Moisturizing treatment of patients with atopic dermatitis and ichthyosis vulgaris improves dry skin, but has a modest effect on gene expression regardless of FLG genotype.

BACKGROUND: Loss-of-function mutations in FLG (encoding filaggrin) are a predisposing factor for atopic dermatitis (AD) and cause ichthyosis vulgaris (IV). Patients with AD and IV display impaired skin barrier and dry skin, and altered epidermal expression of genes in pro-inflammatory and lipid metabolic pathways are often evident. OBJECTIVES: To evaluate the effect of three different moisturizers on skin barrier function and epidermal gene expression in patients with AD/IV in relation to FLG mutation status. METHODS: Patients (n = 43) were classified according to their FLG status: AD with FLG+/+ (n = 14), AD with FLG+/- (n = 14), and AD/IV with FLG-/- (n = 15). Dryness score and transepidermal water loss (TEWL) were monitored on volar forearms, and punch biopsies were taken for analysis of gene expression. Measurements were repeated after 4 weeks of treatment with either of two moisturizers on each forearm. RESULTS: Treatment with any of the three moisturizers significantly reduced dryness score and TEWL in the group as a whole. FLG-/- patients displayed the largest reduction in dryness score. Only minute changes occurred in the mRNA expression of 15 selected epidermal genes. CONCLUSIONS: Moisturizing treatment improves dry skin and certain aspects of abnormal skin barrier function, especially in patients with AD/IV and dual FLG mutations, but does not normalize the epidermal gene expression profile.

The effect of two endogenous retinoids on the mRNA expression profile in human primary keratinocytes, focusing on genes causing autosomal recessive congenital ichthyosis.

Retinoids (natural forms and synthetic derivatives of vitamin A) are used as therapeutic agents for numerous skin diseases such as keratinization disorders (e.g. ichthyoses) and psoriasis. Two endogenous ligands for retinoic acid receptors exist, retinoic acid (atRA) and 3,4-didehydroretinoic acid (ddRA). In primary human epidermal keratinocytes many transcriptional targets for atRA are known, whereas the targets for ddRA are unknown. In an attempt to determine the targets, we compared the effect of atRA and ddRA on transcriptional profiles in undifferentiated and differentiating human primary keratinocytes. First, as expected, many genes were induced or suppressed in response to keratinocyte differentiation. Furthermore, the two retinoids affected substantially more genes in differentiated keratinocytes (>350) than in proliferating keratinocytes (≈20). In differentiating keratinocytes markers of cornification were suppressed suggesting a de-differentiating effect by the two retinoids. When comparing the expression profile of atRA to that of ddRA, no differently regulated genes were found. The array analysis also found that a minor number of miRNAs and a large number of non-coding transcripts were changed during differentiation and in response to the two retinoids. Furthermore, the expression of all, except one, genes known to cause autosomal recessive congenital ichthyosis (ARCI) were found to be induced by differentiation. These results comprehensively document that atRA and ddRA exert similar transcriptional changes in keratinocytes and also add new insights into the molecular mechanism influenced by retinoids in the epidermis. Furthermore, it suggests which ARCI patients could benefit from therapy with retinoids.

X-linked recessive ichthyosis: an impaired barrier function evokes limited gene responses before and after moisturizing treatments.

BACKGROUND: X-linked recessive ichthyosis (XLRI) is due to deletions or inactivating mutations in the steroid sulfatase (STS) gene. This results in an accumulation of cholesterol sulphate affecting the packing of intercorneocyte lipids. XLRI is characterized by dry, scaly skin and increased skin barrier permeability; patients are often dependent on daily use of moisturizers. OBJECTIVES: To examine the biophysical and molecular changes in the skin of patients with XLRI compared with healthy volunteers, and to analyse the effects of moisturizers on the patients' barrier function. METHODS: Patients with XLRI (n=14) and healthy controls (n=14) were included in the study. Skin dryness score, transepidermal water loss (TEWL) and skin surface pH were monitored at baseline, and punch biopsies were obtained for mRNA expression profiles determined by oligonucleotide arrays. Measurements were repeated in the patients with XLRI after a 4-week treatment with three different moisturizers on the volar forearms. RESULTS: Patients with XLRI showed, compared with healthy controls, increased dryness and TEWL, equal skin pH and altered expression of 27 genes. There were no signs of activation of inflammation or repair pathways. Five selected genes were significantly altered also on quantitative polymerase chain reaction analysis. Treatment with the moisturizers showed similar effects: they improved skin dryness but had no effect on TEWL, pH or expression of selected genes. CONCLUSIONS: Despite a dysfunctional skin barrier, the limited number of genes altered in XLRI skin suggests that no inflammatory or repair mechanisms are triggered. Treatment with moisturizers does not have any major impact on the skin barrier properties of patients with XLRI.

Immortalized keratinocytes derived from patients with epidermolytic ichthyosis reproduce the disease phenotype: a useful in vitro model for testing new treatments.

BACKGROUND: Epidermolytic ichthyosis (EI) is a skin fragility disorder caused by mutations in genes encoding suprabasal keratins 1 and 10. While the aetiology of EI is known, model systems are needed for pathophysiological studies and development of novel therapies. OBJECTIVES: To generate immortalized keratinocyte lines from patients with EI for studies of EI cell pathology and the effects of chemical chaperones as putative therapies. METHODS: We derived keratinocytes from three patients with EI and one healthy control and established immortalized keratinocytes using human papillomavirus 16-E6/E7. Growth and differentiation characteristics, ability to regenerate organotypic epidermis, keratin expression, formation of cytoskeletal aggregates, and responses to heat shock and chemical chaperones were assessed. RESULTS: The cell lines EH11 (K1_p.Val176_Lys197del), EH21 (K10_p.156Arg>Gly), EH31 (K10_p.Leu161_Asp162del) and NKc21 (wild-type) currently exceed 160 population doublings and differentiate when exposed to calcium. At resting state, keratin aggregates were detected in 9% of calcium-differentiated EH31 cells, but not in any other cell line. Heat stress further increased this proportion to 30% and also induced aggregates in 3% of EH11 cultures. Treatment with trimethylamine N-oxide and 4-phenylbutyrate (4-PBA) reduced the fraction of aggregate-containing cells and affected the mRNA expression of keratins 1 and 10 while 4-PBA also modified heat shock protein 70 (HSP70) expression. Furthermore, in situ proximity ligation assay suggested a colocalization between HSP70 and keratins 1 and 10. Reconstituted epidermis from EI cells cornified but EH21 and EH31 cells produced suprabasal cytolysis, closely resembling the in vivo phenotype. CONCLUSIONS: These immortalized cell lines represent a useful model for studying EI biology and novel therapies.

Epidermolysis bullosa simplex due to KRT5 mutations: mutation-related differences in cellular fragility and the protective effects of trimethylamine N-oxide in cultured primary keratinocytes.

BACKGROUND: Epidermolysis bullosa simplex (EBS) is a mechanobullous skin fragility disease characterized by cytolysis of basal keratinocytes and intraepidermal blistering often caused by mutations in keratin genes (KRT5 or KRT14). No remedies exist for these disorders presenting a need for development of novel therapies. OBJECTIVES: To identify new genotype-phenotype relationships in vivo and in cultured primary EBS keratinocytes in vitro, and to study the cytoskeletal stabilizing effects of trimethylamine N-oxide (TMAO) in heat-stressed EBS cells. METHODS: Genomic DNA and cDNA samples from three Swedish patients with EBS were analysed for keratin mutations. Primary EBS keratinocyte cultures were established, heat stressed with and without added TMAO, followed by evaluation of cellular fragility. RESULTS: In addition to the previously reported KRT5 mutation (V186L) in one patient, two patients were found to have a novel I183M and recurrent E475G replacements in KRT5. Cultured EBS keratinocytes did not exhibit keratin aggregates or cell loss, except in the patient with the p.I183M mutation who showed 3% aggregates and 2% cell loss. Upon transient heat stress the number of aggregate-containing cells increased to 21%, 27% and 13%, respectively, in the p.I183M, p.E475G and p.V186L mutant cells. Interestingly, pretreatment with TMAO prior to heat stress, dose dependently reduced the number of aggregate-containing cells and cell loss. CONCLUSION: These results revealed a genotype-phenotype correlation in EBS keratinocytes upon heat stress and suggest protein stabilization as a new therapeutic strategy.

Topical treatment with CYP26 inhibitor talarozole (R115866) dose dependently alters the expression of retinoid-regulated genes in normal human epidermis.

BACKGROUND: An alternative approach to retinoid therapy is to inhibit the cytochrome P450 (CYP)-mediated catabolism of endogenous all-trans retinoic acid in the skin by applying retinoic acid metabolism blocking agents such as talarozole (R115866). OBJECTIVES: To study the effects of topical talarozole on retinoid biomarkers in normal skin in a randomized phase I trial. METHODS: Gels containing talarozole (0.35% or 0.07%) and vehicle were applied once daily for 9 days on either buttock of 16 healthy volunteers. Epidermal shave biopsies (for mRNA analysis) and punch biopsies (for histology and immunofluorescence analysis) were collected from the treatment areas. Genes encoding the following were studied by quantitative real-time polymerase chain reaction: cellular retinoic acid binding protein 2 (CRABP2), cytokeratins (KRT2 and KRT4), CYP26A1, CYP26B1, CYP26C1 and CYP2S1, two enzymes in the retinol metabolism (retinal dehydrogenase-2 and retinol acyltransferase) and two proinflammatory cytokines [interleukin (IL)-1alpha and tumour necrosis factor-alpha]. RESULTS: Talarozole treatment increased the mRNA expression of CRABP2, KRT4, CYP26A1 and CYP26B1 dose dependently, and decreased the expression of KRT2 and IL-1alpha compared with vehicle-treated skin. No mRNA change in retinol-metabolizing enzymes was obtained. There was no induction of epidermal thickness or overt skin inflammation in talarozole-treated skin. Immunofluorescence analysis confirmed an upregulation of KRT4 protein, but no upregulation of CYP26A1 and CYP26B1 expression was detected. CONCLUSIONS: Talarozole influences the biomarker pattern consistently with increased retinoic acid stimulation. The low irritancy of talarozole at the two examined dosages is a possible advantage over topical retinoids.

Changes in skin barrier function following long-term treatment with moisturizers, a randomized controlled trial.

BACKGROUND: Moisturizers are commonly used by patients with dry skin conditions as well as people with healthy skin. Previous studies on short-term treatment have shown that moisturizers can weaken or strengthen skin barrier function and also influence skin barrier recovery. However, knowledge of the effects on skin barrier function of long-term treatment with moisturizers is still scarce. OBJECTIVES: To investigate the impact of long-term treatment with moisturizers on the barrier function of normal skin, as measured by transepidermal water loss (TEWL) and susceptibility to an irritant, and to relate those effects to the composition of the designed experimental moisturizers. METHODS: Volunteers (n = 78) were randomized into five groups. Each group treated one volar forearm for 7 weeks with one of the following preparations: (i) one of three simplified creams, containing only a few ingredients in order to minimize the complexity of the system; (ii) a lipid-free gel; (iii) one ordinary cream, containing 5% urea, which has previously been shown to decrease TEWL. The lipids in the simplified creams were either hydrocarbons or vegetable triglyceride oil, and one of them also contained 5% urea. After 7 weeks, treated and control forearms were exposed for 24 h to sodium lauryl sulfate (SLS) using a patch test. TEWL, blood flow and skin capacitance of both SLS-exposed and undamaged skin were evaluated 24 h after removal of patches. Additionally, a 24-h irritancy patch test of all test preparations was performed on 11 volunteers in order to check their possible acute irritancy potential. RESULTS: Changes were found in the barrier function of normal skin after 7 weeks of treatment with the test preparations. The simplified creams and the lipid-free gel increased TEWL and skin response to SLS, while the ordinary cream had the opposite effect. One of the simplified creams also decreased skin capacitance. All test preparations were shown to be nonirritant, both by short-term irritancy patch test and by measurement of blood flow after long-term treatment. CONCLUSIONS: Moisturizers influence the skin barrier function of normal skin, as measured by TEWL and susceptibility to SLS. Moreover, the effect on skin barrier function is determined by the composition of the moisturizer. The ingredients which influence the skin barrier function need to be identified, and the mechanism clarified at the molecular level.

Inefficacy of topical thyroid hormone analogue TriAc in plaque psoriasis: results of a double-blind placebo-controlled trial.

BACKGROUND: Thyroid hormone receptors are expressed in human skin and are believed to be involved in the regulation of epidermal proliferation and differentiation, i.e. processes which are disturbed in psoriatic skin lesions. Ligands of the thyroid hormone receptors have so far not been tested as antipsoriatic agents. TriAc (3,3',5-triiodo-thyroacetic acid) is a well-known thyroid hormone analogue with much reduced cardiac thyrotoxic activity compared with the classical thyroid hormones. OBJECTIVES: To determine the effectiveness and side-effects of topical TriAc in patients with chronic plaque psoriasis. METHODS: Twelve patients with mild to moderate psoriasis were treated with TriAc (0.1% in hydrophilic ointment) and placebo applied twice daily to either of two (or several) bilaterally symmetrical plaques for 8 weeks. The patients and investigator were blinded as to the content of the tubes. Every 2 weeks the treated plaques were evaluated by the patient (using a balanced visual analogue scale for a right-left comparison) and by the investigator (using a psoriasis severity index and a global assessment of each plaque). RESULTS: After 8 weeks of treatment, more than 33% improvement of the psoriasis index occurred in 10 of 12 TriAc-treated and nine of 12 placebo-treated plaques. There were no statistically significant differences between the treatments in terms of reduction of the scores for erythema, scaling, induration or pruritus during the study. Half of the patients considered TriAc superior to placebo, whereas three of 12 were of the opposite opinion (P > 0.05). The global assessment showed marked improvement or remission in six TriAc-treated and five placebo-treated cases (P > 0.05 for difference). No adverse effects were noted. CONCLUSIONS: TriAc in the dosage and formulation studied was safe but no more effective than placebo in treating plaque psoriasis. However, newer thyroid hormone analogues (agonists or antagonists) might be more active and should be further explored in this context.

The expression of cellular retinoid binding proteins in non-APL leukemic cells and its association with retinoid sensitivity.

Retinoic acid (RA) has important effects on cell differentiation and cell growth and on normal embryonic development. Intracellular retinoid signaling induced by endogenous or exogenous RA is regulated by retinoid binding proteins such as CRBPI, CRABPI and CRABPII and there are data suggesting that the expression of these proteins can influence the sensitivity to the growth inhibitory effects of ATRA. In this study, we investigated the basal and ATRA-induced expression of CRBPI and CRABPI and II in leukemic cell lines and in cells from patients with acute myeloid leukemia (AML). CRBPI as well as CRABPI and II were expressed in all tested cell lines and in leukemic cells from all 18 AML-patients. CRABPII mRNA expression was more abundant than CRBPI and CRABPI in both the cell lines and the patient cells but the levels compared the house keeping gene was lower in the patient cells. In all cell lines and in 69% of the patient samples, ATRA did upregulate CRABPII whereas CRBPI exhibited a varying response and CRABPI was more commonly downregulated. The sensitivity to the growth inhibitory effects of ATRA did not correlate with the basal expression of any of these proteins. However, ATRA-induced upregulation of CRABPII did significantly correlate with the ATRA sensitivity (p < 0.005) as well as with ATRA-induced upregulation of the retinoid receptor RARbeta (p < 0.05). We conclude that the retinoid binding proteins CRBPI and CRABPI and II are expressed in myeloid leukemic cells of non-M3 type but that the level of expression does not affect ATRA sensitivity.

Topical retinoic acid alters the expression of cellular retinoic acid-binding protein-I and cellular retinoic acid-binding protein-II in non-lesional but not lesional psoriatic skin.

Therapeutic retinoids have profound effects on psoriatic skin pathology but their interactions with various retinoid-binding proteins in lesional vs non-lesional skin have not been investigated. Using quantitative real-time PCR the mRNA expression of cellular retinol-binding protein I (CRBPI) and retinoic acid-binding protein I/II (CRABPI/CRABPII) was studied in psoriatic and healthy control (=normal) skin after 4 days of occlusive RA/vehicle treatment (n=6). Untreated psoriatic lesions showed a markedly elevated CRABPII/CRABPI ratio, while the CRBPI level was reduced in lesional and non-lesional skin as compared to normal skin. In RA-treated normal and non-lesional skin, the mRNA expression of CRBPI was unaltered while that of CRABPI and CRABPII was reduced by approximately 80% and increased approximately 5-fold, respectively, as compared to vehicle-treated skin. In contrast, lesional skin exposed to RA showed an almost 90% increase in CRBPI transcripts but unaltered expression of CRABPI and CRABPII, yet, the mRNA expression of several inflammatory mediators, e.g. inducible nitric oxide synthase, interferon-gamma and interleukin-1beta, was clearly reduced. Immunohistochemistry localized CRABPII to suprabasal keratinocytes in normal skin and revealed markedly elevated levels in lesional skin. RA treatment induced CRABPII protein expression in normal and non-lesional skin, to similar levels as in untreated lesions. The results indicate that the effects of RA differ in normal/non-lesional psoriatic skin and lesional skin. Whether the high expression of CRABPII in psoriatic skin lesions is due to increased amounts of endogenous retinoids in lesional skin or reflects an abnormal regulation of the CRABPII gene in psoriasis remains to be studied.

Interaction of isotretinoin with endogenous retinoids.

Isotretinoin exhibits antiproliferative and antiandrogenic effects on the sebaceous glands by yet unknown mechanisms. This review focuses on a putative mechanism of action of isotretinoin involving the interaction of isotretinoin with the metabolism of endogenous retinoids and androgens.

Increased retinoid signaling in vascular smooth muscle cells by proinflammatory cytokines.

Retinoids have been shown to modulate inflammation and the immune response in many cell types including macrophages, endothelial cells, and vascular smooth muscle cells. However, present knowledge of whether inflammatory mediators modulate vitamin A status in these cells is limited. To identify the role of inflammation on retinoid metabolism in vascular smooth muscle cells, the cells were exposed to a combination of proinflammatory cytokines: interleukin-1beta, interferon-gamma, and lipopolysaccharides. Without stimulation with proinflammatory cytokines, vascular smooth muscle cells expressed retinol dehydrogenases-2 and 5 mRNA detected by RT-PCR. Stimulation with the combination of cytokines induced a substantial increase of retinol dehydrogenase-5 mRNA. This was associated with increased production of ligands for retinoic acid receptors, when assayed in a retinoic acid receptor-dependent luciferase reporter system. Our results demonstrate that inflammatory mediators activate the retinoid metabolic pathway in vascular smooth muscle cells, which potentially may modulate the inflammatory response in the vascular wall.

Retinoid receptor expression and its correlation to retinoid sensitivity in non-M3 acute myeloid leukemia blast cells.

All-trans-retinoic acid (ATRA) has significantly improved the treatment results in acute promyelocytic leukemia (M3). In non-M3 acute myeloid leukemia (AML), the effects are less clear, and there is a pronounced heterogeneity in the sensitivity to the growth-inhibitory effects of retinoids in leukemic cells from different non-M3 AML patients. Retinoids exert their effects through a number of nuclear receptors [retinoic acid receptors (RARs) and retinoid X receptors (RXRs)]. In this study, we determined the expression of RAR alpha, RAR beta, RAR gamma, and RXR alpha by real-time PCR in four cell lines and in blast cells from patients with non-M3 AML before and after ATRA incubation. All four receptors were expressed in cells from all 18 tested patient samples and in four myeloid cell lines. In the majority of the patient samples as well as in the cell lines, there was a pattern of high expression of RAR alpha and RXR alpha and low expression of RAR beta and RAR gamma. There was no correlation between the basal expression of any of the retinoid receptors and sensitivity to ATRA. A 24-h exposure to ATRA increased the expression of RAR alpha, RAR beta, RAR gamma, and RXR alpha in 46%, 77%, 30%, and 38% of the samples, respectively. The mean increase in receptor expression was most pronounced for RAR beta and RXR alpha. There was a significant correlation between an increase in RAR beta expression in response to ATRA and sensitivity to ATRA (P < 0.014). No such correlations were found for RAR alpha, RAR gamma, and RXR alpha. The expression of the monocytoid marker CD14 was significantly correlated with increased expression of RAR alpha (P = 0.03). We conclude that RAR alpha, RAR beta, RAR gamma, and RXR alpha are expressed in non-M3 AML blast cells and that ATRA-induced expression of RAR beta may be a marker for retinoid sensitivity.

Retinoic acid inhibits nitric oxide synthase-2 expression through the retinoic acid receptor-alpha.

Retinoids are multipotent modulators of cellular functions and suppress cytokine-induced production of nitric oxide (NO) in several cell types. We have explored the mechanisms by which retinoic acid (RA) regulates NO production in rat aortic smooth muscle cells (VSMC), which express NOS2 in response to proinflammatory cytokines. RA inhibited interleukin-1beta (IL-1beta)-induced NOS2 mRNA expression and NO production. These effects were attenuated by the retinoic acid receptor (RAR) antagonist CD3106, indicating that they were mediated through retinoic acid receptors (RARs). The synthetic retinoid agonists CD336 (which specifically binds RARalpha) and CD367 (which binds all RARs) but not agonists specific for RARbeta, RARgamma, or RXRs reduced IL-1beta-induced NOS2 expression and NO production. When transfecting VSMC with a 1570-bp NOS2 promoter fragment fused to a luciferase reporter gene, the NOS2 promoter activity was inhibited by RA. These results indicate that retinoids modulate NO production in VSMC via RARalpha, which inhibits the transcription of the NOS2 gene.

Decreased mRNA levels of retinoic acid receptor alpha, retinoid X receptor alpha and thyroid hormone receptor alpha in lesional psoriatic skin.

Retinoic acid, vitamin D3 and triiodothyronine regulate keratinocyte proliferation and differentiation--processes that are disturbed in psoriatic skin--via binding to nuclear receptors for retinoic acid (RAR-alpha,-gamma), vitamin D3 (VDR), thyroid hormone (TR-alpha,-beta) plus the common heterodimer partners, the 9-cis-retinoic acid receptors (RXR-alpha,-beta). By using a new quantitative real-time polymerase chain reaction assay, the expression of these receptors and three housekeeping genes (cyclophilin, GAPDH and beta-actin) was studied in psoriatic skin. The expression of housekeeping genes was consistently 2.7-4.3 times higher in lesional than in non-lesional skin. When the beta-actin expression was used to normalize the receptor mRNA values, the RARalpha, RXRalpha and TRalpha transcripts were found to be 58-75% lower in lesional vs. non-lesional skin and the RXRalpha:RARgamma ratio was reduced from 3.2 to 1.5. Topical treatment for 4 days with 0.025% all-trans-retinoic acid or calcipotriol under occlusion did not normalize the altered mRNA expression of RARs, RXR and VDR in lesional skin. The results suggest that retinoid and thyroid hormone signalling is abnormal in lesional psoriatic skin, but how this relates to the pathogenesis of the disease is still unclear.

Keratin 4 upregulation by retinoic acid in vivo: a sensitive marker for retinoid bioactivity in human epidermis.

Retinoids affect keratinocyte differentiation and modulate the expression of many epidermal proteins, among them cellular retinoic acid-binding protein II and the family of cytokeratins. The upregulation of the former protein is a well-known phenomenon, whereas the retinoid-induced regulation of epidermal keratin expression is more complex and only partially understood. We studied the effect of topical retinoids on the expression in healthy skin of cellular retinoic acid-binding protein II, tazarotene-induced genes 1 and 2, several epidermal keratins (K1, K2e, and K10), and two mucous keratins (K4 and K13) known to appear in epidermis under certain abnormal conditions. Reverse transcription-polymerase chain reaction experiments showed that the K4 expression was the one most overtly induced by 2 wk of open treatment with 0.05% of retinoic acid and tazarotene. Using real-time quantitative polymerase chain reaction (TaqMan) and normalization of the mRNA values to beta-actin, the increase in K4 was found to be 100-1000-fold. In comparison, the expression of K13 and cellular retinoic acid-binding protein II was increased 10-50-fold, the K1 and K10 mRNA levels remained unchanged, and the K2e level decreased by a factor of 100-1000. In parallel biopsies, immunohistochemistry showed no change in K1, K2e, or K10 staining, but a strong de novo appearance of K4 in the granular layer after retinoid treatment. In a separate study, occlusive application of 0.025% retinoic acid in four healthy subjects produced a maximal K4 mRNA signal after 48 h and strong K4 staining after 80 h. Finally, a dose-response study showed that the de novo appearance of K4 can be utilized as a sensitive test for retinoid bioactivity in epidermis in vivo.

Ultraviolet irradiation depletes cellular retinol and alters the metabolism of retinoic acid in cultured human keratinocytes and melanocytes.

Vitamin A is an intrinsic modulator of proliferation and differentiation in human epidermis, and may be destroyed by ultraviolet radiation (UVR) impinging on the skin. To identify the deleterious effects of a perturbed cellular vitamin A status, we investigated the endogenous retinoid concentrations and the metabolism of [3H]retinol and all-trans [3H]retinoic acid in cultured human keratinocytes and melanocytes exposed to UVR, using high performance liquid chromatography. Before UVR the retinoid content was similar in keratinocytes and melanocytes, but the uptake of [3H]retinol was three-fold higher and the uptake of [3H]retinoic acid was 10-fold higher in the melanocytes. In both cell types, UVR (UVA 360 mJ/cm2 plus UVB 140 mJ/cm2) instantaneously reduced the concentration of retinol by about 50% and that of 3,4-didehydroretinol by about 20%. The retinoid concentrations returned to normal within 1-2 days post-irradiation, despite there being no overt increase in the uptake of [3H]retinol or the biosynthesis of 3,4-didehydroretinol. However, in both types of irradiated cells, the accumulation of the biologically most active metabolite, all-trans [3H]retinoic acid, was about 60% higher than in control cells. Furthermore, the metabolism of authentically supplied [3H]retinoic acid was reduced, especially in irradiated keratinocytes, which probably contributed to the restoration of retinoid levels after UV exposure.

The vitamin A metabolism and expression of retinoid-binding proteins differ in HaCaT cells and normal human keratinocytes.

HaCaT keratinocytes differ from normal human epidermal keratinocytes (HEK) by constitutive expression of differentiation markers which are normally suppressed by vitamin A. In search of an explanation for this discrepancy we compared the vitamin A content, the expression of retinoid-binding proteins, and the vitamin A metabolism in the two cell types. The concentrations of retinol and 3,4-didehydroretinol in cultured HaCaT cells were less than one-fifth those in HEK, and the content of fatty acyl esters was even lower. Similarly, the concentrations of cellular retinol-binding protein and cellular retinoic acid-binding protein (CRBPI and CRABPII, respectively) were 10-30 times lower in HaCaT cells than in HEK corresponding to a reduced mRNA expression of these proteins. Unexpectedly, HaCaT cells expressed RARbeta in addition to RARalpha, RARgamma and RXRalpha, which are nuclear receptors normally found in HEK. Radioactive retinol added to the culture medium appeared only transiently in HaCaT cells, and pulse labeling confirmed a defective cellular retention of retinyl esters. After 24 h of incubation with [3H]retinol, cell-associated radioactivity corresponding to retinol, 3,4-didehydroretinol, all-trans-retinoic acid and 3,4-didehydroretinoic acid was found in both HaCaT cells and HEK. [3H]Retinoic acid showed a more rapid metabolism to 4-hydroxy/4-keto-retinoic acid in HaCaT cells than in HEK, which could be explained by a higher expression of cytochrome p450RAI in the former cells. In conclusion, the abnormal uptake of vitamin A and low levels of retinoid binding proteins in HaCaT cells, linked with an aberrant metabolism of retinol, may help to explain why these cells differentiate also in the presence of retinoids.