Retinoids and state of differentiation modulate CRABP II gene expression in a skin equivalent.

Cellular retinoic acid-binding proteins (CRABPs) are a family of proteins that specifically bind retinoic acid (RA) and have been implicated in mediating its action, although their exact function is still unknown. Two CRABPs have been identified and cloned. CRABP I is present in many tissues and cultured cells; CRABP II, first detected in embryonic and neonatal skin of rats and chicks, is now recognized as the predominant form in human epidermis. Using a human living skin equivalent model composed of a dermis and an epidermis and human cDNAs recently cloned in our laboratory, we have studied the effects of 10(-6) M RA and etretin (ET) on the expression of CRABPs under different culture conditions intended to favor greater or lesser degrees of epidermal differentiation. Total cellular RNA was isolated separately from the dermis and epidermis and processed for northern blot analysis. At a presumptive physiologic RA concentration, only the gene for CRABP II, and not for CRABP I, was expressed. CRABP II transcripts were far more abundant on a per cell basis in epidermal keratinocytes than in dermal fibroblasts under all conditions studied. Epidermal differentiation, stimulated by air exposure of the cultures, tended to enhance CRABP II expression, and treatment with presumptive therapeutic concentrations of the two retinoid compounds tended to decrease CRABP II expression. Opposite effects of air exposure and retinoid treatment were observed on steady state levels of mRNA for selected markers of epidermal differentiation: involucrin, transglutaminase, and spr I. These results are consistent with earlier work at the protein level examining the effect of retinoids on CRABP activity and state of differentiation both in vivo and in vitro. Thus, the skin equivalent appears to be an excellent model system for investigating the role of CRABPs in mediating retinoid effects at the cellular and molecular levels.

Human dermal fibroblasts modulate the effects of retinoids on epidermal growth.

We report the pharmacologic effects of retinoids in a human skin-equivalent model. This sophisticated culture system is composed, as in vivo, of a dermis and epidermis, and provides a unique in vitro system for studying dermal-epidermal interactions and thus, whether normal dermal fibroblasts influence the effects of retinoids on epidermal growth. Epidermalization was initiated on collagen substrates in which fibroblasts were either viable or lysed by osmotic shock. Retinoic acid, isotretinoin, and acitretin at 10(-6) M or 10(-7) M were added to the cultures just after epidermalization, then every two days. Epidermal growth was determined after 2 weeks in terms of the surface area, DNA content, and tritiated thymidine incorporation during the last 24 h of culture. In the absence of viable fibroblasts, retinoic acid and isotretinoin increased epidermal growth, whereas etretin inhibited it. In contrast, in the presence of viable fibroblasts, retinoic acid and isotretinoin inhibited epidermal growth, whereas etretin had no effect. Thus, retinoic acid and isotretinoin had a similar effect on keratinocyte proliferation that contrasted with that of etretin. Taken as a whole, these results show that fibroblasts, present within a collagen substrate, can modulate the pharmacologic effects of retinoids on epidermal growth.

Induction of inosine monophosphate dehydrogenase activity after long-term treatment with mycophenolate mofetil.

OBJECTIVES: To study the pharmacologic activity of mycophenolate mofetil in stable kidney transplant recipients receiving mycophenolate mofetil therapy for different periods from 2 months to 3 years. METHODS: Seventeen patients were enrolled during a 9-month period. Blood samples were collected before administration and 1/2, 1, 2, 4, and 6 hours after administration. Mycophenolic acid, the active metabolite of mycophenolate mofetil, was measured in plasma with use of the enzyme multiplication immunoassay technique (EMIT) assay and the activity of inosine monophosphate dehydrogenase (IMPDH), a key enzyme in the de novo biosynthesis of purines inhibited by mycophenolate mofetil, was determined in whole blood by the estimation of the 3H-release from [2,8-3H]-hypoxanthine. RESULTS: As the length of mycophenolate mofetil therapy increased, the inhibitory effect of mycophenolate mofetil on IMPDH activity was reduced (0.13+/-0.03 for long-term treatment versus 0.46+/-0.06 for short-term treatment; P = .0006) and a stimulatory effect of mycophenolate mofetil on IMPDH activity was observed (1.16+/-0.56 for long-term treatment, versus 0.03+/-0.01 for short-term treatment; P < .0001). These modifications of IMPDH activity were associated with fluctuations in the pharmacokinetics of mycophenolic acid. CONCLUSION: Long-term treatment with mycophenolate mofetil was associated with an induction of IMPDH activity. Such induction could be deleterious if it is followed by a restoration or a stimulation of the proliferative capacity of lymphocytes. These results strongly suggest that the place of mycophenolate mofetil in long-term treatment of kidney transplant patients needs to be carefully assessed.

Calcineurin activity as a functional index of immunosuppression after allogeneic stem-cell transplantation.

BACKGROUND: The authors have previously shown that mononuclear cells derived from patients with resistant chronic graft-versus-host-disease (GVHD) express high calcineurin (CN) activity, suggesting that in vitro assessment of CN activity may be a useful index to estimate the degree of immunosuppression afforded by cyclosporine A (CsA). The goal of this study was to assess CN activity during the first 2 months after allogeneic stem-cell transplantation (SCT) and to correlate its evolution with the occurrence of acute GVHD. METHODS: Thirty-one allogeneic SCT recipients were enrolled during a 21-month period. All received GVHD prophylaxis with CsA (2 mg/kg/day) and methotrexate (on days 1, 3, and 6). CN activity was measured before transplant, and then once weekly, for at least 2 months. RESULTS: Eighteen patients developed acute grade II or higher GVHD at a median time of 22.5 days and were treated with steroids. CN activity was significantly increased in these 18 patients when compared with 13 patients who did not develop GVHD. Analysis involving the receiver operating characteristic curve demonstrated that acute grade II or higher GVHD can be predicted with a sensitivity of 89% and a specificity of 54% with the use of a cutoff value of 28 pmol RII/mg proteins/min of CN activity. CONCLUSIONS: CN activity appears to be a promising therapeutic test to predict acute GVHD after allogeneic SCT. This functional assessment of the in vivo efficacy of CsA opens new insights for CsA dose adjustment-in particular, the administration of its most efficient dose instead of its maximal tolerated dose, as is currently performed.

Expression of inosine monophosphate dehydrogenase type I and type II after mycophenolate mofetil treatment: a 2-year follow-up in kidney transplantation.

The objective of the study was to evaluate the effect of mycophenolate mofetil (MMF) on the regulation of inosine monophosphate dehydrogenase (IMPDH) during the first 2 years after renal transplantation. Twelve patients were enrolled, and 10-h time-course evaluations of the effects of MMF were regularly performed during the study. IMPDH activity and gene expression were measured in whole blood and in mononuclear cells, respectively. Type I IMPDH (IMPDH-I) mRNA was increased during the first 3 months following transplantation and reached its maximal level during acute rejection episodes, whereas type II IMPDH mRNA was stable. Furthermore, although no alteration in the predose samples was observed, patients with prolonged MMF treatment exhibited an increase in the induction potency of both IMPDH activity and gene expression. In vitro experiments confirmed that IMPDH-I is inducible, but preferentially in monocytes than in lymphocytes. This finding suggests that the measurement of IMPDH mRNAs may provide reliable information to predict acute rejection.

Characterization of human cellular retinoic acid-binding proteins-I and -II: ligand binding affinities and distribution in skin.

Cellular retinoic acid-binding proteins (CRABPs) are a family of proteins that specifically bind retinoic acid (RA) and have been implicated in mediating its action, although their exact function is still unknown. Two CRABPs have been identified and cloned. CRABP-I is present in many tissues and cultured cells; and CRABP-II, first detected in embryonic and neonatal skin of rats and chicks, is now recognized as the predominant form in human epidermis. Previous studies of CRABP protein expression and function could not distinguish between the two forms and perhaps for that reason have yielded conflicting results, particularly with regard to RA-binding affinity in human tissues. In the present study, we have used the FLAG technology to generate recombinant CRABP-II and developed an anion-exchange HPLC assay in order to allow an accurate discrimination of the two proteins. CRABP-II eluted first with a retention time of 6 min, and CRABP-I with a retention time of 14 min. Both CRABP-II and CRABP-I were found to be expressed in human skin, CRABP-II by fibroblasts and keratinocytes and CRABP-I by as yet unidentified cells. This divergent origin supports the hypothesis that CRABP-II and CRABP-I differentially mediate RA effects. Binding studies demonstrated that CRABP-I and CRABP-II possess two classes of RA-binding sites: one class of high-affinity binding sites with a constant of dissociation (Kd) of 1.5 nM for CRABP-I and 4.7 nM for CRABP-II and one class of low-affinity binding sites with a Kd of 69 nM for CRABP-I and 101 nM for CRABP-II. These data further elucidate the complex regulation of retinoid effects in human skin.

Keratinocytes and dermal factors activate CRABP-I in melanocytes.

Recognition that cellular retinoic acid binding protein (CRABP)-I and CRABP-II are found in different cell types has provided additional support for the presumably divergent roles of these two proteins in mediating retinoic acid (RA) effects in human skin. CRABP-II is expressed in fibroblasts and keratinocytes, and CRABP-I in as yet unidentified cells, possibly epidermal melanocytes. Recently, we demonstrated that each of these RA-binding proteins in human skin possesses two classes of binding sites, possibly related to the state of phosphorylation of the proteins. We now characterize the cutaneous origin of CRABP-I further using an anion-exchange HPLC assay that allows effective separation of the two proteins in human skin, and a fluorescent in situ hybridization technique. We report that CRABP-I is expressed in isolated melanocytes at the mRNA level, although under these circumstances the protein has minimal RA-binding activity, and that keratinocytic and dermal influences are required for CRABP-I activity in melanocytes. This melanocyte origin for CRABP-I and the improvement by RA of the irregular hyperpigmentation associated with photoaging led us to examine the effects of RA using various cellular associations, from conventional pure cultures of melanocytes grown on plastic dishes to a pigmented skin equivalent consisting of melanocytes and keratinocytes grown on a dermal equivalent. We established that the inhibitory effects of RA on melanogenesis do not result from a direct effect on melanocytes alone but also involve keratinocytes and dermal influence. These data expand our understanding of cell-to-cell signaling in cutaneous pigmentation, and strongly suggest a role for CRABP-I in mediating RA effects on melanogenesis.

Compartmentalised inducible nitric-oxide synthase activity in septic shock.

BACKGROUND: Previous experimental studies support a role for inducible nitric-oxide synthase (iNOS) in the pathogenesis of severe sepsis. The aim of the study was to characterise iNOS activity in different tissues in patients with septic shock. METHODS: 13 consecutive patients with septic shock caused by cellulitis were enrolled. Skin, muscle, fat, and artery samples were obtained from normal, inflamed, and putrescent areas to measure iNOS activity, and concentrations of tumour necrosis factor alpha (TNFalpha) and interleukin 1beta (IL-1beta). In two patients, iNOS activity was also assessed in peripheral blood mononuclear cells (PBMC) incubated with microorganisms causing the sepsis, or in macrophages isolated from suppurating peritoneal fluid incubated with IL-1beta. FINDINGS: Compared with normal and inflamed areas, iNOS activity was increased in putrescent areas for muscle (71-fold [95% CI 20-259] vs normal areas, 69-fold [19-246] vs inflamed areas; p<0.01 for each) and for fat (68-fold [23-199] and 49-fold [18-137], respectively; p<0.01), but not for skin. Compared with normal areas, putrescent areas of arteries showed increased iNOS expression (1280-fold [598-3153]; p<0.01). Compared with normal areas, TNFalpha and IL-1beta were increased in putrescent areas of arteries (223-fold and 41-fold, respectively; p<0.01 for each). PBMCs and tissue macrophages expressed iNOS. Plasma nitrite/nitrate concentrations inversely correlated with mean arterial pressure and systemic vascular resistance. INTERPRETATION: In human septic shock we found that iNOS activity is compartmentalised at the very site of infection and parallels expression of TNFalpha and IL-1beta. PBMCs and tissue macrophages can be a cellular source for iNOS.

Adenovirus-mediated transfer of the atrial natriuretic peptide gene in rat pulmonary vascular smooth muscle cells leads to apoptosis.

Atrial natriuretic peptide (ANP) exhibits relaxant and growth-inhibiting effects on vascular smooth muscle cells (VSMCs). To obtain ANP gene expression in VSMCs, we built a recombinant adenovirus containing the ANP cDNA controlled by the adenovirus major late promotor (AdMLP-ANP). After pulmonary VSMC treatment with AdMLP-ANP at a multiplicity of infection ranging from 5 to 100 TCID(50)/cell, immunoreactive ANP was detectable in the cell culture medium at a level that reached 101 +/- 27 pmol/well after 2 days. The newly expressed ANP was biologically active, as evidenced by its ability to induce cyclic guanosine monophosphate accumulation in target cells and to mimic the effect of exogenous ANP (10(-8) to 10(-7) mol/L). Cell growth and survival of AdMLP-ANP-infected cells were decreased and were associated with the promotion of VSMC apoptosis. These effects, which occurred at a multiplicity of infection of 10 to 100 TCID(50)/cell, were observed neither in cells infected with the control adenoviral constructs (AdMLP-betaGAL and AdMLP-gD) nor in cells treated with exogenous ANP (10(-7) to 10(-6) mol/L). These results showing VSMC apoptosis in response to ANP gene expression may have important implications for the prevention of vascular remodeling by gene therapy.