Id proteins are dynamically expressed in normal epidermis and dysregulated in squamous cell carcinoma.

Helix-loop-helix genes regulate many developmental pathways, and growing evidence associates dysregulated expression with tumorigenesis. We observed Id-1, Id-2, and Id-3 mRNA expression in proliferating human keratinocytes in vitro with subsequent down-regulation with differentiation. Immunohistochemical analysis of human tissue sections identified cytoplasmic Id-1 expression and nuclear Id-2 and Id-3 expression in the proliferating layers of the epidermis. Furthermore, we observed a columnar pattern of Id-2 and Id-3 staining, which may relate to the epidermal proliferative unit. In squamous cell carcinoma of the head and neck, Id protein immunoreactivity was observed in the majority of malignant keratinocytes in the most poorly differentiated sections, with reduced staining in well-differentiated disease.

The effects of inflammatory cytokines on the isolated human sebaceous infundibulum.

The human sebaceous pilosebaceous infundibulum was isolated and maintained in medium for up to 7 d. Freshly isolated infundibula were found to express keratins 1, 5, 6, 16, and 17, as determined by immunohistochemistry. In addition, freshly isolated infundibula expressed filaggrin, profilaggrin, involucrin, cornifin alpha, and loricrin. This pattern of expression was retained over 7 d. The addition of 100 U interferon (IFN)-gamma per ml over 3 d and 1 nM phorbol myristate acetate over 24 h resulted in the expression of intercellular adhesion molecule (ICAM)-1 and HLA-DR by infundibular keratinocytes, as determined by immunohistochemistry. Ten nanograms tumor necrosis factor-alpha per ml and 10 ng IL-6 per ml both caused expression of ICAM-1 alone. IL-1alpha had no effect on the expression of ICAM-1 or HLA-DR over 3 d, but addition of 1 ng IL-1alpha per ml over 7 d in culture resulted in hypercornification of the keratinocytes of the infundibulum, apparently brought about by early keratinocyte cornification. These data suggest that the isolated, maintained, infundibulum is a good model for studying the effects of inflammatory cytokines on the infundibulum, and that IL-1alpha acts on infundibular keratinocytes to promote cornification.

The improved organ maintenance of the human sebaceous gland: modeling in vitro the effects of epidermal growth factor, androgens, estrogens, 13-cis retinoic acid, and phenol red.

We have previously reported that human sebaceous glands can be maintained for up to 14 d as whole organs with full retention of the physiological rate and pattern of new cell formation, but we have also reported that the newly formed cells did not differentiate normally, causing a progressive loss of lipogenesis in vitro. We now show that this abnormal sebocyte differentiation was attributable to the presence of epidermal growth factor (EGF) and phenol red in our maintenance medium. In their absence, human sebaceous glands apparently retain in vivo rates of cell division and lipogenesis over 7 d of maintenance in addition to a retention of in situ morphology. This is reversible on the re-addition of 10 ng EGF/ml and 10 mg phenol red/ml. The addition of 600 pM 17 beta-estradiol results in a significant fall in the rate of lipogenesis over 7 d of maintenance, without affecting the rate of cell division. This effect is apparently due to abnormal differentiation of newly formed sebocytes. Neither 1 nM testosterone nor 1 nM dihydrotestosterone (DHT) has any effect on rates of cell division of lipogenesis over 7 d. In the presence of phenol red, however, 1 nM testosterone or 1 nM DHT cause a significant reduction in the rate of lipogenesis over 7 d of maintenance. One micromolar 13-cis retinoic acid caused a significant reduction in the rate of lipogenesis over 7 d in both the presence and absence of phenol red. These findings show that we can model the physiological effects of steroids, EGF, and 13-cis retinoic acid in vitro.

Lipogenesis in the human sebaceous gland: glycogen and glycerophosphate are substrates for the synthesis of sebum lipids.

The lipid class compositions of freshly isolated and overnight maintained human sebaceous glands were determined using high performance thin layer chromatography and were found to be broadly similar. The lipid classes identified in freshly isolated glands were broadly similar to those generated from various radiolabeled precursors in vitro, although lower amounts of wax/sterol esters and cholesterol were observed in vitro. We determined the glycogen content of sebaceous glands and showed that during their incubation with several radiolabeled substrates, with the exception of [U-14C]glucose, significant glycogen breakdown occurred, thus providing acetyl-CoA and NADPH that could affect the pattern of lipids synthesized. We examined glycogen-depleted glands and found that their rates and pattern of lipogenesis resemble that of nondepleted glands, except that the squalene/triacylglycerides ratio for acetate and glutamine doubled from 1.6:1 to 3.4:1 and from 0.7:1 to 1.4:1, respectively. We have shown that exogenous glycerol reduces the squalene/triacylglycerides ratio from acetate from 3.4:1 to 1.7:1, suggesting that glycogen-derived glycerophosphate is important in triacylglycerides synthesis from acetate. Moreover, glands express a glycerokinase activity that may fully account for the rate of triacylglycerides synthesis seen in vitro. We conclude that glycerophosphate generation may be as important as NADPH generation in explaining the directing effects of different substrates within the sebaceous gland.

Metabolic studies on isolated hair follicles: hair follicles engage in aerobic glycolysis and do not demonstrate the glucose fatty acid cycle.

The matrix cells of the hair follicle have one of the highest rates of cell division in the mammalian body, but their fuel metabolism is poorly understood, due mainly to the difficulty in obtaining viable intact follicles from the skin. We have previously shown that viable and intact rat hair follicles can be isolated by shearing, and in this study we now report on their fuel metabolism. In this study we have shown that the hair follicle exhibits aerobic glycolysis, in that of the total glucose utilized by the hair follicle, only 10% is oxidized to CO2. We have also shown that, in the absence of glucose, the hair follicle is capable of utilizing other fuels such as palmitate and beta-hydroxybutyrate. However, neither palmitate or beta-hydroxybutyrate had any effect on the rate of glucose utilization or on [U-14C] glucose oxidation, showing that glucose sparing via the glucose fatty acid cycle does not operate in the hair follicle. Measurements of glucose flux through the pentose phosphate pathway accounted for only 3% of the total glucose utilized by the hair follicle, although this value represented 32% of the total glucose oxidized. Both palmitate and beta-hydroxybutyrate inhibited glucose flux through the pentose phosphate pathway.

Effects of EGF on the morphology and patterns of DNA synthesis in isolated human hair follicles.

We have previously reported that human hair grows at a normal rate in vitro for up to 10 d. We have also reported that, on gross observation, epidermal growth factor appears to induce a catagen-like effect on cultured hair follicles, but we have not characterized the details of this. We now report that when isolated human hair follicles are maintained in the presence of epidermal growth factor, the rate of hair follicle elongation is significantly stimulated but hair fiber production is inhibited. Light microscopy showed that epidermal growth factor stimulated a thickening and vacuolation of the cells of the lower outer root sheath of the hair follicle and that the matrix cells of the hair follicle underwent an upward migration resulting in the formation of a 'club hair'-like structure that remained connected to the dermal papilla by a thin strand of epithelial cells. [Methyl-3H] thymidine autoradiography was carried out to investigate the patterns of DNA synthesis and showed that epidermal growth factor inhibited DNA synthesis in the hair follicle matrix cells but dramatically stimulated DNA synthesis in the outer root sheath. We conclude from these studies that epidermal growth factor may be inducing an artificial 'catagen-like' effect by stimulating outer root sheath proliferation, which uncouples the normal patterns of proliferation and migration that occur in the anagen hair follicle and that result in an anagen-to-catagen-like transition. Moreover, these results also suggest that, under certain conditions, outer root sheath cells in the hair follicle may be capable of downward migration.

Cyclical changes in rat vibrissa follicles maintained In vitro.

In mammals hair growth is cyclical; however, the factors that regulate the hair growth cycle are still poorly understood. The recent development of methods for culturing hair follicles in vitro has proved an important tool to investigate many aspects of the regulation of hair follicle growth. At present, however, these models are based on the culture of anagen hair follicles and have only partially been used to address the cyclical nature of hair growth. In this study we have made use of the fact that in rodents the hair growth cycle is synchronized, well characterized, and relatively short. We have isolated vibrissa follicles from 12 d old rats and confirmed by histology that these follicles are in the anagen stage of their first hair growth cycle. We have then maintained these follicles in vitro, on Gelfoam supports, for up to 23 d (35 d of age) and compared their histology with in vivo follicles from equivalent age littermates. We observed that 12 d old follicles maintained in vitro for up to 23 d show changes in morphology that suggest that cultured rat vibrissa follicles retain cyclical activity in vitro. Cyclical changes in hair follicle morphology were only seen in follicles maintained on gelfoam supports and moreover, hair follicle size appears to be a key feature in determining the ability of the follicle to cycle in vitro. All follicles that showed cyclical changes in vitro, however, appeared to remain blocked in pro-anagen. These data suggest that the vibrissa follicle is a in vitro good model system with which to investigate hair cycle control. J Invest Dermatol 115:1152-1155 2000

The activity of HMG-CoA reductase and acetyl-CoA carboxylase in human apocrine sweat glands, sebaceous glands, and hair follicles is regulated by phosphorylation and by exogenous cholesterol.

Human apocrine and sebaceous glands function to secrete lipids, predominantly triglycerides, fatty acids, cholesterol and its esters, and, in the sebaceous gland, squalene. The enzymes that catalyze the important regulatory steps in cholesterol and fatty acid biosyntheses, 3-hydroxy-3-methylglutaryl-coenzyme A (HMG-CoA) reductase and acetyl-CoA carboxylase, respectively, were therefore studied in isolated human skin appendages, and their relevant kinetic parameters determined. The enzyme activities that were observed can account for previously described rates of incorporation of radiolabeled substrates into the appropriate lipids by glands in vitro. Reduced enzyme activities following homogenization in the presence of fluoride indicated that both of these enzymes in skin appendages are inactivated by phosphorylation. The activity of the enzyme known to catalyze this phosphorylation, the AMP-activated protein kinase, was also measured. Compactin was shown to inhibit HMG-CoA reductase in homogenates of these appendages. Conversely, incubation of whole sebaceous glands with compactin resulted in the stimulation of enzyme activity, which suggests that these appendages can respond to diminishing cholesterol levels. The effect of exogenous low density lipoprotein and 25-hydroxycholesterol on HMG-CoA reductase activity from skin appendages was investigated. HMG-CoA reductase activity in both apocrine and sebaceous glands was reduced following incubation with either low density lipoprotein or 25-hydroxycholesterol. Low density lipoprotein receptor and lipoprotein lipase mRNA expression was also detected in skin appendages. These results indicate that apocrine and sebaceous glands have the capacity to sequester dietary cholesterol and fatty acids that may have important implications for the understanding of both acne and axillary odor.

Effects of insulin and insulin-like growth factors on cultured human hair follicles: IGF-I at physiologic concentrations is an important regulator of hair follicle growth in vitro.

Insulin stimulated hair follicle growth in a dose-dependent manner over the range of 0.01 to 100 micrograms/ml. Maximum rates of hair follicle growth were observed when follicles were maintained in medium containing 10 micrograms/ml insulin, which is supraphysiologic. Hair follicles maintained in the absence of insulin or at physiologic levels showed premature entry into a catagen-like state. Insulin-like growth factor (IGF)-I and -II had no significant effect on hair follicle growth when maintained in the presence of 10 micrograms/ml insulin. However, in the absence of insulin, both IGF-I (0.01-100 ng/ml) and IGF-II (0.01-100 ng/ml) stimulated hair follicle growth in a dose-dependent manner. IGF-I was more potent than either insulin or IGF-II, stimulating maximum rates of hair follicle growth at 10 ng/ml, whereas IGF-II gave maximum stimulation at 100 ng/ml. The rates of hair follicle growth stimulated by 10 ng/ml IGF-I were identical to those stimulated by 10 micrograms/ml insulin. IGF-II (100 ng/ml), however, was unable to stimulate hair follicle growth to the same extent as insulin. Both IGF-I (10 ng/ml) and IGF-II (100 ng/ml) were more potent than insulin at preventing hair follicles from entering into a catagen-like state. Growth hormone had no effect on hair follicle growth or morphology in the absence of insulin. These data suggest that in vitro IGF-I may be an important physiologic regulator of hair growth and possibly the hair growth cycle. Moreover, the removal of insulin from tissue culture medium may be a useful method of generating large numbers of catagen hair follicles for further in vitro studies.

Modeling acne in vitro.

To help elucidate the factors responsible for the infundibular changes seen in acne, the human sebaceous pilosebaceous infundibulum was isolated by microdissection and maintained for 7 d in keratinocyte serum-free medium supplemented with 50 micrograms/ml bovine pituitary extract, 100 units/ml penicillin and streptomycin, 2.5 micrograms/ml amphotericin B and CaCl2(10H2O) to give a final Ca2+ concentration of 2 mM. Infundibular structure was maintained over 7 d in this medium; the pattern of cell division mimicked that in vivo. The rate of cell division was significantly higher than previously described for infundibula maintained in supplemented William's E medium, and moreover did not fall over 7 d. The addition of 1 ng/ml interleukin-1 alpha (IL-1 alpha) caused hypercornification of the infundibulum similar to that seen in comedones; this could be blocked by 1000 ng/ml interleukin-1 receptor antagonist (IL-1ra). In about 20% of subjects there was spontaneous hypercornification of the infundibulum that could be blocked by 1000 ng/ml IL-1ra, suggesting that the infundibulum is capable of synthesising IL-1 alpha. The addition of 5 ng/ml epidermal growth factor or 5 ng/ml transforming growth factor-alpha to the medium caused a disorganisation of the keratinocytes of the infundibulum that resulted in rupturing similar to that seen in the more severe, purulent grades of acne. The addition of 1 microM 13-cis retinoic acid caused a significant reduction in the rate of DNA synthesis and apparent parakeratosis. We are now, therefore, able to model histologically the major infundibular changes in acne.

Metabolism of freshly isolated human hair follicles capable of hair elongation: a glutaminolytic, aerobic glycolytic tissue.

The metabolism of the human hair follicle was investigated in vitro under conditions that maintained glycogen and adenosine triphosphate (ATP) content and the growth rate of the follicle at values observed in vivo. We have shown that only 10% of the total glucose utilized was oxidized to CO2 and 40% of this was oxidized via the pentose phosphate shunt. Although fatty acids and ketone bodies were oxidized by the hair follicle, they are poor energetic substitutes for glucose. Nor will fatty acids or ketone bodies sustain hair growth in vitro. Glutamine, however, was shown, both biochemically and by comparing growth rates, to be an important fuel with 23% of uptake being oxidized, generating a possible 2.16 +/- 0.32 nmoles ATP/follicle/h (mean +/- SEM) (glucose metabolism generates 4.54 +/- 0.61 nmoles ATP/follicle/h). Sixty-four percent of the glutamine taken up was calculated to be metabolized to lactate, showing that the hair follicle engages in both glycolysis and glutaminolysis. The glucose-fatty acid cycle appears to be unimportant in the hair follicle but our data indicates that a glucose-glutamine cycle does operate.

The role of IGF-I in human skin and its appendages: morphogen as well as mitogen?

Previous studies have investigated the expression of insulin-like growth factor-I (IGF-I) and its receptor in cultured skin cells or in whole skin. In order to fully understand the role of IGF-I in the skin and its appendages, however, a comprehensive study that details the expression of IGF-I and the IGF-I receptor in sections of human skin is needed. Therefore, we now report an immunocytochemical and in situ hybridization localization study of the cell types expressing IGF-I and its receptor in human adult skin and its appendages. We have observed that (i) dermal fibroblasts produce IGF-I, (ii) the epidermal basal keratinocytes are IGF-I negative but IGF-I receptor positive, and (iii) the keratinocytes of the stratum granulosum produce IGF-I. These observations indicate either that the mitogenesis of the basal keratinocytes is regulated by IGF-I expressed both in the dermis and in the stratum granulosum, or that dermal fibroblasts are responsible for sequestering IGF-I to the basal keratinocytes and that the stratum granulosum-derived IGF-I may be an autocrine regulator of epidermal differentiation. The distribution of IGF-I and its receptor in the hair follicle indicates that IGF-I may be a morphogen, not a mitogen, at those sites, because their proliferating cells, but not their differentiating cells, are IGF-I receptor negative. Further, IGF-I receptor expression by the dermal papilla appears to be switched off during the transition from anagen to catagen, which implies a regulatory role for IGF-I during the hair growth cycle.

Evidence for sodium-coupled acid-base transport across the basolateral membrane of the reabsorptive duct of the human eccrine sweat gland.

Intracellular pH was measured in isolated nonperfused ducts of human eccrine sweat glands in vitro to investigate basolateral acid-base transport mechanisms. Bath sodium removal led to a bicarbonate-independent, 4-acetamido-4'-isothiocyanatostilbene-2,2'-disulfonic acid insensitive acidification. The recovery of this acidification was ethylisopropyl amiloride sensitive, suggestive of basolateral sodium:hydrogen exchange. Whereas bath chloride removal led to a small acidification this was not 4-acetamido-4'-isothiocyanatostilbene-2,2'-disulfonic acid sensitive and its causes remain unclear. Elevation of bath potassium to depolarize the basolateral membrane led to a small alkalinization but this was not mimicked by addition of barium or chloride removal. As chloride removal and barium addition would be expected to cause larger depolarizations than potassium elevation these observations do not support a major role for electrogenic acid-base transport. In conclusion, although this study does not support a major role for electrogenic acid-base transport, it has demonstrated the basolateral presence of sodium-coupled acid-base transport in the reabsorptive duct of the human eccrine sweat gland, which most likely represents a sodium:hydrogen exchanger involved in regulation of intracellular pH.

In the absence of streptomycin, minoxidil potentiates the mitogenic effects of fetal calf serum, insulin-like growth factor 1, and platelet-derived growth factor on NIH 3T3 fibroblasts in a K+ channel-dependent fashion.

There is considerable evidence to suggest that the opening of K+ channels plays an important role in stimulating mitogenesis. K+ channel blockers have been shown to inhibit mitogenesis in vitro, mitogens increase cytosolic membrane K+ channel permeability, K+ channel openers stimulate hair growth in vivo, and the Ras/Raf signal transduction pathway induces K+ channel activity. Paradoxically, however, K+ channel openers such as minoxidil have been reported in vitro not to modulate, or even to inhibit, mitogenesis in a range of cell types. Only untherapeutic concentrations have stimulated mitogenesis. These experiments, however, appear to have been carried out in the presence of aminoglycoside antibiotics, which inhibit potassium channel activity. We now report that in the absence of aminoglycoside antibiotics, minoxidil at 10 microg/ml (0.05 mM) causes a significant stimulation of proliferation of NIH 3T3 fibroblasts maintained over a 10-d period in 5% fetal calf serum-supplemented medium. Further, we show that in the presence of 100 microg streptomycin per ml, minoxidil at 10 microg/ml produces an initial inhibition of proliferation, which apparently confirms, in NIH 3T3 fibroblasts, that the inhibition of mitogenesis by minoxidil in the presence of streptomycin is an artifact. The potentiation of NIH 3T3 cell growth by minoxidil can be attributed to the opening of potassium channels, because the potassium channel blocker tolbutamide (5 mM) or combinations of the blockers tolbutamide (1 mM)/tetraethylammonium (2 mM) or glibenclamide (1 microM)/apamin (10 nM) block the minoxidil-induced stimulation of growth. We also demonstrate that minoxidil is able to significantly potentiate the mitogenic effects of both platelet-derived growth factor and insulin-like growth factor 1 on NIH 3T3 fibroblasts in the presence of CPSR-2 (a cytokine free serum substitute). Thus we have shown that minoxidil potentiates the mitogenic effects of fetal calf serum in vitro on NIH 3T3 fibroblasts by opening potassium channels and is also able to potentiate the mitogenic effects of the growth factors platelet-derived growth factor and insulin-like growth factor 1.

Lipogenesis by isolated human apocrine sweat glands: testosterone has no effect during long-term organ maintenance.

Lipid synthesis by freshly isolated human apocrine glands has been measured by the incorporation of [U-14C] acetate. Incorporation is linear over 6 h at 1010 +/- 282 pmol/mg wet weight/h (n = 11; mean +/- sem). The lipid classes, as percentages of the total lipid synthesized, were found by TLC to be cholesterol 12.3 +/- 2.0, mono-glycerides 7.5 +/- 1.5, 1,2 di-glycerides 3.0 +/- 0.9, 1,3 di-glycerides 3.5 +/- 0.5, tri-glycerides 28.4 +/- 1.8, free fatty acids 2.0 +/- 0.4, lysolecithin 15.4 +/- 3.9, sphingomyelin 9.9 +/- 4.3, phosphatidyl-choline 8.4 +/- 0.4, phosphatidyl-ethanolamine -inositol and -serine 1.8 +/- 0.1, phosphatidic acid and cardiolipin 3.3 +/- 0.5, and unidentified 3.3 +/- 0.5 (mean +/- sem, n = 5). Glands were maintained on permeable supports. After 10 d maintenance, electron microscopy showed that the cellular architecture had been preserved, that the ATP contents were the same as in freshly isolated glands, and that [U-14C] acetate incorporation was not significantly altered at 851 +/- 237 pmol/mg/h (n = 18). The addition of 3 microM testosterone had no effect on acetate incorporation at 844 +/- 231 pmol/mg/h (n = 18). The lipid classes and their proportions were similar to the values for fresh glands after 10 d maintenance both with and without testosterone.

Rapid isolation in large numbers of intact, viable, individual hair follicles from skin: biochemical and ultrastructural characterization.

A rapid, novel method is described by which large numbers of intact, viable, individual hair follicles may be isolated from rat skin. Follicles are freed from the surrounding connective tissue by shearing, which is effected by repeated cutting with a loosely fitting pair of scissors, and collected individually under liquid using gentle aspiration. Ultrastructural analysis indicates that the follicles are sheared away from the surrounding dermis in the region of the connective tissue capsule which encircles the hair. The follicles appear viable by light and electron microscopy and, within 2 h of isolation, retain the capacity to incorporate [3H]thymidine into DNA and [35S]methionine into proteins as judged by autoradiography. A histologic comparison indicates that the structural integrity of follicles isolated by this new method is significantly superior to those plucked from the animal at the same time. The method affords the isolation of large numbers of hair follicles, without resort to enzyme treatments, suitable for biologic studies in the absence of other skin appendages and dermis.

cDNA cloning of a human androgen-induced mRNA exhibiting an early and protein synthesis-independent induction.

The detailed mechanism of action of androgens remains unknown. We have used an androgen-dependent human prostate cancer cell line and a subtractive cDNA hybridisation strategy to enrich for androgen-dependent sequences. This yielded a cDNA clone which exhibits the characteristics of a primary trans-activated target for androgens. This androgen-regulated gene encodes a polyadenylated 4.5 kb mRNA which is induced 30-50-fold within 3 h of treatment with 10 nM dihydrotestosterone. The induction does not require continued protein synthesis as it is maintained in the presence of protein synthesis inhibitors.