Neurosteroid Biosynthesis

The specific interactions of steroids with binding sites at neuronal membranes and the ability of various steroids to modulate the brain function has prompted the investigation of the steroidogenic potential of CNS structures. The pioneering work of Baulieu et al. (65) demonstrated that glial cells can convert cholesterol to PREG and give origin to steroid metabolites, which are potential modulators of neuronal function. It has been shown that oligodendrocytes, a glioma cell line, and Schwann cells have the ability to metabolize cholesterol to PREG, the first step in steroid biosynthesis. However, discrepancies exist among the levels of the enzymatic activity, the amount of immunoreactive protein, and mRNA of the P450scc present in brain. In addition, despite the high levels of DHEA (the first neurosteroid described) found in brain, no one has demonstrated the presence of the 17a-hydroxylase cytochrome P450 activity in brain (P450c17). Thus, it seems that brain steroid synthesis may not fit the well-defined scheme of adrenal, gonadal, and placental steroidogenesis, and that new pathways should be explored. Understanding the mechanisms of PREG and DHEA formation is paramount to all speculation and hypotheses about neurosteroids and their role in brain function. The levels of these steroids and their sulfated and lipoidal forms in brain are distinct from the peripheral steroid levels (66,67), and their function as neuroactive steroids at the GABAa and NMDA receptor level has been well-established (68).

We examined neurosteroid synthesis using the C6-2B subclone of the rat C6 glioma cell line. Figure 1 shows that these cells express both the glial fibrillary acidic protein (GFAP), a general marker for glial cells in culture, and galactocerebroside C (Gal C), a specific cell-surface marker for oligodendrocytes in culture. In agreement with our previous findings, using isolated mitochondria in immunoblot studies (55), C6-2B cells also express P450scc protein, but not the P450c17 protein. We then demonstrated that these cells were able to synthesize PREG from the substrates mevanolactone and cholesterol (69).

In order to examine the activity of the P-450scc present in C6 cells hydroxylated analogs of cholesterol were used, which will freely cross the mitochondrial membranes thus accessing the P-450scc in the inner membrane. Three different hydroxylated cholesterols (25-, 22-, and 20-OH-cholesterol) stimulated mitochondrial production of PREG by three to fivefold. Glial-cell P-450scc activity was also tested with hydroxylated cholesterol analogs in the presence of aminoglutethimide, an inhibitor of adrenal, testis, and ovarian P-450scc. Aminoglutethimide inhibited the PREG formation in a concentration-dependent manner from 25- and 22-OH-cholesterol, but failed to affect the conversion of 20-OH-cholesterol into PREG (55). These findings suggest a functional analogy between adrenal and glial P-450scc. Thus, in addition to the P450scc protein, C6-2B cells also express P450scc activity. More recently, we demonstrated that human glioma cells in culture are also able to synthesize neurosteroids in a similar manner to C6-2B cells, thus validating the use of the C6-2B cell model system. Although C6-2B glial cell P450scc protein levels closely correlated those found in the adrenal gland, the P450scc mRNA was undetectable by Northern-blot and RNase protection assays. P450scc mRNA could be detected only after 35 cycles of PCR. Furthermore, the P450scc enzymatic activity in glial cells was 10-fold less than the activity of the adrenal enzyme. These discrepancies may be owing to differences in transcriptional regulation, mRNA stability, and antibody specificity and affinity. Alternatively, the possibility exists that neurosteroids, all or in part, may be formed from alternative precursors using alternative pathways (70,71).

It is important to note that P450scc activity is related to the oligodendrocyte differentiation process (72). Cholesterol accumulation in brain is also related to differentiation (73) and coincides with the rate of myelinization (74). Interestingly, all three activities reach their maximum in the rat at 20 d of age and cholesterol accumulation in brain declines after maturation of the CNS structures, such as myelin and nerve endings. These findings demonstrate a temporal relationship among cholesterol accumulation, steroid synthesis, myelinization, and nerve-ending formation. The functional consequence(s) of this relationship is under investigation.

Despite the overwhelming evidence that P450scc activity is localized to glial cells of the brain, an earlier study demonstrated the presence of P450scc immunoreactivity in select neuronal populations (75) and more recently P450scc mRNA was detected by the

Fig. 1. Characterization of C6-2B rat tumor glioma cells. Cells were immunostained with antibodies: (B) anti-GFAP(1:160); (C) anti-GALC (1:100); (D) anti-PBR (1:200); (E) anti-P450scc (1:200); and (F) anti-P450c17 (1:200). Normal rabbit serum control is shown in (A). Horseradish-peroxidase conjugated secondary antibody was used for the detection of the anti-serum-antigen complex.

Fig. 1. Characterization of C6-2B rat tumor glioma cells. Cells were immunostained with antibodies: (B) anti-GFAP(1:160); (C) anti-GALC (1:100); (D) anti-PBR (1:200); (E) anti-P450scc (1:200); and (F) anti-P450c17 (1:200). Normal rabbit serum control is shown in (A). Horseradish-peroxidase conjugated secondary antibody was used for the detection of the anti-serum-antigen complex.

polymerase chain reaction (PCR) technique in primary cerebellar granule neurons (76). In addition, using isolated adult rat retina as a model system, we observed that the neuronal ganglion cells express P450scc protein and activity, and thus are able to synthesize steroids (77).

Was this article helpful?

0 0

Post a comment