The cell biology of human hair follicle pigmentation

Although we have made significant progress in understanding the regulation of the UVR‐exposed epidermal‐melanin unit, we know relatively little about how human hair follicle pigmentation is regulated. Progress has been hampered by gaps in our knowledge of the hair growth cycle’s controls, to which hair pigmentation appears tightly coupled. However, pigment cell researchers may have overly focused on the follicular melanocytes of the nocturnal and UVR‐shy mouse as a proxy for human epidermal melanocytes. Here, I emphasize the epidermis‐follicular melanocyte pluralism of human skin, as research models for vitiligo, alopecia areata and melanoma, personal care/cosmetics innovation. Further motivation could be in finding answers to why hair follicle and epidermal pigmentary units remain broadly distinct? Why melanomas tend to originate from epidermal rather than follicular melanocytes? Why multiple follicular melanocyte sub‐populations exist? Why follicular melanocytes are more sensitive to aging influences? In this perspective, I attempt to raise the status of the human hair follicle melanocyte and highlight some species‐specific issues involved which the general reader of the pigmentation literature (with its substantial mouse‐based data) may not fully appreciate.     

Polyclonal origin and hair induction ability of dermal papillae in neonatal and adult mouse back skin

Hair follicle development and growth are regulated by Wnt signalling and depend on interactions between epidermal cells and a population of fibroblasts at the base of the follicle, known as the dermal papilla (DP). DP cells have a distinct gene expression signature from non-DP dermal fibroblasts. However, their origins are largely unknown. By generating chimeric mice and performing skin reconstitution assays we show that, irrespective of whether DP form during development, are induced by epidermal Wnt activation in adult skin or assemble from disaggregated cells, they are polyclonal in origin. While fibroblast proliferation is necessary for hair follicle formation in skin reconstitution assays, mitotically inhibited cells readily contribute to DP. Although new hair follicles do not usually develop in adult skin, adult dermal fibroblasts are competent to contribute to DP during hair follicle neogenesis, irrespective of whether they originate from skin in the resting or growth phase of the hair cycle or skin with β-catenin-induced ectopic follicles. We propose that during skin reconstitution fibroblasts may be induced to become DP cells by interactions with hair follicle epidermal cells, rather than being derived from a distinct subpopulation of cells.     

RADIATION DEPIGMENTATION OF MOUSE HAIR: A STUDY OF FOLLICULAR MELANOCYTE POPULATIONS

The radiation depigmentation of mouse hair has been studied by a technique enabling melanocyte per follicle counts to be made. Distributions for normal skin show a large peak corresponding to the zigzag hair type. Changes in the frequency distributions of melanocytes per follicle after irradiation are presented for Strong F and DBA-1 mice irradiated in anagen or telogen stages of hair growth. These distributions clearly suggest the existence of some precursor cells, and the dose-response curves obtained by defining radiation survivors as follicles containing more than ten melanocytes gives the sensitivity of these cells to inactivation. D₀ values are 180–220 rads. A melanocyte-melanoblast model is proposed for the follicular melanocyte cycle which can be outlined as follows: The telogen follicle contains a small number of amelanotic melanocytes that survived through catagen. These cells possess the ability to repopulate the follicle with melanocytes. In catagen functional and/or amelanotic melanocytes are lost at random. Genes for dilution (possibly only when modified by other coat colour genes) and radiation both increase the chance of melanocyte loss at catagen by altering the melanocyte-dermal papilla relationship. One way in which this is affected is by a shortening of the dendrites. A feedback may operate in the follicle so that the full complement of melanocytes is achieved whatever number of melanocytes persists in telogen.     

Migration of melanoblasts into the developing murine hair follicle is accompanied by transient c-Kit expression.

Disruption of the c-Kit/stem cell factor (SCF) signaling pathway interferes with the survival, migration, and differentiation of melanocytes during generation of the hair follicle pigmentary unit. We examined c-Kit, SCF, and S100 (a marker for precursor melanocytic cells) expression, as well as melanoblast/melanocyte ultrastructure, in perinatal C57BL/6 mouse skin. Before the onset of hair bulb melanogenesis (i.e., stages 0-4 of hair follicle morphogenesis), strong c-Kit immunoreactivity (IR) was seen in selected non-melanogenic cells in the developing hair placode and hair plug. Many of these cells were S100-IR and were ultrastructurally identified as melanoblasts with migratory appearance. During the subsequent stages (5 and 6), increasingly dendritic c-Kit-IR cells successively invaded the hair bulb, while S100-IR gradually disappeared from these cells. Towards the completion of hair follicle morphogenesis (stages 7 and 8), several distinct follicular melanocytic cell populations could be defined and consisted broadly of (a) undifferentiated, non-pigmented c-Kit-negative melanoblasts in the outer root sheath and bulge and (b) highly differentiated melanocytes adjacent to the hair follicle dermal papilla above Auber's line. Widespread epithelial SCF-IR was seen throughout hair follicle morphogenesis. These findings suggest that melanoblasts express c-Kit as a prerequisite for migration into the SCF-supplying hair follicle epithelium. In addition, differentiated c-Kit-IR melanocytes target the bulb, while non-c-Kit-IR melanoblasts invade the outer root sheath and bulge in fully developed hair follicles.     

Maintenance of distinct melanocyte populations in the interfollicular epidermis

Hair follicles and sweat glands are recognized as reservoirs of melanocyte stem cells (MSCs). Unlike differentiated melanocytes, undifferentiated MSCs do not produce melanin. They serve as a source of differentiated melanocytes for the hair follicle and contribute to the interfollicular epidermis upon wounding, exposure to ultraviolet irradiation or in remission from vitiligo, where repigmentation often spreads outwards from the hair follicles. It is unknown whether these observations reflect the normal homoeostatic mechanism of melanocyte renewal or whether unperturbed interfollicular epidermis can maintain a melanocyte population that is independent of the skin's appendages. Here, we show that mouse tail skin lacking appendages does maintain a stable melanocyte number, including a low frequency of amelanotic melanocytes, into adult life. Furthermore, we show that actively cycling differentiated melanocytes are present in postnatal skin, indicating that amelanotic melanocytes are not uniquely relied on for melanocyte homoeostasis.     

Different Populations of Melanocytes Are Present in Hair Follicles and Epidermis

Melanocytes in human skin reside both in the epidermis and in the matrix and outer root sheath of anagen hair follicles. Comparative study of melanocytes in these different locations has been difficult as hair follicle melanocytes could not be cultured. In this study we used a recently described method of growing hair follicle melanocytes to characterize and compare hair follicle and epidermal melanocytes in the scalp of the same individual. Three morphologically and antigenically distinct types of melanocytes were observed in primary culture. These included (1) moderately pigmented and polydendritic melanocytes derived from epidermis; (2) small, bipolar, amelanotic melanocytes; and (3) large, intensely pigmented melanocytes; the latter two were derived from hair follicles. The three sub-populations of cells all reacted with melanocyte-specific monoclonal antibody. Epidermal and amelanotic hair follicle melanocytes proliferated well in culture, whereas the intensely pigmented hair follicle melanocytes did not. Amelanotic hair follicle melanocytes differed from epidermal melanocytes in being less differentiated, and they expressed less mature melanosome antigens. In addition, hair follicle melanocytes expressed some antigens associated with alopecia areata, but not antigens associated with vitiligo, whereas the reverse was true for epidermal melanocytes. Thus, antigenically different populations of melanocytes are present in epidermis and hair follicle. This could account for the preferential destruction of hair follicle melanocytes in alopecia areata and of epidermal melanocytes in vitiligo.     

Characterization of melanocyte-specific inducible Cre recombinase transgenic mice

Conditional Cre-mediated recombination has emerged as a robust method of introducing somatic genetic alterations in an organ-specific manner in the mouse. Here, we generated and characterized mice harboring a 4-hydroxytamoxifen (OHT)-inducible Cre recombinase-estrogen receptor fusion transgene under the control of the melanocyte-specific tyrosinase promoter, designated Tyr::CreER(T2). Cre-mediated recombination was induced in melanocytes in a spatially and temporally controlled manner upon administration of OHT and was documented in embryonic melanoblasts, follicular bulb melanocytes, dermal dendritic melanocytes, epidermal melanocytes of tail skin, and in putative melanocyte stem cells located within the follicular bulge. Functional evidence suggestive of recombination in follicular melanocyte stem cells included the presence of Cre-mediated recombination in follicular bulb melanocytes 1 year after topical OHT administration, by which time several hair cycles have elapsed and the melanocytes residing in this location have undergone multiple rounds of apoptosis and replenishment. These Tyr:: CreER(T2) transgenic mice represent a useful resource for the evaluation of melanocyte developmental genetics, the characterization of melanocyte stem cell function and dynamics, and the construction of refined mouse models of malignant melanoma.     

Melanocyte subpopulation turnover during the human hair cycle: an immunohistochemical study

The human hair cycle is characterized by successive phases of growth and involution that imply tissue regression and regeneration. As a consequence, the hair melanin unit has to be renewed in a cyclic manner. Actually, the behavior of human hair follicle melanocytes throughout the hair cycle has been poorly studied. Thus, the origin of melanocytes present in the bulb after human hair regeneration is still not clarified, and neither are the events that control the melanin biosynthesis activity in the human hair bulb. In this study, we showed at the cellular level that in human pigmented hair follicles, the expression of tyrosinase and tyrosinase-related protein-1 (TRP-1) was detectable during the anagen phases III/IV through VI, only in those melanocytes which were located in the bulb. During the catagen phase, the two evaluated melanogenic enzymes were detectable no more, although melanocytes were still present in the preceding bulbar area. The epithelial column of catagen follicles and the capsule of telogen follicles also contained inactive melanocytes as evidenced by pMel-17 labeling. At the induction of a new anagen hair follicle, some melanocytes were committed to cell division, but only when located in the nascent bulb close to the dermal papilla. Our results emphasize the close relationship between melanogenesis and the hair cycle and suggest that in humans, melanogenesis is restricted to anagen hair follicles not because of the regulation of tyrosinase activity, but because of melanogenic enzyme expression, e.g., tyrosinase and TRP-1. Furthermore, the fact that in the newly developing anagen hair follicles, cell-division commitment and tyrosinase and TRP-1 expression were observed in melanocytes only when located in the nascent bulb suggests a highly regio-specific melanocyte stimulation in early the anagen phase.     

Hair cycle dynamics: the case of the human hair follicle

The existence of a growth and regeneration cycle makes the hair follicle a true paradigm of tissue homeostasis. Analysis of about 9000 cycles led us to propose a stochastic model of human hair dynamics. The existence of hair cycles implies that stem cells must be cyclically activated and hair melanin unit has to be renewed. Using different markers, we were able to identify two distinct epithelial stem cell reservoirs, located in the upper and lower thirds of the anagen hair follicle outer root sheath. These two reservoirs fuse during the regression phase and individualize again in the new forming anagen hair follicle. Using a set of antibodies specific of melanocyte lineage and melanogenesis, pigmentation unit turnover was followed throughout the entire hair cycle. In the terminal anagen hair, active melanocytes were localized on top of the dermal papilla, while amelanotic melanocytes were identified in the upper third of the outer root sheath (ORS). Those amelanotic melanocytes located in upper ORS probably represented a melanocyte reservoir for successive hair generation, since at the induction of anagen phase, some melanocytes were committed to cell division and melanogenesis was turned on, but only in the nascent hair bulb, close to the dermal papilla.     

Bi-compartmental communication contributes to the opposite proliferative behavior of Notch1-deficient hair follicle and epidermal keratinocytes

Notch1-deficient epidermal keratinocytes become progressively hyperplastic and eventually produce tumors. By contrast, Notch1-deficient hair matrix keratinocytes have lower mitotic rates, resulting in smaller follicles with fewer cells. In addition, the ratio of melanocytes to keratinocytes is greatly reduced in hair follicles. Investigation into the underlying mechanism for these phenotypes revealed significant changes in the Kit, Tgfbeta and insulin-like growth factor (IGF) signaling pathways, which have not been previously shown to be downstream of Notch signaling. The level of Kitl (Scf) mRNA produced by Notch1-deficient follicular keratinocytes was reduced when compared with wild type, resulting in a decline in melanocyte population. Tgfbeta ligands were elevated in Notch1-deficient keratinocytes, which correlated with elevated expression of several targets, including the diffusible IGF antagonist Igfbp3 in the dermal papilla. Diffusible stromal targets remained elevated in the absence of epithelial Tgfbeta receptors, consistent with paracrine Tgfbeta signaling. Overexpression of Igf1 in the keratinocyte reversed the phenotype, as expected if Notch1 loss altered the IGF/insulin-like growth factor binding protein (IGFBP) balance. Conversely, epidermal keratinocytes contained less stromal Igfbp4 and might thus be primed to experience an increase in IGF signaling as animals age. These results suggest that Notch1 participates in a bi-compartmental signaling network that controls homeostasis, follicular proliferation rates and melanocyte population within the skin.     

THE FINE STRUCTURE OF NORMAL AND NEOPLASTIC MELANOCYTES IN THE SYRIAN HAMSTER, WITH PARTICULAR REFERENCE TO CARCINOGEN-INDUCED MELANOTIC TUMORS

The dermal melanocyte system of the Syrian hamster is particularly responsive to the melanogenetic and tumor-inducing effects of 7,12-dimethylbenz(a)anthracene (DMBA). The melanocytes of the hair follicles appear to be susceptible to the melanogenetic effect of DMBA but not to its tumor-inducing effect. The epidermal melanocytes are non-pigmented and are unresponsive to both melanogenetic and carcinogenic effects of DMBA. The pigmented granules of the dermal melanocytes of both the golden and the white hamster have an identical substructure and pattern of melanization which occurs in an orderly fashion on a delicate fibrillar component. The hair melanocytes have larger pigment granules with a more complicated fibrillar substructure. The epidermal melanocytes do not possess pigment granules but are recognized by their dendritic shape, the absence of desmosomes and tonofilaments, and the presence of racket-shaped or rod-shaped organelles. The melanin granules in neoplastic melanocytes of the golden hamster differ from corresponding normal melanocytes only in their larger size. In the white hamster, however, the melanin granules in tumors produced under identical experimental conditions are so bizarre and atypical that consideration was given to the possibility that a genetic difference in the melanization pattern between the two varieties becomes apparent in carcinogen-induced melanotic tumors. No definite conclusions could be reached as to the precise origin of the melanin granules in either normal or neoplastic melanocytes.     

Molecular biology of hair morphogenesis: development and cycling

In mammals, hair follicles produce hairs that fulfill a number of functions including thermoregulation, collecting sensory information, protection against environmental trauma, social communication, and mimicry. Hair follicles develop as a result of epithelial-mesenchymal interactions between epidermal keratinocytes committed to hair-specific differentiation and cluster of dermal fibroblasts that form follicular papilla. During postnatal life, hair follicles show patterns of cyclic activity with periods of active growth and hair production (anagen), apoptosis-driven involution (catagen), and relative resting (telogen). During last decade, substantial progress has been achieved in delineating molecular mechanisms that control hair follicle development and cyclic activity. In this review, we summarize the data demonstrating that regulation of hair follicle development in the embryo and control of hair follicle growth during postnatal life are highly conserved and both require involvement of similar molecular mechanisms. Since many of the molecules that control hair follicle development and cycling are also involved in regulating morphogenesis and postnatal biology of other ectodermal derivatives, such as teeth, feathers, and mammary glands, basic principles and molecular mechanisms that govern hair follicle development and growth may also be applicable for other developmental systems.     

The patchwork mouse phenotype: implication for melanocyte replacement in the hair follicle.

Mice homozygous for the recessive patchwork (pwk) mutation are characterized by a variegated pigment pattern with a mixture of unpigmented and normally pigmented hairs. The pigmented hair bulbs contain functional melanocytes. By contrast, the unpigmented hair bulbs contain no melanocytes. This lack results from the death of melanoblasts in the hair follicle at the end of embryogenesis. Here, we report that melanoblasts and melanocytes are found in the epidermis of pwk/pwk mice. Furthermore, these epidermal pigment cells are able to colonize new hair follicles after skin wounding. Despite the presence of epidermal pigment cells with a colonization potential, a follicle that had produced an unpigmented hair produces a new unpigmented hair during the successive hair growth cycles. This hair color continuity is also true for the pigmented hair follicles. Thus, in normal conditions, the hair acts as an independent functional unit as regards its pigment cells population.     

The Patchwork Mouse Phenotype: Implication for Melanocyte Replacement in the Hair Follicle

Mice homozygous for the recessive patchwork (pwk) mutation are characterized by a variegated pigment pattern with a mixture of unpigmented and normally pigmented hairs. The pigmented hair bulbs contain functional melanocytes. By contrast, the unpigmented hair bulbs contain no melanocytes. This lack results from the death of melanoblasts in the hair follicle at the end of embryogenesis. Here, we report that melanoblasts and melanocytes are found in the epidermis of pwk/pwk mice. Furthermore, these epidermal pigment cells are able to colonize new hair follicles after skin wounding. Despite the presence of epidermal pigment cells with a colonization potential, a follicle that had produced an unpigmented hair produces a new unpigmented hair during the successive hair growth cycles. This hair color continuity is also true for the pigmented hair follicles. Thus, in normal conditions, the hair acts as an independent functional unit as regards its pigment cells population.     

Mesoderm-Ectoderm Interaction in the Production of the Agouti Pigmentation Pattern in Mice

Alleles at the agouti locus in the mouse determine the synthesis of either phaeomelanin or eumelanin by follicular melanocytes by altering the hair follicle environment. The method of dermal-epidermal recombination of mouse skin from C57BL/6J a/a and C57BL/6J A(w-J)/A(w-J) embryos was used in this study to establish the precise site of agouti gene action within the hair follicle. The pigmentary pattern of hairs formed in the recombination skin grafts was specific for the genotype of the dermal (mesodermal) component of the hair follicle. The genotype of the epidermal (ectodermal) component had no influence on the type of hair pigmentary pattern. These results indicate that future studies on gene mechanisms should focus on the dermis as the determining factor in altering the hair follicle environment.     

The M4 muscarinic acetylcholine receptor plays a key role in the control of murine hair follicle cycling and pigmentation

Cholinergic receptors of the muscarinic class (M1-M5) are expressed in epidermal keratinocytes and melanocytes as well as in the hair follicle. Knockout (KO) mice of all five receptors have been created and resulted in different phenotypes. KO mice with a deletion of the M4 muscarinic acetylcholine receptor (M4R) present a striking hair phenotype, which we have analyzed here in greater detail by quantitative histomorphometry. Earlier studies revealed a retarded hair follicle morphogenesis in M4R KO mice, compared to age-matched wild type controls. On day 17, when mice enter the first hair growth cycle, the KO mice still showed a slightly retarded catagen phase. Subsequently, hair follicles of the KO mice stayed in a highly significantly prolonged telogen phase, while wild type mice had already far progressed in the hair cycle by entry into anagen. Most strikingly, the M4R KO mice did not engage in follicular melanogenesis and failed to produce pigmented hair shafts. The current pilot study suggests that the M4R plays a fundamental role in the control of the murine hair follicle cycling and is an essential signaling element in the control of hair follicle pigmentation.     

Inducible deletion of epidermal Dicer and Drosha reveals multiple functions for miRNAs in postnatal skin

MicroRNAs (miRNAs) regulate the expression of many mammalian genes and play key roles in embryonic hair follicle development; however, little is known of their functions in postnatal hair growth. We compared the effects of deleting the essential miRNA biogenesis enzymes Drosha and Dicer in mouse skin epithelial cells at successive postnatal time points. Deletion of either Drosha or Dicer during an established growth phase (anagen) caused failure of hair follicles to enter a normal catagen regression phase, eventual follicular degradation and stem cell loss. Deletion of Drosha or Dicer in resting phase follicles did not affect follicular structure or epithelial stem cell maintenance, and stimulation of anagen by hair plucking caused follicular proliferation and formation of a primitive transient amplifying matrix population. However, mutant matrix cells exhibited apoptosis and DNA damage and hair follicles rapidly degraded. Hair follicle defects at early time points post-deletion occurred in the absence of inflammation, but a dermal inflammatory response and hyperproliferation of interfollicular epidermis accompanied subsequent hair follicle degradation. These data reveal multiple functions for Drosha and Dicer in suppressing DNA damage in rapidly proliferating follicular matrix cells, facilitating catagen and maintaining follicular structures and their associated stem cells. Although Drosha and Dicer each possess independent non-miRNA-related functions, the similarity in phenotypes of the inducible epidermal Drosha and Dicer mutants indicates that these defects result primarily from failure of miRNA processing. Consistent with this, Dicer deletion resulted in the upregulation of multiple direct targets of the highly expressed epithelial miRNA miR-205.     

Enrichment of epidermal stem cells by rapid adherence and analysis of the reciprocal interaction of epidermal stem cells with neighboring cells using an organotypic system

Keratinocytes have the ability to adhere to extracellular matrix rapidly. With this in mind, in this study we isolated keratinocytes known as rapidly adhering (RA) cells. To compare epidermal regenerative abilities, skin substitutes were reconstructed by adding keratinocytes or RA cells to two groups of bioengineered dermis made by fibroblasts and hair follicle dermal cells respectively. After transplantation, the results illustrated that the skin substitutes including RA cells were integrated into the host tissue. Furthermore, with hair follicle dermal cells' influences, the RA cells could form structures very similar to normal hair follicles. These results indicate that RA cells are predominately comprised of epidermal stem cells. The results also demonstrated that besides the reciprocal interaction of epidermal stem cells with dermal cells, the interaction of epidermal stem cells with keratinocytes were critical in epidermis morphogenesis and self-renewal, and application of RA cells could optimize engineering of skin substitutes.     

Restoration of hair growth by surgical implantation of follicular dermal sheath.

The capacity of lower follicle dermal sheath to restore hair growth was tested by removing the lower halves of follicles, and then immediately implanting material containing dermal sheath cells from these bases, into the remaining upper epidermal follicle cavity. Over 60% of recipient follicles produced stout emergent vibrissa fibres and some operations resulted in multiple hair production from a single follicle. Histological examination revealed new dermal papillae within large bulb structures which were sited below the level of amputation--a feature that indicated that the new dermal papilla was derived from implanted material. For many follicles, the failure to produce emergent fibres could be accounted for after histological examination. These results provide clear evidence that lower follicle dermal sheath cells are capable of replacing those of the dermal papilla and it shows that they can do so in the context of the upper follicle. However, because elements of lower follicle epidermis were present in the implant material, the interactive sequence of events cannot be established. Dermal sheath cells have immense potential for papilla cell replacement: questions remain as to whether the distinction between sheath and papilla cells is one of context, or whether the transition requires specific external influences.     

STX17在不同毛色小鼠皮肤中差异表达与在毛囊中的定位

突触融合蛋白17 (STX17)是一种囊泡蛋白,参与细胞中物质的运输.为研究Stx17在不同毛色皮肤中是否存在差异表达及明确它在毛囊中的定位,进行了普通PCR、real-time PCR、免疫组化和蛋白免疫印迹实验对小鼠皮肤组织和体外培养黑素细胞的Stx17基因及蛋白的检测.普通PCR检测得出小鼠皮肤和黑色素细胞总RNA有Stx17 CDS区序列的表达;荧光定量检测显示,在白、灰、黑3种组织中Stx17均有表达,在灰色腹部表达量最高,是黑色皮肤的1.682倍,昆明鼠白色皮肤中表达量最低,是黑色皮肤的0.115倍;皮肤组织免疫组化结果显示,STX17表达于毛囊的上皮根鞘,且毛囊上段和中段表达量高于下段,黑色素细胞的免疫组化分析得出,STX17在黑色素细胞的细胞质和细胞膜上均有表达;蛋白免疫印迹结果显示,在白色、灰色和黑色皮肤均有STX17蛋白阳性条带且灰色皮肤中表达量最高,黑色皮肤次之,白色皮肤中表达量是最低的,这与荧光定量检测结果一致,体外培养的小鼠黑色素细胞中也有STX17蛋白阳性条带.实验结果表明,小鼠Stx17基因在皮肤组织、毛囊角化细胞以及黑色素细胞中均有表达,Stx17可能参与毛色的形成,且在小鼠腹部毛色变浅中起到了一定的作用.