START domain proteins are conserved α/β helix-grip fold that play a role in the non-vesicular and intracellular transport of lipids and sterols. The mechanism and conformational changes permitting the entry of the ligand into their buried binding sites is not well understood. Moreover, their functions and the identification of cognate ligands is still an active area of research. Here, we report the solution structure of STARD6 and the characterization of its backbone dynamics on multiple time-scales through (15)N spin-relaxation and amide exchange studies. We reveal for the first time the presence of concerted fluctuations in the Ω1 loop and the C-terminal helix on the microsecond-millisecond time-scale that allows for the opening of the binding site and ligand entry. We also report that STARD6 binds specifically testosterone. Our work represents a milestone for the study of ligand binding mechanism by other START domains and the elucidation of the biological function of STARD6.

Steroidogenic acute regulatory protein (StAR)-related lipid transfer (START) domain proteins display diverse expression patterns and cellular localisations. They bind a large variety of lipids and sterols and are involved in lipid metabolism, lipid transfer and cell signalling. The START domain tertiary structure is an α-helix/β-grip fold module of approximately 210 amino acids delimiting an internal cavity forming the binding site. However, the determinants that dictate ligand specificity and the mechanism of ligand entry and exit are ill-defined. Herein, we review and discuss the current knowledge on ligand specificity and binding mechanism of START domains. More specifically, we highlight that the conserved residues of STARD1, STARD3, STARD4, STARD5 and STARD6 START domains binding sterol play an important structural role for the global protein fold, whereas the residues forming the cavity that fits the shape of their respective ligand are divergent, suggesting their participation in ligand specificity. We also explore the potential binding of steroids to STARD6 in the context of ligand selectivity.

Steroidogenic acute regulatory (StAR)—related lipid transfer proteins possess a START (steroidogenic acute regulatory-related lipid transfer) domain. START domains are conserved protein modules involved in the non-vesicular intracellular transport of lipids and cholesterol in mammals. Fifteen mammalian proteins, divided in five subfamilies, are reported to possess a START domain. Members of the STARD4 subfamily, i.e. STARD4, 5 and 6 are essentially single START domains and are thought to be involved in the intracellular transport of cholesterol. No structure of a cholesterol-bound START domain from this family has been resolved yet. The determination of the structure of such a complex would contribute to a better understanding of the mechanism of ligand binding and transport by START domains, two unresolved aspects of their structural biology. In this context, we have undertaken the structure determination of a ligand-bound form of STARD5 by NMR. Here, we report the 1 H, 13 C and 15 N backbone resonance assignments of the ligand-free STARD5.

We present herein a review of our recent results on the characterization of the binding sites of STARD1, STARD5 and STARD6 using NMR and other biophysical techniques. Whereas STARD1 and STARD6 bind cholesterol, no cholesterol binding could be detected for STARD5. However, titration of STARD5 with cholic acid and chenodeoxycholic acid led to specific binding. Using perturbation of the (1)H-(15)N-HSQC spectra and the sequence specific NMR assignments, we identified the amino acids in contact with those ligands. The most perturbed residues in presence of ligands are lining the internal cavity of the protein. Interestingly, these residues are not conserved in STARD1 and STARD6 and could therefore be key structural determinants of the specificity of START domains toward their ligands. We highlight three tissues expressing STARD5 that are affected by bile acids.

STARD5 is a member of the STARD4 sub-family of START domain containing proteins specialized in the non-vesicular transport of lipids and sterols. We recently reported that STARD5 binds primary bile acids. Herein, we report on the biophysical and structural characterization of the binding of secondary and conjugated bile acids by STARD5 at physiological concentrations. We found that the absence of the 7α-OH group and its epimerization increase the affinity of secondary bile acids for STARD5. According to NMR titration and molecular modeling, the affinity depends mainly on the number and positions of the steroid ring hydroxyl groups and to a lesser extent on the presence or type of bile acid side-chain conjugation. Primary and secondary bile acids have different binding modes and display different positioning within the STARD5 binding pocket. The relative STARD5 affinity for the different bile acids studied is: DCA > LCA > CDCA > GDCA > TDCA > CA > UDCA. TCA and GCA do not bind significantly to STARD5. The impact of the ligand chemical structure on the thermodynamics of binding is discussed. The discovery of these new ligands suggests that STARD5 is involved in the cellular response elicited by bile acids and offers many entry points to decipher its physiological role.

Steroidogenic acute regulatory-related lipid transfer (START) domain proteins are involved in the nonvesicular intracellular transport of lipids and sterols. The STARD1 (STARD1 and STARD3) and STARD4 subfamilies (STARD4–6) have an internal cavity large enough to accommodate sterols. To provide a deeper understanding on the structural biology of this domain, the binding of sterols to STARD5, a member of the STARD4 subfamily, was monitored. The SAR by NMR [1H-15N heteronuclear single-quantum coherence (HSQC)] approach, complemented by circular dichroism (CD) and isothermal titration calorimetry (ITC), was used. Titration of STARD5 with cholic (CA) and chenodeoxycholic acid (CDCA), ligands of the farnesoid X receptor (FXR), leads to drastic perturbation of the 1H-15N HSQC spectra and the identification of the residues in contact with those ligands. The most perturbed residues in presence of ligands are lining the internal cavity of the protein. Ka values of 1.8·10−4 M−1 and 6.3·104 M−1 were measured for CA and CDCA, respectively. This is the first report of a START domain protein in complex with a sterol ligand. Our original findings indicate that STARD5 may be involved in the transport of bile acids rather than cholesterol.

Using pharmaceutical and overexpression approaches we have previously reported that in H295R cells, (a) angiotensin II (AII) activates PKCɛ, PKCα and p44/42 MAPK pathway, (b) PKCɛ, PKCα and p44/42 MAPK overexpression inhibits AII-induced CYP11B2 gene transcription and (c) overexpression of PKCɛ inhibits CYP11B2 gene transcription through p44/42 MAPK activation [LeHoux, J.G., Dupuis, G., Lefebvre, A., 2001. Control of CYP11B2 gene expression through differential regulation of its promoter by atypical and conventional protein kinase C isoforms. J. Biol. Chem. 276 (11), 8021–8028; LeHoux, J.G., Lefebvre, A., 2006. Novel protein kinase C-epsilon inhibits human CYP11B2 gene expression through ERK1/2 signalling pathway and JunB. J. Mol. Endocrinol. 36 (1), 51–64]. The aim of the present work was to evaluate the physiological role of endogenous PKCɛ and PKCα isoforms in the activation of p44/42 MAPK by AII. A 50% reduction of PKCɛ protein by siRNA-PKCɛ resulted in 35% inhibition of AII-induced p44/42 MAPK activation. Knockdown of PKCɛ stimulated AII-induced CYP11B2 transcription indicating that the PKCɛ is not involved in the activation of CYP11B2 gene expression by AII. Furthermore, knockdown of PKCα enhanced AII-stimulated CYP11B2 transcription without altering p44/42 MAPK indicating that inhibition of AII-stimulated CYP11B2 gene by PKCα does not involve the p44/42 MAPK signalling pathway. These results thus establish that physiologically, PKCɛ and PKCα act through different signalling pathways to inhibit AII-stimulated CYP11B2 gene expression.

We previously reported that H295R cells co-express three diacylglycerol (DAG)-dependent protein kinase Cs (PKCs), namely conventional (c) PKCalpha and novel (n) PKCepsilon and PKCtheta. The aim of the present work was to evaluate the implication of DAG-dependent PKCs in the activation of p44/42 MAP kinase (MAPK) by angiotensin II (Ang II) and to define the role of this pathway towards CYP11B2 regulation in H295R cells. The PKC inhibitor bisindolylmaleimide 1 (Bis) inhibited Ang II-induced p44/42 MAPK phosphorylation whereas the cPKC inhibitor Gö6976 failed to do so, thus ruling out the participation of PKCalpha. Ang II activated nPKCepsilon and did not affect nPKCtheta, pinpointing PKCepsilon as the mediator of Ang II in p44/42 MAPK activation. Overexpression of wild-type ERK1 and ERK2 significantly reduced basal as well as Ang II-stimulated human -2023CYP11B2-CAT activity; conversely, the two dominant negative mutants increased them. Overexpression of constitutively active (ca) PKCsuppressed Ang II-induced -2023CYP11B2-CAT activity. Infection of H295R cells with adenoviruses (Adv) expressing caPKCepsilon activated endogenous MEK1/2 and p44/42 MAPK. Adv-caPKCepsilon inhibited Ang II-stimulated aldosterone synthase mRNA levels and this action was reversed by the MEK1 inhibitor, PD98059. Also, Ang II increased JunB protein levels and this effect was inhibited by PD98059 and Bis. Adv-caPKCepsilon enhanced JunB protein levels and PD98059 attenuated the increase. JunB overexpression abolished the Ang II-induced promoter activity within -138 bp of the 5'-flanking region of CYP11B2. Collectively, these results demonstrate that PKCepsilon inhibits CYP11B2 transcription through the p44/42 MAPK pathway and JunB in H295R cells.

We have previously reported that the protein kinase C ligand 12‐O‐tetradecanoyphorbol‐13‐acetate (TPA) inhibited the angiotensin II (AII) stimulated CYP11B2 gene expression in the adrenocortical H295R cell line. Here we report that TPA increased the level of phospho‐p44/42 MAPK but AII did not. The MEK1 inhibitor PD98059 was found to increase the level of aldosterone synthase mRNA and the activity of a human CYP11B2(‐2023 bp)‐promoter construct. The cotransfection of H295R with ERK 1 and the hCYP11B2 promoter resulted in the inhibition of the promoter activity. TPA but not AII increased the level of the transcription factor JunB in nuclear extracts and the increase was partially abolished by the MEK1 inhibitor PD98059. The cotransfection of H295R with JunB and the hCYP11B2 promoter abolished the AII stimulating effect. Taken together these results suggest that TPA inhibits the AII‐dependent activation of CYP11B2 via the p44/42 MAPK signaling pathway leading to an increase of the level of nuclear JunB.

We reported previously that the protein kinase C (PKC) inhibitor GF109203X stimulated the hamster CYP11B2 promoter activity in transfected NCI-H295 cells, PKC alpha, -epsilon, and -zeta were detected in hamster adrenal tons glomerulosa and NCI-H295 cells, and PKC theta in NCI-H295 cells, 12-O-Tetradecanoylphorbol-13-acetate (TPA) inhibited basal and stimulated cytochrome P450 aldosterone synthase mRNA expression by angiotensin (AII), dibutyryl cyclic adenosine 3':5'-monophosphate (Bt(2)cAMP), or KCl in NCI-H295 cells. Basal CYP11B2 promoter activity was inhibited in cells cotransfected with constitutively active (CA) PKC alpha, -epsilon, and -theta mutants, whereas it was increased with CA-PKC zeta. Dominant negative (DN) PKC alpha, -theta, -epsilon, and -zeta mutants stimulated the promoter activity, AII-, KCl-, and Bt(2)cAMP-stimulatory effects were abolished in cells cotransfected with CA-PKC alpha, -epsilon or -theta. The effect of Bt(2)cAMP was abolished by CA-PKC zeta but AII and KCl were still able to enhance the promoter activity, DN-PKC alpha, -epsilon, -theta, or -zeta did not inhibit these effects, Go6976 enhanced promoter activity, providing further evidence that PKC alpha was involved. Various CYP11B2 promoter constructs were used to identify the area associated with TPA and PKC inhibition. TPA and CA-PKC alpha, -epsilon, or -theta abolished the effects of AII, KCl, and Bt(2)cAMP on the activity of -102 and longer constructs. In summary, our findings suggest that the hamster CYP11B2 gene is under differential control by conventional (alpha) and atypical (zeta) PKC.

Bisindolylmaleimide, a protein kinase C (PKC) inhibitor, was shown to stimulate the hamster CYP11B2 promoter activity in transfected NCI-H295 cells. In this study we have found that TPA, an activator of PKC, also inhibited the hamster CYP11B2 promoter activity. DAG-dependent PKC α and PKC ω, and atypical PKC were detected in hamster adrenal zona glomerulosa, whereas the isoforms α, β were found in NCI-H295 cells. CYP11B2 promoter activity was inhibited in cells co-transfected with constitutively active PKC α and mutants, whereas it was increased with the constitutively active PKC mutant. Dominant negative PKC α, mutants stimulated the promoter activity. Gö6976, a specific inhibitor of classical PKCs, enhanced promoter activity, providing further evidence that PKC α, the only classical PKC revealed in hamster adrenal and NCI-H295 cells, was involved in the promoter inhibition.

Previous studies have shown that the hamster CYP11B2 gene promoter is under the influence of angiotensin II (AII), cAMP and potassium (K+). However, very little is known about the mechanisms by which these compounds regulate the transcription of the CYP11B2 gene. Therefore we analysed the 5'-flanking region of the hamster CYP11B2 gene using a transient transfection expression system in NCI-H295 adrenocortical cells, which are known to respond to K+, cAMP and AII. The first 486 bp before the transcription initiation site were introduced upstream of the chloramphenicol acetyl transferase gene. NCI-H295 cells transfected with this - 486 construct showed increased CAT activity upon treatment by K+, AII, forskolin and cAMP. The calcium channel antagonist nifedipine partially blocked the enhancing effects of AII, forskolin and cAMP by 35%, 30% and 30% respectively, whereas it completely blacked the stimulatory effects of KCl (1). These results thus show the involvement of calcium channels in the regulation of CYP11B2 gene transcription by K+, and their partial involvement in the regulation of this gene by AII, forskolin and cAMP in NCI-H295 cells.

We studied the regulation of the hamster CYP11B2 gene in the NCI-H295 cell line, which is known to produce aldosterone in response to stimulation by angiotensin II (AII) and KCl. Ten deletion plasmids harboring the 5'-untranslated region of the CYP11B2 gene were used for chloramphenicol acetyltransferase (CAT) assays. Transient transfections showed progressively increasing basal promoter activity by constructs beyond the TATA box, with a peak occurring with the -167 bp construct which contains putative Adl, Ad2, Ad5 and the newly reported -143/-161 cis-element sequences. The promoter activity was lower with the construct containing the putative Ad3 cis-element and increased with longer constructs. This indicates the presence of both inhibitory and stimulatory cis-elements in this area of the gene. Expression of the reporter gene of all constructs was stimulated by AII and KCl, with the exception of the construct containing only the TATA box, which showed 6-fold and 10-fold increases occurring with the -167 bp deletion plasmid. The patterns of increase in CAT activity with AII and KCI treatment were similar, showing that these two regulators can stimulate hamster CYP11B2 promoter activity through common cis-elements. The calcium channel antagonist nifedipine blocked the stimulatory effects of KCl on CAT activity, showing the involvement of calcium channels in the regulation of CYP11B2 gene transcription by KCl. 12-O-Tetradecanoylphorbol 13-acetate, a known stimulator of the protein kinase C (PKC) signaling pathway, was without significant effect on CAT activity. Bisindolylmaleimide, a specific inhibitor of PKC, had a significant enhancing effect (3.4- to 6-fold), indicating that PKC may negatively regulate the expression of the hamster CYP11B2 gene in NCI-H295 cells. A mutation was induced in the sequence -143/-161 of the - 350 bp construct in order to determine its importance in the regulation of hamster CYP11B2 promoter activity. The stimulatory effects of AII, KCl, forskolin and bisindolylmaleimide on CAT activity were significantly less in the mutant than in the wild type. These results confirm that this cis-element is necessary in maintaining a high level of transcriptional activity in stimulated NCI-295H cells. In conclusion, using NCI-295H transfected cells, we have found that the 5'-untranslated region of the hamster CYP11B2 gene possesses transcriptional activity with stimulatory and also inhibitory cis-elements; CYP11B2 promoter activity can be stimulated by AII, KCl, forskolin, dibutyryl cAMP and bisindolylmaleimide. Our results suggest that this gene is positively regulated through the protein kinase A signaling pathway and through calcium channels, whereas PKC may have a negative regulatory effect upon the transcription of the CYP11B2 gene. Furthermore, we have shown that the cis-element -143/-161 in the 5'-untranslated region of the hamster CYP11B2 gene is important in maintaining a high level of promoter activity in stimulated NCI-295H cells.

A CYP11B2 gene encoding cytochrome P450 aldosterone synthase (P450aldo) was isolated from a hamster genomic library. The gene, which contained 9 exons, was composed of 9,045 bp, of which 3,722 bp were located in the 5' untranslated region (5' UTR). A TATA box sequence (gataaa) and other putative cis elements, previously named Ad1 to Ad6, were identified in the 5' UTR of the hamster gene comparable to the CYP11B2 gene of other animal species. Footprint analysis showed protection by nuclear protein extracts from hamster adrenal zona glomerulosa (ZG) in the regions containing the above mentioned cis elements. In addition, a new protected cis element, between -143 and -161 bp, was demonstrated, and gel-shift assays revealed that the sequence of this new cis element was specifically retarded by factors in the nuclear extracts of hamster adrenal ZG. We then examined the transcriptional activity of the 5' UTR of the CYP11B2 gene, using chloramphenicol acyltransferase (CAT) as the reporter gene. Ten deletion plasmids were constructed using a modified pCAT vector. Transient transfections of the chimeric reporter constructs into Y1 cells showed that the highest basal promoter activity was obtained with the construct containing up to -134 bp. Increasing the length of the regulatory region of CYP11B2 gene to -167 bp resulted in less than two-thirds of the maximal activity, indicating the probability of putative inhibitory cis elements in this area of the gene. Forskolin stimulated the expression of the reporter gene of deletion plasmids excepting the construct containing only the TATA box, and the highest activity also occurred with the -134 bp construct. TPA had no stimulatory effects on any of the constructs, and interestingly it slightly inhibited CAT activity. In contrast to TPA, staurosporine, an inhibitor of the PKC pathway, stimulated CAT activity. To conclude, the promoter region of the hamster CYP11B2 gene transfected in Y1 cells is responsive to forskolin, indicating that the gene is controlled by the PKA signaling pathway. Paradoxically, staurosporine, but not TPA, stimulates the promoter activity of the CYP11B2 gene, indicating that PKC might, at least in Y1 cells, act as a negative regulator on the aldosterone synthase promoter. Moreover, a new cis element was shown to exert a negative effect on basal as well as on stimulated activities of the hamster promoter CYP11B2 gene.

In the current work we studied the effects of a low sodium intake on P450 aldosterone synthase (P450aldo) in the adrenal cortex of male hamsters by Western blotting analysis. We also investigated the zonal distribution of P450aldo with a specific antibody using immunofluorescence and immuno-gold electron microscopy. Western blotting analysis revealed a progressive induction of P450aldo in the adrenals of hamsters kept on a low sodium diet, with two-, four- and eightfold increases after 2, 4 and 21 days on the diet. Immunofluorescence microscopy showed that P450aldo was confined to the zona glomerulosa (ZG) cells. Electron microscopy showed P450aldo to be located in the mitochondria of ZG cells. When hamsters were maintained on a low sodium intake for 2, 11 and 21 days, P450aldo was still found only in the ZG; the ZG appeared either unchanged or sometimes slightly enlarged. Moreover, at days 11 and 21, the intensity of the immunofluorescent signal was much stronger in the ZG of hamsters on the low sodium intake than in controls. Hence, immunocytochemistry using the colloidal-gold technique showed P450aldo to be more abundant in the mitochondria of the experimental animals than in controls. To conclude, P450aldo is present only in the ZG of hamster adrenals and sodium restriction appears to induce its expression by stimulating production within individual ZG cells rather than by stimulating a proliferation of the ZG cells.

In this study, we report the cloning of a StAR cDNA from a hamster adrenal cDNA library. The library was screened using a PCR fragment specific for the hamster adrenal StAR cDNA. Several clones of different lengths were obtained and one of these was sequenced. Northern blotting analysis revealed the presence of the StAR mRNA in male and female adrenals, in testes and ovaries, but not in the liver or kidneys of either sex. Whole hamster adrenals revealed the presence of four mRNAs of 0.65, 1.7, 3.1 and 5.25 kb, respectively. In addition, ACTH regulates the expression of StAR mRNA in hamster adrenals. Indeed, when groups of hamsters were injected with ACTH and sacrificed at different times after treatment, only the 0.65 kb form of the StAR mRNA did not increase, whereas the other forms increased at varying levels. These results might suggest that the expression of the StAR protein in hamster adrenals depends upon different genes, different promoters, or different polyadenylation signal sites. In conclusion, these results indicate that in vivo, StAR is regulated by ACTH, suggesting the participation of this protein in controlling the transformation of cholesterol to pregnenolone, a key regulatory step in corticosteroidogenesis.

We have isolated a hamster adrenal P450C11 cDNA which shared 90 and 84% homology, respectively, with the nucleotide sequence and the amino acid sequence of the hamster adrenal P450aldo. Both P450C11 and P450aldo cDNA coding sequences were inserted in the plasmid pBluescript SK, transcribed and then translated using a rabbit reticulocyte system in the presence of [35S]methionine. The reaction products were immunoprecipitated with an anti-bovine P450C11 antibody for P450C11 and with an anti-hamster P450aldo for P450aldo. Immunoprecipitated proteins were analyzed by polyacrylamide gel electrophoresis. A single 35S-labeled protein band was detected for P450C11 and for P450aldo, respectively. P450C11 and P450aldo cDNAs were then both inserted into the expression vector pCMV5 containing a viral sequence specific for the attachment of ribosomes to mRNA. These constructions were transfected in COS-1 cells. 24 h after transfection, the presence of P450C11 and P450aldo mRNAs was determined by Northern blot analysis. In a time study experiment we found that P450C11 transformed the labeled-steroid into [14C]corticosterone, [14C]19-OH-deoxycorticosterone and [14C]18-OH-deoxycorticosterone in ratios of 1:1.11:0.07, after 2 h of incubation; no [14C]aldosterone could be detected. Cells transfected with plasmids harboring the P450aldo cDNA transformed [14C]deoxycorticosterone to [14C]corticosterone, [14C]aldosterone, [14C]18-OH-corticosterone, [14C]18-OH-deoxycorticosterone, [14C]19-OH-deoxycorticosterone and [14C]11-dehydrocorticosterone in ratios of 1:0.25:0.45:0.04:0.04:0.04 after 12 h of incubation. These results indicate that one P450 catalyzes the ultimate step of glucocorticoid formation and a separate P450 is involved in the final steps of aldosterone formation in hamster adrenals. The capacity of the hamster adrenal P450C11 to hydroxylate at positions 11β and 19 in nearly equal ratio makes this animal an excellent model to study the mechanism of synthesis and inhibition of 19-OH-deoxycorticosterone, the precursor of 19-nor-deoxycorticosterone, a very potent mineralocorticoid involved in the development of essential hypertension.

The zonal distribution of aldosterone synthase cytochrome P450 (P450aldo) in the adrenal cortex of male hamsters was investigated by immunofluorescence and electron microscopy, using an anti-P450aldo peptide antibody. On cryostat sections the immunolocalization of P450aldo was confined to the zona glomerulosa cells. On semi-thin plastic sections, P450aldo was shown to be located in mitochondria. Studies in electron microscopy, using the colloidal gold technique, confirmed that P450aldo was located in mitochondria.