MATERIALS AND METHODS

Continuous submerged organotypic 'sandwich' cultures were established. These consist of a pseudodermis (collagen I mixed with and contracted by human interfollicular dermal fibroblasts) on which one of two upper layers is placed: either a mixture of Matrigel basement membrane matrix (BD Biosciences, Bedford, MA, U.S.A.) and follicular dermal papilla fibroblasts (DPC), with outer root sheath keratinocytes (ORSK) layered on the top ('layered' system), or a mixture of Matrigel, DPC and ORSK ('mixed' system). Morphological and functional characteristics of these 'folliculoid sandwiches' were then assessed by routine histology, histomorphometry and immunohistochemistry.

RESULTS

In both 'layered' and 'mixed' systems, the ORSK formed spheroid epithelial cell aggregates, which retained their characteristic keratin expression pattern (i.e. cytokeratin 6). In the 'mixed' sandwich model the size of the epithelial cell aggregates was smaller, but the numbers of ORSK were significantly higher than in the 'layered' model at day 14 in the culture. ORSK proliferated better in the 'mixed' than in the 'layered' sandwich system, regardless of the calcium or serum content of the media, whereas apoptosis of ORSK was lowest in the 'mixed' system in serum-free, low calcium medium. The kinetics of proliferation and apoptosis of DPC, which retained their characteristic expression of versican, were similar in both systems. However, proliferation and apoptosis of DPC were higher in the presence of serum and/or under high calcium conditions.

CONCLUSIONS

Our results underscore the importance of structural design and medium composition for epithelial-mesenchymal interactions as they occur in the human HF. Specifically, we report a new organotypic submerged 'folliculoid sandwich' system with serum-free, low calcium medium and a mixture of interacting human DPC and ORSK, which offers several advantages over previously available assays. This system allows the standardized assessment of the effects of a test agent on the proliferation, apoptosis and key marker expression of human ORSK and DPC under substantially simplified in vitro conditions which approximate the in vivo situation.

Increased airway responsiveness, allergy-type-I skin responses and systemic anaphylaxis in a humanized-severe combined immuno-deficiency mouse model

BACKGROUND

In patients with allergic bronchial asthma, a strong relationship between elevated serum IgE antibody titres and the development of increased airway responsiveness (AR) has been demonstrated. To further elucidate the relationship between human (hu) IgE and development of increased AR, we developed an in vivo model utilizing immuno-compromised severe combined immuno-deficiency (SCID) mice.

METHODS

SCID mice were either reconstituted with peripheral blood mononuclear cells (PBMC) from non-atopic, healthy or atopic individuals sensitized against house dust mite allergen (Der p), or passively sensitized with plasma from non-atopic, healthy or atopic individuals.

RESULTS

In both systems, atopic hu-SCID mice developed increased AR. The following results suggest that these responses were mediated via IgE antibodies: increased AR did not occur after transfer of either PBMC or IgE-negative plasma from non-atopic individuals; increased AR occurred simultaneous with increased serotonin release detected 15 min after allergen-aerosol challenge in bronchoalveolar lavage fluid; and increased AR required at least two allergen-aerosol challenges. SCID mice reconstituted with serum containing anti-Der p IgE antibodies developed positive immediate-type skin test responses to intradermal injection of Der p as well as anti-hu-IgE antibody. In addition, IgE binding to skin mast cells was demonstrated by immunohistochemistry. Furthermore, intravenous challenge of hu anti-Der p positive SCID mice with Der p resulted in systemic anaphylaxis.

CONCLUSION

These data provide evidence that passive immunization of SCID mice with hu IgE alters AR and that T cells and eosinophils were not a requirement for the development of increased AR in this model.

Enhanced inhibition of tumour growth and metastasis, and induction of antitumour immunity by IL-2-IgG2b fusion protein

Cytokine-immunoglobulin (Ig)-fusion proteins have attracted increasing interest as antitumour agents. Here, we have investigated the antimetastatic and antitumour responses elicited in vivo by mammary adenocarcinoma cells (TS/A) engineered to secrete interleukin (IL)-2-IgG fusion proteins. TS/A cells were transfected with DNA coding for IL-2-IgG2b, IgG2b or IL-2, and injected subcutaneously into syngeneic mice. Animals injected with TS/A-IL-2 or TS/A-IL-2-IgG2b both efficiently rejected tumours, whereas treatment with parental cells or TS/A-IgG2b was lethal. Interestingly, only mice vaccinated with IL-2-IgG2b fusion protein-secreting cells showed a long-lasting protective immunity against a later challenge with parental tumour cells. Moreover, the metastatic potential of TS/A-IL-2-IgG2b-transfected cells was dramatically decreased compared with TS/A-IL-2-cells, with a virtual absence of lung metastases after intravenous injection. Adenocarcinomas secreting IL-2-IgG2b exhibited a more prominent, early and persistent infiltration of CD4+, CD8+ and natural killer (NK) cells than TS/A-IL-2 cells. Therefore, upon transfection into adenocarcinoma cells, the IgG2b part of IL-2 fusion protein exerts intriguing added antitumour properties over IL-2 alone, thus contributing to a long-lasting tumour immunity, probably by the recruitment of specific immune effector cells. These findings suggest a promising new oncotherapeutic strategy for poorly immunogenic tumours: vaccination with tumour cells engineered to secrete IL-2-IgG2b fusion protein.

The IL-15R alpha chain signals through association with Syk in human B cells

The alpha-chain of the IL-15R (IL-15Ralpha) serves as the specific, high-affinity receptor for IL-15. It is expressed by lymphoid and nonlymphoid cells, including B cell lymphoma lines. In this study, we have further explored IL-15Ralpha-mediated signaling in activated primary B cells and in Raji cells, a human B-lymphoblastoid cell line which expresses the IL-15Ralpha and IL-2Rgamma chains, but lacks the IL-2Rbeta chain. Stimulation of Raji cells with IL-15 induces their proliferation and rescues them from C2-ceramide-induced apoptosis. By immunoprecipitation and Western blotting, we show that treatment of Raji cells and activated primary B cells with IL-15 induces coprecipitation of Syk kinase with the IL-15Ralpha chain. Upon association, the activated Syk kinase phosphorylates the IL-15Ralpha chain as well as phospholipase Cgamma, which coprecipitates with Syk. Furthermore, transfection of Raji cells with stem-loop Syk antisense oligonucleotides prevents IL-15Ralpha and phospholipase Cgamma phosphorylation as well as the inhibition of apoptosis by IL-15. Mutation of a defined region of the intracellular signaling portion of IL-15Ralpha (Tyr227) abrogates both the IL-15Ralpha/Syk association and IL-15Ralpha phosphorylation. Taken together, this suggests that Syk kinase physically and functionally associates with the IL-15Ralpha chain in B cells and that Syk plays a key role in mediating IL-15-induced signal transduction, thus accounting for the distinct functional consequences of IL-15 vs IL-2 binding to B cells.

Mutant laboratory mice with abnormalities in hair follicle morphogenesis, cycling, and/or structure: annotated tables

Numerous transgenic, targeted mutagenesis (so-called knockouts), conditional (so-called "gene switch") and spontaneous mutant mice develop abnormal hair phenotypes. The number of mice that exhibit such abnormalities is increasing exponentially as genetic engineering methods become routine. Since defined abnormalities in hair follicle morphogenesis, cycling and/or structure in such mutant mice provide important clues to the as yet poorly understood functional roles of many gene products, it is useful to summarize and classify these mutant mice according to their hair phenotype. This review provides a corresponding, annotated table of mutant mice with hair abnormalities, classifying the latter into 6 categories, 1) abnormally low number of hair follicles, 2) disorders of hair morphogenesis, 3) of hair follicle cycling, 4) of hair follicle structure 5) of sebaceous gland structure, and 6) hair growth disorders as a consequence of immunological abnormalities. This annotated table should serve as a useful source of reference for anyone who is interested in the molecular controls of hair growth, for investigators who are looking for mouse models to explore or compare the functional activities of their gene of interest, and for comparing the hair phenotype of newly generated mouse mutants with existing ones.

Indications for a 'brain-hair follicle axis (BHA)': inhibition of keratinocyte proliferation and up-regulation of keratinocyte apoptosis in telogen hair follicles by stress and substance P

It has long been suspected that stress can cause hair loss, although convincing evidence of this has been unavailable. Here, we show that in mice sonic stress significantly increased the number of hair follicles containing apoptotic cells and inhibited intrafollicular keratinocyte proliferation in situ. Sonic stress also significantly increased the number of activated perifollicular macrophage clusters and the number of degranulated mast cells, whereas it down-regulated the number of intraepithelial gd T lymphocytes. These stress-induced immune changes could be mimicked by injection of the neuropeptide substance P in nonstressed mice and were abrogated by a selective substance P receptor antagonist in stressed mice. We conclude that stress can indeed inhibit hair growth in vivo, probably via a substance P-dependent activation of macrophages and/or mast cells in the context of a brain-hair follicle axis.

Noggin is required for induction of the hair follicle growth phase in postnatal skin

During postnatal development, the hair follicle (HF) shows cyclic activity with periods of relative resting, active growth (anagen), and regression. We demonstrate that similar to the HF induction in embryonic skin, initiation of a new hair growth phase in postnatal skin requires neutralization of the inhibitory activity of bone morphogenetic protein 4 (BMP4) by the BMP antagonist noggin. In the resting HF, BMP4 mRNA predominates over noggin in the epithelium and mesenchyme, and the BMP receptor IA is prominently expressed in the follicular germ. Anagen development is accompanied by down-regulation of the BMP4 and increased noggin mRNA in the HF. Furthermore, administration of noggin protein induces new hair growth phase in postnatal telogen skin in vivo. In contrast, BMP4 induces selective arrest of anagen development in the non-tylotrich (secondary) HF. As a hair growth inducer, noggin increases Shh mRNA in the HF whereas BMP4 down-regulates Shh. This suggests that modulation of BMP4 signaling by noggin is essential for hair growth phase induction in postnatal skin and that the hair growth-inducing effect of noggin is mediated, at least in part, by Shh.

The nude mouse skin phenotype: the role of Foxn1 in hair follicle development and cycling

The original nude mouse mutation has proven to be an incredibly valuable biomedical tool since its discovery in 1966. Initially its value was as a tool to study the immune system. The immunodeficiency in this mutant mouse made nude mice valuable as hosts for xenografts, primarily for cancer research. More recently, the most obvious clinical feature of this mutant mouse, lack of hair, has been capitalized on to define the role of Foxn1 in normal and pathological skin and hair follicle physiology.

A new strategy for modulating chemotherapy-induced alopecia, using PTH/PTHrP receptor agonist and antagonist

Parathyroid hormone (PTH) related peptide (PTHrP) and the PTH/PTHrP receptor (PTH/PTHrP-R) show prominent cutaneous expression, where this signaling system may exert important paracrine and/or autocrine functions, such as in hair growth control. Chemotherapy-induced alopecia - one of the fundamental unsolved problems of clinical oncology - is driven in part by defined abnormalities in hair follicle cycling. We have therefore explored the therapeutic potential of a PTH/PTHrP-R agonist and two PTH/PTHrP-R antagonists in a mouse model of cyclophosphamide-induced alopecia. Intraperitoneal administration of the agonist PTH(1-34) or the antagonists PTH(7-34) and PTHrP(7-34) significantly altered the follicular response to cyclophosphamide in vivo. PTH(7-34) and PTHrP(7-34) shifted it towards a mild form of "dystrophic anagen", associated with a significant reduction in apoptotic (TUNEL+) hair bulb cells, thus mitigating the degree of follicle damage and retarding the onset of cyclophosphamide-induced alopecia. PTH(1-34), in contrast, forced hair follicles into "dystrophic catagen", associated with enhanced intrafollicular apoptosis. We had previously shown that an induced shift in the follicular damage-response towards "dystrophic catagen" mitigates cyclophosphamide-induced alopecia, whereas a shift towards "dystrophic catagen" initially enhanced the hair loss, yet subsequently promoted accelerated hair follicle recovery. Therefore, this study in an established animal model of chemotherapy-induced alopecia, which closely mimics human chemotherapy-induced alopecia, strongly encourages the exploration of PTH/PTHrP-R agonists and antagonists as novel therapeutic agents in chemotherapy-induced alopecia.

A comprehensive guide for the accurate classification of murine hair follicles in distinct hair cycle stages

Numerous strains of mice with defined mutations display pronounced abnormalities of hair follicle cycling, even in the absence of overt alterations of the skin and hair phenotype; however, in order to recognize even subtle, hair cycle-related abnormalities, it is critically important to be able to determine accurately and classify the major stages of the normal murine hair cycle. In this comprehensive guide, we present pragmatic basic and auxiliary criteria for recognizing key stages of hair follicle growth (anagen), regression (catagen) and quiescence (telogen) in C57BL/6NCrlBR mice, which are largely based on previous work from other authors. For each stage, a schematic drawing and representative micrographs are provided in order to illustrate these criteria. The basic criteria can be employed for all mouse strains and require only routine histochemical techniques. The auxiliary criteria depend on the immunohistochemical analysis of three markers (interleukin-1 receptor type I, transforming growth factor-beta receptor type II, and neural cell-adhesion molecule), which allow a refined analysis of anatomical hair follicle compartments during all hair cycle stages. In contrast to prior staging systems, we suggest dividing anagen III into three distinct substages, based on morphologic differences, onset and progression of melanogenesis, and the position of the dermal papilla in the subcutis. The computer-generated schematic representations of each stage are presented with the aim of standardizing reports on follicular gene and protein expression patterns. This guide should become a useful tool when screening new mouse mutants or mice treated with pharmaceuticals for discrete morphologic abnormalities of hair follicle cycling in a highly reproducible, easily applicable, and quantifiable manner.

Patterns of proliferation and apoptosis during murine hair follicle morphogenesis

In this study, we have correlated cutaneous apoptosis and proliferation in neonatal mice during hair follicle morphogenesis. We have applied a novel triple- staining technique that uses Ki67 immunoreactivity as a marker of proliferation as well as TUNEL and Hoechst 33342 staining as apoptosis markers. We have also assessed the immunoreactivity of interleukin-1 beta-converting enzyme, caspase 1, a key enzyme in the execution of apoptosis, and of P-cadherin, which has been suggested as a key adhesion receptor in segregating proliferating keratinocytes. The TUNEL data were systematically compared with high resolution light microscopy and transmission electron microscopy data. Virtually all keratinocytes of the developing hair bud were strongly Ki67(+), suggesting that the hair bud is not an epidermal invagination but primarily the product of localized keratinocyte proliferation. As hair follicle development advanced, three distinct foci of proliferation became apparent: the distal outer root sheath around the hair canal, the mid outer root sheath, and the proximal hair matrix. Of these proliferating hair follicle keratinocytes only defined subsets expressed P-cadherin. TUNEL(+) cells in the hair follicle were not found before stage 5 of murine hair follicle morphogenesis. During the early stages of hair follicle development, interleukin-1 beta-converting enzyme immunoreactivity was present on all keratinocytes, but virtually disappeared from the proximal hair follicle epithelium later on. High resolution light microscopy/transmission electron microscopy revealed scattered and clustered apoptotic keratinocytes in all epithelial hair follicle compartments throughout hair follicle development, including its earliest stages. This highlights striking differences in the demarcation of apoptotic hair follicle keratinocytes between the TUNEL technique and high resolution light microscopy/transmission electron microscopy and suggests a role for apoptosis in sculpting the hair follicle even during early hair follicle development.

Hair-cycle-associated remodeling of the peptidergic innervation of murine skin, and hair growth modulation by neuropeptides

As the neuropeptide substance P can manipulate murine hair growth in vivo, we here further studied the role of sensory neuropeptides in hair follicle biology by determining the distribution and hair-cycle-dependent remodeling of the sensory innervation in C57BL/6 mouse back skin. Calcitonin-gene-related peptide, substance P, and peptide histidine methionine (employed as vasoactive intestinal peptide marker) were identified by immunohistochemistry. All of these markers immunolocalized to bundles of nerve fibers and to single nerve fibers, with distinct distribution patterns and major hair-cycle-associated changes. In the epidermis and around the distal hair follicle and the arrector pili muscle, only calcitonin-gene-related peptide immunoreactive nerve fibers were visualized, whereas substance P and peptide histidine methionine immunoreactive nerve fibers were largely restricted to the dermis and subcutis. Compared to telogen skin, the number of calcitonin-gene-related peptide, substance P, and peptide histidine methionine immunoreactive single nerve fibers increased significantly (p < 0.01) during anagen, including around the bulge region (the seat of epithelial stem cells). Substance P significantly accelerated anagen progression in murine skin organ culture, whereas calcitonin-gene-related peptide and a substance-P-inhibitory peptide inhibited anagen (p < 0.05). The inhibitory effect of calcitonin-gene-related peptide could be antagonized by coadministrating substance P. In contrast to substance P, calcitonin-gene-related peptide failed to induce anagen when released from subcutaneous implants. This might reflect a differential functional assignment of the neuropeptides calcitonin-gene-related peptide and substance P in hair growth control, and invites the use of neuropeptide receptor agonists and antagonists as novel pharmacologic tools for therapeutic hair growth manipulation.

Controls of hair follicle cycling

Nearly 50 years ago, Chase published a review of hair cycling in which he detailed hair growth in the mouse and integrated hair biology with the biology of his day. In this review we have used Chase as our model and tried to put the adult hair follicle growth cycle in perspective. We have tried to sketch the adult hair follicle cycle, as we know it today and what needs to be known. Above all, we hope that this work will serve as an introduction to basic biologists who are looking for a defined biological system that illustrates many of the challenges of modern biology: cell differentiation, epithelial-mesenchymal interactions, stem cell biology, pattern formation, apoptosis, cell and organ growth cycles, and pigmentation. The most important theme in studying the cycling hair follicle is that the follicle is a regenerating system. By traversing the phases of the cycle (growth, regression, resting, shedding, then growth again), the follicle demonstrates the unusual ability to completely regenerate itself. The basis for this regeneration rests in the unique follicular epithelial and mesenchymal components and their interactions. Recently, some of the molecular signals making up these interactions have been defined. They involve gene families also found in other regenerating systems such as fibroblast growth factor, transforming growth factor-beta, Wnt pathway, Sonic hedgehog, neurotrophins, and homeobox. For the immediate future, our challenge is to define the molecular basis for hair follicle growth control, to regenerate a mature hair follicle in vitro from defined populations, and to offer real solutions to our patients' problems.

Graying: gerontobiology of the hair follicle pigmentary unit

The visual appearance of humans derives predominantly from their skin and hair color. The phylogenetically ancient biochemical [corrected] pathway underling this phenomenon is called melanogenesis and results in the production of melanin pigments in neural crest-derived melanocytes, followed by its transfer to epithelial cells. While melanin from epidermal melanocytes clearly protects human skin by screening harmful ultraviolet radiation, the biologic value of hair pigmentation is less clear. In addition to important roles in social/sexual communication, one potential benefit of pigmented scalp hair in humans may be the rapid excretion of heavy metals, chemicals, toxins from the body by their selective binding to melanin. The hair follicle and epidermal melanogenic systems are broadly distinct, though open. The primary distinguishing feature of follicular melanogenesis, compared to the continuous melanogenesis in the epidermis, is the tight coupling of hair follicle melanogenesis to the hair growth cycle. This cycle appears to involve periods of melanocyte proliferation (during early anagen), maturation (mid to late anagen) and melanocyte death via apoptosis (during early catagen). Thus, each hair cycle is associated with the reconstruction of an intact hair follicle pigmentary unit... at least for the first 10 cycles or so. Thereafter, gray and white hairs appear, suggesting an age-related, genetically regulated exhaustion of the pigmentary potential of each individual hair follicle. Melanocyte aging may be associated with reactive oxygen species-mediated damage to nuclear and mitochondrial DNA with resultant accumulation of mutations with age, in addition to dysregulation of anti-oxidant mechanisms or pro/anti-apoptotic factors within the cells. While the perception of "gray hair" derives in large part from the admixture of pigmented and white hair, it is important to note that individual hair follicles can indeed exhibit pigment dilution or true grayness. This dilution is due to a reduction in tyrosinase activity of hair bulbar melanocytes, sub-optimal melanocyte-cortical keratinocyte interactions, and defective migration of melanocytes from a reservoir in the upper outer root sheath to the pigment-permitting microenvironment close to the dermal papilla of the hair bulb. Animal models with mutations in apoptotic survival factors (e.g. bcl-2) and in melanogenic enzymes (TRP-1) are providing valuable insights into the aging hair pigmentary unit. It is from these and other advances, including our ability to grow hair follicle melanocytes in vitro, that the possibility of reversing canities has been raised. Indeed, it is not too uncommon to see spontaneous repigmentation along the same individual hair shaft in early canities. Moreover, melanocytes taken from gray and white hair follicles can be induced to pigment in vitro. One of the surprising results of pigment loss in canities is the alteration in keratinocyte proliferation and differentiation, providing the tantalizing suggestion that melanocytes in the hair follicle contribute far more that packages of melanin alone. Furthermore, there have been some unconfirmed reports in the literature suggesting that canities may link (although not causally) with more systemic alterations in homeostasis e.g. osteoporosis. Here, we review the current state of knowledge of the development, regulation and control of the human hair follicle pigmentary unit during life.

High-dose proinflammatory cytokines induce apoptosis of hair bulb keratinocytes in vivo

BACKGROUND

Hair loss following skin inflammation may in part be mediated by keratinocyte (KC) apoptosis. While the effects of different cytokines or other apoptosis stimulating agents such as interferon (IFN)-gamma or tumour necrosis factor (TNF)-alpha on KC apoptosis in vitro have been addressed in several studies, little is known about the effects of proinflammatory cytokines on KC apoptosis in vivo.

OBJECTIVES

To study the effects of intradermally injected TNF-alpha, interleukin (IL)-1beta and IFN-gamma on KC apoptosis in the back skin of C57BL/6 mice.

METHODS

Apoptosis in epidermal and hair bulb KCs was analysed by immunohistology using TUNEL staining.

RESULTS

Injection of TNF-alpha induced a significantly higher number of apoptotic cells within the epidermis than vehicle; all three proinflammatory cytokines together further increased their number. Intrafollicular hair bulb KCs were much more susceptible to apoptosis induction by TNF-alpha or IL-1beta; their injection significantly upregulated apoptosis after 6 h, which was further increased after 24 h. The combination of all cytokines together accelerated intrafollicular apoptosis after 6 h by doubling the number of apoptotic cells per hair bulb, compared with the effects of TNF-alpha or IL-1beta alone.

CONCLUSIONS

These data suggest that programmed cell death of proliferating KCs in vivo can be induced by proinflammatory cytokines.

A role for p75 neurotrophin receptor in the control of apoptosis-driven hair follicle regression

To examine the mechanisms that underlie the neurotrophin-induced, apoptosis-driven hair follicle involution (catagen), the expression and function of p75 neurotrophin receptor (p75NTR), which is implicated in apoptosis control, were studied during spontaneous catagen development in murine skin. By RT-PCR, high steady-state p75NTR mRNA skin levels were found during the anagen-catagen transition of the hair follicle. By immunohistochemistry, p75NTR alone was strongly expressed in TUNEL+/Bcl2- keratinocytes of the regressing outer root sheath, but both p75NTR and TrkB and/or TrkC were expressed by the nonregressing TUNEL-/Bcl2+ secondary hair germ keratinocytes. To determine whether p75NTR is functionally involved in catagen control, spontaneous catagen development was compared in vivo between p75NTR knockout (-/-) and wild-type mice. There was significant catagen retardation in p75NTR knockout mice as compared to wild-type controls (P<0.05). Instead, transgenic mice-overexpressing NGF (promoter: K14) showed substantial acceleration of catagen (P<0.001). Although NGF, brain-derived neurotrophic factor (BDNF), and neurotrophin 3 (NT-3) accelerated catagen in the organ-cultured skin of C57BL/6 mice, these neurotrophins failed to promote catagen development in the organ-cultured p75NTR null skin. These findings suggest that p75NTR signaling is involved in the control of kerotinocyte apoptosis during catagen and that pharmacological manipulation of p75NTR signaling may prove useful for the treatment of hair disorders that display premature entry into catagen.

Patterns of hairless (hr) gene expression in mouse hair follicle morphogenesis and cycling

The hr (hairless) gene encodes a putative transcription factor with restricted expression in the skin and brain. Mutations in the hr locus cause papular atrichia in humans and complete hair loss in mice and other mammals. To further elucidate the role of hr in skin biology, and to identify potential target cells for hr regulation, we studied hr mRNA localization during hair follicle (HF) morphogenesis and cycling in normal C57BL/6J mice. In situ hybridization revealed that hr expression was present in the suprabasal cell layers of the epidermis, whereas the basal and highly differentiated keratinocytes of the granular layer were hr-negative. During the early stages of HF morphogenesis, hr mRNA was detected in the developing hair peg. Later, it became concentrated in the HF infundibulum, in the HF matrix, and in the inner root sheath (IRS), whereas the dermal papilla (DP) and outer root sheath were consistently hr mRNA-negative. During catagen, hr gene expression gradually declined in the regressing IRS, shortly but dramatically increased in the zone of developing club hair, and became up-regulated in the epithelial cells adjacent to the DP. The co-localization of hr mRNA with the site of the morphological defects in mutant skin implicates hr as a key factor in regulating basic cellular processes during catagen, including club hair formation, maintenance of DP-epithelial integrity, IRS disintegration, and keratinocyte apoptosis in the HF matrix.

Cathepsin L deficiency as molecular defect of furless: hyperproliferation of keratinocytes and pertubation of hair follicle cycling

Lysosomal cysteine proteinases of the papain family are involved in lysosomal bulk proteolysis, major histocompatibility complex class II mediated antigen presentation, prohormone processing, and extracellular matrix remodeling. Cathepsin L (CTSL) is a ubiquitously expressed major representative of the papain-like family of cysteine proteinases. To investigate CTSL in vivo functions, the gene was inactivated by gene targeting in embryonic stem cells. CTSL-deficient mice develop periodic hair loss and epidermal hyperplasia, acanthosis, and hyperkeratosis. The hair loss is due to alterations of hair follicle morphogenesis and cycling, dilatation of hair follicle canals, and disturbed club hair formation. Hyperproliferation of hair follicle epithelial cells and basal epidermal keratinocytes-both of ectodermal origin-are the primary characteristics underlying the mutant phenotype. Pathological inflammatory responses have been excluded as a putative cause of the skin and hair disorder. The phenotype of CTSL-deficient mice is reminiscent of the spontaneous mouse mutant furless (fs). Analyses of the ctsl gene of fs mice revealed a G149R mutation inactivating the proteinase activity. CTSL is the first lysosomal proteinase shown to be essential for epidermal homeostasis and regular hair follicle morphogenesis and cycling.

Inhibition of keratinocyte apoptosis by IL-15: a new parameter in the pathogenesis of psoriasis?

Keratinocytes (KC) are important source of and targets for several cytokines. Although KC express IL-15 mRNA, the functional effects of IL-15 on these epithelial cells remain to be dissected. Investigating primary human foreskin KC and HaCaT cells, we show here by semiquantitative RT-PCR and flow cytometric analysis that both translate IL-15 and IL-15R mRNA and express IL-15 and IL-15Ralpha protein on the cell surface, suggesting that human KC can employ IL-15 for juxtacrine signaling. While IL-15 exerted no significant effect on KC proliferation and IL-6 or IL-8 secretion, IL-15 inhibited both anti-Fas and methylcellulose-induced KC apoptosis in vitro. This is in line with the recognized potent anti-apoptotic effects of IL-15. IL-2, whose receptor shares two components with the IL-15R, failed to inhibit KC apoptosis. Together with the role of IL-15 in sustaining chronic immune reactions, this invited the question of whether a reduction of KC apoptosis by IL-15 may be involved in the pathogenesis of psoriasis, a chronic hyperproliferative inflammatory skin disease characterized by abnormally low KC apoptosis in the epidermis. Remarkably, compared with nonlesional psoriatic skin and skin of healthy volunteers, lesional psoriatic epidermis showed high IL-15 protein expression in the epidermis and enhanced binding activity for IL-15. Therefore, antagonizing the inhibitory effects of IL-15 on KC apoptosis deserves exploration as a novel therapeutic strategy in psoriasis management.