Klaus Sellheyer, MD, Clinical Professor of Dermatology, Department of Dermatology A61, Cleveland Clinic Foundation, 9500 Euclid Avenue, Cleveland, OH 44195, USA. E-mail: email@example.com
Our knowledge on stem cells of the hair follicle has increased exponentially after the bulge was characterized as the stem cell niche two decades ago. In contrast, little is known about stem cells in the nail unit. Whereas hair follicles are plentiful and easy to access, the human body has only twenty nails and they are rarely biopsied. Therefore, examining fetal material offers unique advantages. In the following mini-review, our current knowledge on nail stem cells is summarized and analogies to the hair follicle stem cells are drawn.
Stem cells of the hair follicle have been well-characterized in recent years and have been located in the hair bulge . In adult skin this follicular stem cell niche is difficult to recognize morphologically, however, and consists of only a single cell layer in the external root sheath at the level of the isthmus (Figure 1a). In fetal skin, in contrast, the bulge is distinctive (Figure 1b). Immunohistochemically stem cells can be identified relatively simply with stem cell markers such as cytokeratin 15, cytokeratin 19 or PHLDA1 . Stem cells are also found in the basal cell layer of the interfollicular epidermis, in addition to the hair follicle. These, too, can be stained immunohistochemically . While the literature on stem cells especially of the hair follicle has grown exponentially since the discovery of the bulge as the stem cell niche over two decades ago , the situation with respect to the nail organ is entirely different. Few publications have appeared on the subject. In the mouse, stem cells were described in a portion of the basal cell layer of the nail matrix that adjoins the nail bed . In a single fingernail of a 15-week-old human fetus the stem cell marker cytokeratin 19 stained cells of the basal cell layer of the proximal nail fold as well as also in the basal and suprabasal layer of the ventral matrix . It is unclear if these are stem cells or Merkel cells.
Nail organ and hair follicle
The nail organ and the hair follicle are related skin appendages and have more in common than differences . In diseases, particularly genodermatoses, they are frequently affected together . The similarities are especially clear when – as Achten  proposed – the nail organ is compared with a longitudinal section of a hair follicle turned by 90° (Figure 1c). While anatomical analogies between the hair bulb and nail matrix or between nail plate and hair shaft are evident, in the nail organ an equivalent to the follicular hair bulge with its stem cells cannot be identified in conventional hematoxylin-eosin stained sections. Nonetheless, it can be expected that the nail organ also possesses a stem cell niche analogous to the hair bulge. Of the various anatomical compartments of the nail organ the ventral proximal nail fold and the nail bed most readily come into question conceptually as possible stem cell niches (Figure 1d). Both are – comparable to the hair bulge – protected from the surroundings and parts of the nail organ. If one pursues the hair-nail analogy, the nail matrix is a relatively unlikely stem cell niche, even though this has been proposed. The cells in contrast to those of the hair bulge have a high mitotic activity. This is a characteristic of differentiated cells, while stem cells rarely divide .
Immunohistochemical study of the fetal nail organ
With 100,000 hair follicles on human scalp, dermatologic research possesses an easily bioptically accessible model of cutaneous stem cells. The nail, in contrast, exists in only twenty copies and biopsies of its vital structures are usually performed only to rule out malignant processes, particularly melanoma. This is an important reason for our lack of knowledge about the stem cells of the nail.
Studies on the nail of the human fetus provide a solution. A developmental biologic study has the further advantage of analyzing dynamics of stem cells. A systematic study on stem cells in the embryonic and fetal nail was recently published . A proposal for the location of the stem cell niche was made, and as it is the only published systematic study on the human nail stem cells to date, it will be discussed in detail.
In contrast to animal models, where stem cells can be labeled radioactively or in other manners in order to trace them in their course, this is not possible in humans for ethical reasons. Essentially one is dependent on immunohistochemical studies, particularly as vital material from the human nail organ for studies with the experimental methods of basic science is difficult to obtain. Furthermore, due to the lack of nail organ stem cell-specific stem cell markers, one also depends on the use of markers that have been sufficiently validated on the human hair follicle. In view of the nail-hair analogy this should not represent a problem conceptually, but must, nonetheless, be mentioned. In the study mentioned above, the expression of the following six follicular stem cell markers were studied in the developing nail organs of 17 human fetuses between the 12th and 23rd week of gestation .
1Cytokeratin 15 (clone LHK15): highly specific marker of the hair bulge that, nonetheless, stains stem cells of the interfollicular epidermis as well [2, 11].
2Cytokeratin 15 (clone C8/144B): employed in lymphoma diagnostics to detect CD8-positive cytotoxic T cells and by chance was characterized as a stem cell marker due to its cross-reactivity with cytokeratin 15 .
3Cytokeratin 19: stains stem cells as well as their direct descendants, namely the transient amplifying cells that do not yet represent fully differentiated cells of the hair follicle and can be detected in the entire length of the outer root sheath [2, 11].
4PHLDA1 (TDAG51): characterized with DNA microarrays and is expressed twice as intensely as cytokeratin 15 in the human hair bulge . The abbreviation stands for pleckstrin-homology-like domain, family A, member 1. The underlying gene, also known as “T-cell death-associated gene 51” (TDAG51) is involved in the regulation of apoptosis .
5CD200: also found over-expressed in the human hair bulge in DNA microarray studies  The protein has an immunosuppressive function and protects the follicular stem cell niche from autoimmune attacks .
6Nestin: mesenchymal stem cell marker found in the fibrous connective tissue sheath of the follicle as well as in the papilla of the growing hair follicle [16, 17]. It was employed to detect a possible equivalent of the follicular papilla in the nail organ.
When these six follicular stem cell markers are employed in the human nail organ, a very interesting distribution pattern results that characterizes the ventral proximal nail fold as stem cell niche . Three markers, cytokeratin 15 (clone LHK15), cytokeratin 19 and PHLDA1 proved to be especially informative (Figure 2a). All three antibodies stain the basal cell layer of the epidermis directly proximal from the developing nail early in development (13th and 14th week of gestation), with staining detectable somewhat later and less intensely with PHLDA1. In the 14th week of gestation it becomes clear that this is the primordium of the dorsal proximal nail fold. In the fetal scalp also the basal cell layer of the interfollicular epidermis is stained by the antibodies , while the expression is no longer detectable in this compartment in the adult scalp .
The later nail bed is totally negative with all markers over the entire gestational period studied with exception of cytokeratin 19 that is transiently found in the nail bed of young fetuses, at first in the entire thickness, somewhat later only in a suprabasal cell layer and at the end of the second trimester no longer (Figure 2a). The lack of expression of stem cell markers over the entire developmental period excludes the nail bed as stem cell niche.
In contrast, cytokeratin 15 (clone LHK15), cytokeratin 19 and PHLDA1 are consistently expressed, even if – as in the case of PHLDA1 – only later during embryogenesis (but therefore all the more intensely) in the ventral proximal nail fold (Figure 2a). In the oldest gestational week examined (23rd week) only the outer cell layer of the ventral proximal nail fold – in analogy to the adult hair follicle (Figure 1a) – is positive for the three stem cell markers. Here, too, cytokeratin 19 is again an exception staining the suprabasal cells but not the epidermal ridges of the nail matrix (Figure 2a). As cytokeratin 19 marks not only stem cells but additionally also transient amplifying cells, it can be assumed that the suprabasal cells of the matrix represent a part of the equivalent of the follicular outer root sheath.
The nail matrix itself would only consist of the ridges reaching into the dermis, a concept that is supported by the staining behavior with the proliferation marker Ki67. Ki67-positive cells are actually found predominantly in the epithelial ridges, but not in suprabasal cells . In analogy, Ki67–labeled cells accumulate in the hair follicle exclusively below Auber's line that is found at the level of the greatest diameter of the hair bulge . While cytokeratin 15 (clone LHK15), cytokeratin 19 and PHLDA1 are consistently expressed in the ventral proximal nail fold, this does not hold true for the dorsal proximal nail fold (Figure 2a). Consistent with the -literature  only cytokeratin 15 (clone LHK15) is continually detectable in the basal cell layer of the dorsal and ventral proximal nail fold, but not cytokeratin 19 and PHLDA1. The former is found in the dorsal proximal nail fold in later gestational weeks only in scattered cells at the tips of the eccrine ridges and thus probably demarcates Merkel cells . PHLDA1 stains only confined cell accumulations in the dorsal proximal nail fold in the later course of development as well as the eccrine ridges, but is otherwise negative (Figure 2a) [10, 20].
Cytokeratin 15 (clone C8/144B) and CD200 were not expressed in the fetal nail organ at all and in the fetal hair follicle only late in development . As both proteins are involved, either directly like CD200 or indirectly like the C8 antigen, via apoptosis regulation in immunosuppression, it can be assumed that the lack of staining can be attributed to the insufficiently developed immune system in the fetus. Nestin, likewise, does not mark stem cells in the nail. An equivalent to the nestin-positive hair papilla also cannot be detected with nestin in the nail . This may also be related to the fact that the nail, in contrast to the hair follicle, does not follow a cycle.
Despite the fact that the study discussed did not examine the adult nail organ, the older gestational weeks studied -already have all characteristics of postnatal nails and an -extrapolation of results to adult nails seems justified. Due to the consistent expression of cytokeratin 15 (clone LHK15), cytokeratin 19 and PHLDA1 in the ventral proximal nail fold in contrast to all other structures, this represents the stem cell niche of the human nail organ (Figure 1b) and thus the equivalent of the hair bulge (Figure 2c). Subsequent studies on adult human nails are, nonetheless, desirable.