Faculty Bibliography

Smith-Lemli-Opitz syndrome (SLOS) is attributable to mutations in the gene coding for 7-dehydrocholesterol reductase. Low to absent enzyme activity accounts for the accumulation of both 7-dehydrocholesterol and 8-dehydrocholesterol in plasma and other tissues. Since oxysterols can participate in the regulation of cholesterol homeostasis, we examined the possibility that they are formed from these dehydrocholesterol intermediates. In patients with SLOS, we found serum levels of 27-hydroxy-7-dehydrocholesterol ranging from 0.1 to 0.25micro M and evidence for circulating levels of 27-hydroxy-8-dehydrocholesterol (0.04-0.51 micro M). Picomolar quantities of 27-hydroxy-7-dehydrocholesterol were identified in normal individuals. Biologic activities of 27-hydroxy-7-dehydrocholesterol were found to include inhibition of sterol synthesis and the activation of nuclear receptor LXRalpha but not that of LXRbeta. These activities occurred at concentrations found in plasma and presumably at those existing in tissues. Thus, patients with SLOS have increased levels of metabolites derived from intermediates in cholesterol synthesis that are biologically active and may contribute to the regulation of cholesterol synthesis in vivo

The stroma of the human postmenopausal ovary is postulated to produce androgens, but evidence for and against this idea exits in the literature. The purpose of this study was to determine whether key steroidogenic enzymes involved in androgen synthesis are expressed in the postmenopausal ovarian stroma. Stromal cells were isolated from postmenopausal ovaries and expression for genes involved in steroidogenesis [steroidogenic acute regulatory protein (StAR), P450scc, 3 beta-hydroxysteroid dehydrogenase (3 beta-HSD) P450c17, and P450c27] as well as for several growth factor binding proteins [gremlin, IGF binding protein-4, follistatin, and secreted frizzled-related protein (sFRP)-1 and -4], were compared with cultured human theca cells and dermal fibroblasts. Production of steroids (pregnenolone, progesterone, and hydroxysterol metabolites) and the metabolism of [3H] pregnenolone by ovarian stromal cells were also assessed. Isolated ovarian stromal cells from different subjects had a uniform morphology within and across cultures. Quantitative real time RT-PCR revealed that StAR, P450scc, and 3 beta-HSD transcripts were, respectively 30, 25, and 45 times more abundant in theca cells than in stromal cells. Mean levels of P450scc and 3 beta-HSD transcripts in stromal cells were similar to those found in dermal fibroblasts, whereas StAR transcripts in stromal cells were 285-fold more abundant than in fibroblasts. There was no significant expression of P450c17 in ovarian stromal cells or fibroblasts ( approximately 2000-fold less than in theca cells). Western analysis demonstrated the presence of the 30-kDa StAR mature protein in the cultured stromal cells, whereas P450c17 protein was not detectable. Ovarian stromal cells did not metabolize [3H] pregnenolone in the presence or absence of 8-Br-cAMP. Furthermore, pregnenolone and progesterone secretion by stromal cells was also undetectable, even in the presence of 22-hydroxycholesterol. P450c27 protein was detected in ovarian stromal cells and its metabolic products (i.e. 27-hydroxycholesterol and cholestenoic acid) were found in the culture media, reflecting functional cholesterol 27-hydroxylase activity. Follistatin, gremlin, IGF binding protein-4, and sFRP-1 and -4 transcripts were detected in the stromal cells in relative amounts significantly higher than theca cells, but not significantly different from fibroblasts, except for sFRP-1, which was significantly higher in stromal cells. Our observations demonstrate that stromal cells of the postmenopausal ovary have a signature biochemical and molecular phenotype that can be distinguished from fibroblasts. These cells do not appear to have significant steroidogenic potential in vitro, but they do metabolize cholesterol into hydroxysterols. We conclude that the predominant stromal cells of the postmenopausal ovary are not a significant site of androgen biosynthesis

As a result of an alternative exon 1, the gene for human hydroxysteroid sulfotransferase (SULTB1) encodes for two peptides differing only at their amino termini. The SULT2B1b isoform preferentially sulfonates cholesterol. Conversely, the SULT2B1a isoform avidly sulfonates pregnenolone but not cholesterol. The outstanding structural feature that distinguishes the SULT2B1 isoforms from the prototypical SULT2A1 isozyme is the presence of extended amino- and carboxyl-terminal ends in the former. Investigating the functional significance of this unique characteristic reveals that removal of 53 amino acids from the relatively long carboxyl-terminal end that is common to both SULT2B1 isoforms has no effect on the catalytic activity of either isoform. On the other hand, removal of 23 amino acids from the amino-terminal end that is unique to SULT2B1b results in loss of cholesterol sulfotransferase activity, whereas removal of 8 amino acids from the amino-terminal end that is unique to SULT2B1a has no effect on pregnenolone sulfotransferase activity. Deletion analysis along with site-directed mutagenesis of SULT2B1b reveal that the amino acid segment 19-23 residues from the amino terminus and particularly isoleucines at positions 21 and 23 are crucial for cholesterol catalysis. In the gene for SULT2B1, exon 1B encodes for only the unique amino-terminal region of SULT2B1b; however, exon 1A encodes for the unique amino-terminal end of SULT2B1a plus an additional 48 amino acids. Thus, if the gene for SULT2B1 employs exon 1B, cholesterol sulfotransferase is synthesized, whereas if exon 1A is used, pregnenolone sulfotransferase is produced

The CYP27 gene is expressed in arterial endothelium, macrophages, and other tissues. The gene product generates sterol intermediates that function as ligands for nuclear receptors prior to their transport to the liver for metabolism, mostly to bile acids. Most attention has been given to 27-hydroxycholesterol as a ligand for LXR activated receptors and to chenodeoxycholic acid as a ligand for farnesoid X activated receptors (FXRs). Expression of the pathway in macrophages is essential for normal reverse cholesterol transport. Thus, ABC transporter activity is upregulated, which enhances cholesterol efflux. Absence of these mechanisms probably accounts for the accelerated atherosclerosis that occurs in cerebrotendinous xanthomatosis. Accumulation of 27-hydroxycholesterol in human atheroma is puzzling and may reflect low levels of oxysterol 7alpha-hydroxylase activity in human macrophages. The same enzyme determines the proportion of mono-, di-, and tri-hydroxy bile acids synthesized in the liver. Oxysterol 7alpha-hydroxylase deficiency is a molecular basis for cholestatic liver disease. Chenodeoxycholic acid, the major normal end product, downregulates expression of cholesterol 7alpha-hydroxylase via the FXR/short heterodimer protein nuclear receptor and thus limits total bile acid production. The challenge is to quantify in a physiologic setting the magnitude of the pathway in different tissues and to further evaluate the biologic roles of all the intermediates that may function as ligands for orphan nuclear receptors or via other regulatory mechanisms

Although a variety of oxidation products of cholesterol occur in vitro, enzyme-catalyzed oxidations can occur at only 5 sites on the cholesterol molecule: C7alpha, C22R, C24S, C25, and C27. The genes coding for the synthesis of these enzymes were cloned, the tissue expressions of the mRNAs were identified, and the enzymes were characterized. The biologic properties of the hydroxycholesterol molecules that are initially generated and their metabolites are under study. Downregulation of cholesterol synthesis via the SREBP/SCAP regulatory pathway is common to the initial hydroxycholesterols, but more variations exist with respect to these intermediates functioning as ligands for the nuclear receptor LXRalpha. Because this receptor regulates the expression of cholesterol 7alpha-hydroxylase and ABC transporter proteins, hydroxycholesterols and their intermediate steroid metabolites modulate a number of biologic processes. Metabolism of 22S-hydroxycholesterol to steroid hormones differs from that of the other hydroxycholesterols which form mostly steroid acidic products, otherwise known as bile acids. In vivo estimates of their production rates in intact humans indicate that 24S and 25-hydroxycholesterol account for no more than 7% of total bile acid production per day. Current evidence indicates that cholesterol 7alpha-hydroxycholesterol generated in the liver is the major source of bile acids in older adults. It is also known that the cholesterol 27-hydroxylation pathway is the only one expressed in fetal and neonatal life. Precisely when the proportions contributed by these two metabolic pathways to bile acid synthesis begin to shift and the role of the cholesterol 27-hydroxylase pathway in reverse cholesterol transport mandate further study

Despite the fact that most human subjects synthesize about twice as much cholic acid as chenodeoxycholic acid, available evidence suggests that 7 alpha-hydroxycholesterol, the first intermediate in the major pathway for bile acid synthesis, is converted about equally to these two bile acids. Synthesis through the main alternate pathway can not explain this discrepancy because 27-hydroxycholesterol, the first intermediate in that pathway, is converted preferentially to chenodeoxycholic acid. To examine the validity of these contradictory observations, we administered (24-(14)C)-cholic acid and (24-(14)C)-chenodeoxycholic acid together with (7 beta-(3)H)-7 alpha-hydroxycholesterol on one occasion and (22,23-(3)H)-27-hydroxycholesterol on a separate occasion to eight normal human subjects. Synthesis of the two primary bile acids was determined by means of standard isotope dilution kinetics of the carbon 14-specific activities of biliary bile acids. Conversion of (7 beta-(3)H)-7 alpha-hydroxycholesterol and (22,23-(3)H)-27-hydroxycholesterol to bile acid was calculated from the tritium/carbon 14 ratio in cholic and chenodeoxycholic acid. For synthesis, the mean +/- SEM cholic/chenodeoxycholic ratio was 1.82 +/- 0.26. For apparent conversion of (7 beta-(3)H)-7 alpha-hydroxycholesterol to bile acid, the mean +/- SEM cholic/ chenodeoxycholic ratio was 1.02 +/- 0.09, whereas for (22,23(3)H)-27-hydroxycholesterol, the mean +/- SEM cholic/chenodeoxycholic ratio was 0.38 +/- 0.03. These data imply that, on average, more than 40% of cholic acid in these subjects was synthesized through a pathway that bypassed initial 7 alpha-hydroxylation. However, consideration of all potential candidates for such a pathway raises doubts that any of them contributes substantially to bile acid synthesis

The enzyme cholesterol 27-hydroxylase, expressed by arterial endothelium and monocytes/macrophages, is one of the first lines of defense against the development of atherosclerosis. By catalyzing the hydroxylation of cholesterol to 27-hydroxycholesterol, which is more soluble in aqueous medium, the enzyme promotes the removal of cholesterol from the arterial wall. Prior studies have suggested that immune reactants play a role in the pathogenesis of atherosclerosis; we report here that immune reactants, IFN-gamma and immune complexes bound to C1q, but not interleukin-1 and tumor necrosis factor, diminish the expression of cholesterol 27-hydroxylase in human aortic endothelial cells, peripheral blood mononuclear cells, monocyte-derived macrophages, and the human monocytoid cell line THP-1. In addition, our studies demonstrate that immune complexes down-regulate cholesterol 27-hydroxylase only after complement fixation via interaction with the 126-kD C1qRp protein on endothelial cells and THP-1 cells. These results are consistent with the prior demonstration that IFN-gamma contributes to the pathogenesis of atherosclerosis and suggest a role for C1q receptors in the atherogenic process. Moreover, these observations suggest that one mechanism by which immune reactants contribute to the development of atherosclerosis is by down-regulating the expression of the enzymes required to maintain cholesterol homeostasis in the arterial wall

In humans, the biotransformation of cholesterol and its hydroxylated metabolites (oxysterols) by sulfonation is a fundamental process of great importance. Nevertheless, the sulfotransferase enzyme(s) that carries out this function has never been clearly identified. Cholesterol is a relatively poor substrate for the previously cloned hydroxysteroid sulfotransferase (HST), i.e. dehydroepiandrosterone (DHEA) sulfotransferase (HST1). Recently, cloning of a single human gene that encodes for two proteins related to HST1 was reported. These newly cloned sulfotransferases (HST2a and HST2b), while exhibiting sequence similarity to other members of the soluble sulfotransferase superfamily, also contain unique structural features. This latter aspect prompted an examination of their substrate specificity for comparison with HST1. Thus, HST1, HST2a, and HST2b were overexpressed as fusion proteins and purified. Furthermore, a novel procedure for the isolation of cholesterol and oxysterol sulfonates was developed that was used in association with HPLC to resolve specific sterol sulfonates. HST1 preferentially sulfonated DHEA and, to a lesser extent, oxysterols; whereas cholesterol was a negligible substrate. The reverse, however, was the case for the HST2 isoforms, particularly HST2b, which preferentially sulfonated cholesterol and oxysterols, in contrast to DHEA, which served as a poor substrate for this enzyme. RT-PCR analysis revealed distinct patterns of HST1, HST2a, and HST2b expression. It was particularly notable that both HST2 isoforms, but not HST1, were expressed in skin, a tissue where cholesterol sulfonation plays an important role in normal development of the skin barrier. In conclusion, substrate specificity and tissue distribution studies strongly suggest that HST2a and HST2b, in contrast to HST1, represent normal human cholesterol and oxysterol sulfotransferases. Furthermore, this study represents the first example of the sulfonation of oxysterols by a specific human HST

Expression of the gene coding for the synthesis of 25(R), 26-hydroxycholesterol in many tissues and the finding that this sterol can be the sole pathway for the production of bile acids have led to a renewed interest in this metabolic pathway. A further impetus for exploring the normal biologic roles that are served by expression of the CYP27A1 gene is the knowledge that mutations in humans are associated with accelerated atherosclerosis and with severe neurologic impairment. The molecular mechanisms governing these phenotypic expressions are not known but in light of the traditional role of steroids as ligands for receptors that regulate gene expression it seems likely that the intermediates in this pathway modulate a number of enzymatic activities that remain to be elucidated