The presence of 5 alpha-reductase (5 alpha-R) in skin may indicate that the androgen regulation of sebaceous glands and sebum production requires the local conversion of testosterone to dihydrotestosterone. The goals of this study were to identify which isozyme of 5 alpha-R (type 1 or type 2) is expressed in sebaceous glands from facial areas, scalp, and non-acne-prone areas; to determine if 5 alpha-R activity is concentrated in sebaceous glands; to assess whether there are regional differences in this enzyme's activity; and to test the effects of azasteroid inhibitors and 13-cis retinoic acid on 5 alpha-R in these tissues. Sebaceous glands were microdissected from facial skin, scalp, and non-acne-prone skin (arm, breast, abdomen, leg), and the activity of 5 alpha-R was determined. A total of 49 samples from 23 male and 21 female subjects without acne (age range, 16 to 81 years, 56 +/- 20 years [mean +/- SD]) was analyzed. The biochemical properties of the enzyme in each of the samples tested are consistent with those of the type 1 5 alpha-R. Minimal to no type 2 5 alpha-R was detected. The level of 5 alpha-R activity was significantly higher in the sebaceous glands compared to whole skin in facial skin (p = 0.047), scalp (p = 0.039), and non-acne-prone skin (p = 0.04). Enzyme activity in sebaceous glands from facial skin and scalp was significantly higher than in a comparable amount of sebaceous gland material obtained from non-acne-prone areas (32 +/- 6 [mean +/- SEM]), 35 +/- 7 (mean +/- SEM) versus 6.0 +/- 3.0 (mean +/- SEM) pmol/min/mg protein, p = 0.014 and 0.007, respectively). Finasteride and 13-cis retinoic acid were poor inhibitors of the enzyme with 50% inhibitory concentration values greater than 500 nM. These data demonstrate that in the skin from older patients without acne the type 1 isozyme of 5 alpha-R predominates, its activity is concentrated in sebaceous glands and is significantly higher in sebaceous glands from the face and scalp compared to non-acne-prone areas, and the action of 13-cis retinoic acid in the control of acne is not at the level of 5 alpha-R. Furthermore, we suggest that specific inhibition of the type 1 5 alpha-R may offer a viable approach to the management of sebum production and, hence, acne.

Steroid Sa-reductase (EC 1.3.1.22) catalyzes the reduction of testosterone to dihydrotestosterone. The Sareductase found in human scalp has been compared with the enzyme found in prostate. The scalp reductase has a broad pH optimum centered at pH 7.0. This is distinctiy different from the pH optimum of 5.5 observed with the prostatic form of the enzyme. These enzymes also differ in the K.. for testosterone, which is 25-fold higher for the scalp reductase. The most significant difference between the two enzymes is their affinity for inhibitors. Two 4-azasteroids and a 3-carboxyandrostadiene are potent inhibitors of the prostatic reductase but are weak inhibitors of the scalp reductase. In contrast, several N-4-methylazasteroids are good inhibitors of the scalp reductase. These findings support a proposal that different isozymes of 5a-reductase may exist in scalp and prostate. The scalp reductase was also compared to 5a-reductase 1, one of the two enzymes recently cloned from human prostate [Andersson, S.

As an initial step toward understanding the transcriptional regulation of cholesterol 7 alpha-hydroxylase (CYP7) in man, we isolated and functionally characterized the 5'-flanking region of the human CYP7 gene. The nucleotide sequences of the first exon and 1.6 kb preceding the exon were determined and found to contain a TATA box at position -30, a modified CAAT box at position -92, three potential hepatocyte nuclear factor 3 (HNF-3) recognition sites at nucleotides -316, -288, and -255, respectively, and a modified sterol response element at position -271. DNA sequences containing 1.3 kb of the 5'-flanking region and 29 nucleotides of the first exon were linked to the chloramphenicol acetyltransferase gene and transiently transfected into several cell lines. Promoter activity was very strong in the human hepatoma cell line HepG2 but absent in cells of nonhepatic origin. Mutational analysis of the promoter identified several regions that function in the transcriptional regulation of CYP7. Introduction of a fragment containing the region from -432 to -220 upstream of a heterologous promoter, in either orientation, resulted in a tremendous stimulation of activity in HepG2 cells. DNase I footprint analysis identified three regions within this fragment which were protected from digestion. The overexpression of HNF-3 in HepG2 cells resulted in a 4-fold stimulation of CYP7 transcriptional activity. We suggest that the region between -432 and -220 functions as a cell-specific enhancer whose activity is controlled, in part, by HNF-3.

No abstract

Cellular processes responsible for maintaining cholesterol homoeostasis are highly regulated. To determine whether two of these processes, cholesterol biosynthesis and receptor-mediated uptake of low-density lipoprotein (LDL), are co-ordinately regulated in human liver, we employed a human hepatoma cell line (HepG2) and measured the accumulation of mRNA for LDL receptor, 3-hydroxy-3-methylglutaryl-CoA (HMG-CoA) reductase and HMG-CoA synthase under a variety of conditions. Genomic Southern-blot analysis demonstrated that the integrity of these genes is maintained in the transformed cell. Treatment of HepG2 cells with mevalonate, 25-hydroxycholesterol, LDL, lovastatin or miconazole resulted in a similar effect on the accumulation of all three mRNAs at the concentrations tested. The onset of the response to drug, whether repression or induction of mRNA accumulation, occurred after approximately the same period of exposure for each mRNA. We conclude that the expression of the LDL receptor, HMG-CoA reductase and HMG-CoA synthase is co-ordinately regulated in HepG2 cells.