Potential applications of using tissue-specific EVs in targeted therapy and vaccinology.

Many cell types secrete spherical membrane bodies classified as extracellular vesicles (EVs). EVs participate in intercellular communication and are present in body fluids, including blood, lymph, and cerebrospinal fluid. The time of EVs survival in the body varies depending on the body's localisation. Once the EVs reach cells, they trigger a cellular response. Three main modes of direct interaction of EVs with a target cell were described: receptor-ligand interaction mode, a direct fusion of EVs with the cellular membrane and EVs internalisation. Studies focused on the medical application of EVs. Medical application of EVs may require modification of their surface and interior. EVs surface was modified by affecting the parental cells or by the direct amendment of isolated EVs. The interior modification involved introducing materials into the cells or direct administrating isolated EVs. EVs carry proteins, lipids, fragments of DNA, mRNA, microRNA (miRNA) and long non-coding RNA. Because of EVs availability in liquid biopsy, they are potential diagnostic markers. Modified EVs could enhance the treatment of diseases such as colorectal cancer, Parkinson's disease, leukaemia or liver fibrosis. EVs have specific tissue tropisms, which makes them convenient organ-directed carriers of nucleic acids, drugs and vaccines. In conclusion, recently published works have shown that EVs could become biomarkers and modern vehicles of advanced drug forms.

Gene Expression in MC3T3-E1 Cells Treated with Diclofenac and Methylprednisolone.

Nonsteroidal anti-inflammatory drugs (NSAIDs) and glucocorticoids (GCs) are often used to treat articular-skeletal disorders. The extended use of NSAIDs and GCs have adverse effects on bone metabolism, reducing bone quality and impairing fracture healing. In the present study, we used mouse pre-osteoblast cells MC3T3-E1 to demonstrate the effects of diclofenac (DF) and methylprednisolone (MP) on cell proliferation and gene expression. Cells were incubated with three doses of DF or MP: 0.5 µM, 5 µM, and 50 µM. MP decreased cell viability even after 24 h, but DF inhibited cell viability after only seven days of treatment. The cells were lysed after one, two, three, and seven days of treatment, and gene expression was analyzed by reverse transcription and quantitative PCR (RT-qPCR) assays. DF did not significantly affect the expression of the osteogenic marker genes. MP modified the expression of Osx, Runx, and Col1a1. We concluded that MP is a more potent inhibitor of mouse pre-osteoblast differentiation and viability than is DF. Our results suggest that prolonged DF treatment could be less harmful to osteoblasts than MP treatment.

Trabecular bone remodelling in the femur of C57BL/6J mice treated with diclofenac in combination with treadmill exercise.

PURPOSE: Analgesic treatment with diclofenac deteriorates bone structure and decreases biomechanical properties. This bone loss has been though to be reversed by training. The impact of exercise on bone treated with diclofenac (DF) has reminded elusive. In the present study, we assayed the combined impact of exercises and DF on mouse femur. METHODS: The femur samples we obtained from 30 days treated C57BL/6J female mice. The training group ran on a horizontal treadmill at 12 m/min by 30 min a day (5% grade/slope). The group of ten mice treated with DF received the drug subcutaneously every day (5 mg/kg of body weight/day). The combined group ran on the treadmill and obtained DF. After 30 days, we sacrificed mice and studied their femurs using microcomputed tomography (µCT), dynamic mechanical analysis (DMA) and nanoindentation. RESULTS: We observed that treadmill running and DF decreased trabecular bone volume and mineral density. Combined effect of training and DF was not additive. A significant interaction of both parameters suggested protective effect of training on bone loss provoked by DF. The femur cortical bone shell remained untouched by the training and treatment. The training and the DF treatment did not alter the storage modulus E' significantly. The unchanged storage modulus would be suggesting on the unaltered bone strength. CONCLUSIONS: We concluded that even relatively short time of training with concomitant DF treatment could be protective on trabecular bone. Although viscoelastic properties of the entire femur were not modulated, femur trabecular tissue was thinned by treatment with DF and protected by training.

Spexin Promotes the Proliferation and Differentiation of C2C12 Cells In Vitro-The Effect of Exercise on SPX and SPX Receptor Expression in Skeletal Muscle In Vivo.

SPX (spexin) and its receptors GalR2 and GalR3 (galanin receptor subtype 2 and galanin receptor subtype 3) play an important role in the regulation of lipid and carbohydrate metabolism in human and animal fat tissue. However, little is still known about the role of this peptide in the metabolism of muscle. The aim of this study was to determine the impact of SPX on the metabolism, proliferation and differentiation of the skeletal muscle cell line C2C12. Moreover, we determined the effect of exercise on the SPX transduction pathway in mice skeletal muscle. We found that increased SPX, acting via GalR2 and GalR3 receptors, and ERK1/2 phosphorylation stimulated the proliferation of C2C12 cells (p < 0.01). We also noted that SPX stimulated the differentiation of C2C12 by increasing mRNA and protein levels of differentiation markers Myh, myogenin and MyoD (p < 0.01). SPX consequently promoted myoblast fusion into the myotubule (p < 0.01). Moreover, we found that, in the first stage (after 2 days) of myocyte differentiation, GalR2 and GalR3 were involved, whereas in the last stage (day six), the effect of SPX was mediated by the GalR3 isoform. We also noted that exercise stimulated SPX and GalR2 expression in mice skeletal muscle as well as an increase in SPX concentration in blood serum. These new insights may contribute to a better understanding of the role of SPX in the metabolism of skeletal muscle.

The Regulation of Collagen Processing by miRNAs in Disease and Possible Implications for Bone Turnover.

This article describes several recent examples of miRNA governing the regulation of the gene expression involved in bone matrix construction. We present the impact of miRNA on the subsequent steps in the formation of collagen type I. Collagen type I is a main factor of mechanical bone stiffness because it constitutes 90-95% of the organic components of the bone. Therefore, the precise epigenetic regulation of collagen formation may have a significant influence on bone structure. We also describe miRNA involvement in the expression of genes, the protein products of which participate in collagen maturation in various tissues and cancer cells. We show how non-collagenous proteins in the extracellular matrix are epigenetically regulated by miRNA in bone and other tissues. We also delineate collagen mineralisation in bones by factors that depend on miRNA molecules. This review reveals the tissue variability of miRNA regulation at different levels of collagen maturation and mineralisation. The functionality of collagen mRNA regulation by miRNA, as proven in other tissues, has not yet been shown in osteoblasts. Several collagen-regulating miRNAs are co-expressed with collagen in bone. We suggest that collagen mRNA regulation by miRNA could also be potentially important in bone metabolism.

Transforming growth factor ß mediates communication of co-cultured human nucleus pulposus cells and mesenchymal stem cells.

Intervertebral disc (IVD) consists of surrounding tissue annulus fibrosus and central nucleus pulposus, which are partially degenerative in scoliotic IVDs. Successful regeneration of scoliotic alterations requires cognition of critical paracrine mediators of cell-to-cell contact in the IVD. In this work, we hypothesized that transforming growth factor ß (TGF-ß) is involved in the intercellular communication of nucleus pulposus cells (NPCs) and mesenchymal stem cells (MSCs). We observed that in cultured NPCs TGF-ß1 stimulated COL1A1 expression, encoding collagen I, and in MSCs stimulated COL1A1 and SOX9 expressions. We subsequently co-cultured NPCs and MSCs together using direct and indirect transwell systems. The expression of miR-140 and miR-145 were decreased in co-cultured NPCs. We observed that direct co-culture system stronger than the indirect system decreased expression of three miRNA. The expression of COL1A1, ACAN, encoding aggrecan, and SOX9 genes was increased in MSCs co-cultured with NPCs. Co-cultures were incubated with two inhibitors of TGF-ß type I receptor: SB-431542 and SB-525334. In co-cultured NPCs, SB-431542 and SB-525334 annulated downregulation of miR-140 and miR-145. In MSCs these inhibitors diminished stimulation of COL1A1, ACAN, and SOX9. We concluded that stimulation of COL1A1, ACAN, and SOX9 in co-cultured MSCs and regulation of miR-140 and miR-145 in NPCs were TGF-ß-dependent and TGF-ß is involved in the communication of NPCs and MSCs in co-culture. © 2018 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. J Orthop Res 36:3023-3032, 2018.

Quantification of the asymmetric migration of the lipophilic dyes, DiO and DiD, in homotypic co-cultures of chondrosarcoma SW-1353 cells.

DiO and DiD are lipophilic cell labelling dyes used in the staining of cells in vivo and in vitro. The aim of the present study was to quantify the asymmetrical distribution of dyes in co­cultured cells and to measure the intercellular transfer of DiO and DiD. DiO and DiD were applied separately to stain two identical populations of SW­1353 human chondrosarcoma cells that were subsequently co­cultured (homotypic co­culture). The intercellular migration of dyes in the co­cultured cells was measured by flow cytometry and recorded under a fluorescent microscope. DiD and DiO caused no effect on the proliferation of cells, the degradation rate of the two dyes was comparable and crossover effects between dyes were negligible. The results of the present study suggested that asymmetrical intercellular migration of DiD and DiO was responsible for the asymmetrical distribution of these dyes in co­cultured cells. To take advantage of the lipophilic dyes migration in the double-stained co-cultured cells we suggest to apply mixed-dyes controls prior to the flow cytometric analysis. These controls are performed by staining cells with a 1:1 mix of the two dyes and would enable the estimation of the intensity of intercellular contact in co­culture systems. A 1:1 premix of DiO and DiD was applied to estimate cellular effect on intercellular exchange of lipid dyes in co­cultures incubated with cycloheximide and cytochalasin B. The cellular effect contributed 6­7% of intercellular migration of the lipophilic dyes, DiO and DiD. The majority of the observed intercellular transfer of these dyes was due to non­cellular, passive transfer.

Relative levels of let-7a, miR-17, miR-27b, miR-125a, miR-125b and miR-206 as potential molecular markers to evaluate grade, receptor status and molecular type in breast cancer.

MicroRNAs (miRNAs/miRs) are a class of short, single­stranded nucleic acids, which have been investigated as potential molecular markers for various types of cancer. The gold­standard and most sensitive method for comparing miRNA levels in cancer tissues is reverse transcription­quantitative polymerase chain reaction (RT­qPCR). This technique uses stably expressed genes for normalisation. The aim of the present study was to improve this model of analysis in the context of RT­qPCR results. A total of six known miRNAs (let­7a, miR­17, miR­27b, miR­125a, miR­125b and miR­206), RNU6B RNA and five mRNAs [erb­b2 receptor tyrosine kinase 2 (ERBB2), hydroxymethylbilane synthase and polymerase (RNA) II (DNA directed) polypeptide A] were analysed pair­wise, in order to determine which biomarker pairs best correlated with the histological groups of 27 breast cancer samples. The lowest P­values and the highest area under the curve values in the receiver operating characteristic analysis were used to select the optimum ratios for discrimination among groups. Among the 21 pairs, miR­17/miR­27b and miR­125a/RNU6B best discriminated three groups of samples with different tumour grades (G classification). miR­125b/miR­206 best discriminated two groups of samples with different tumour sizes (pT), let­7a/RNU6B best discriminated two groups of samples with different lymph node status (pN), and let­7a/miR­125b best discriminated groups of samples with negative and positive oestrogen and progesterone receptor status. No pair of miRNAs was found to discriminate well between groups with either a negative or positive human epidermal growth factor receptor 2 (HER2) status. However, one miRNA/mRNA pair, miR­125a/ERBB2, discriminated HER2­negative from HER2­positive groups. The breast cancer samples investigated in the present study were grouped by immunohistological methods into three molecular classes: Luminal, HER2 positive and basal (L, H and B, respectively). In order to discern L from H and L from B, two miRNA pairs were selected: miR­125a/miR­125b and miR­125a/miR­206. In conclusion, the pair­wise method of RT­qPCR data analysis may be a reasonable alternative to the standard method of using stably expressed reference genes, such as RNU6B RNA, for normalisation. This method may increase the classification power of miRNA biomarkers in breast cancer diagnostics.

Anti­tuberculosis drugs decrease viability and stimulate the expression of chondrocyte marker genes in human nucleus pulposus cells.

Isoniazid (INH), rifampicin (RIF), ethambutol (ETH) and pyrazinamide (PYR) are first­line drugs used in anti­tuberculosis (TB) therapy. However, no studies have been conducted concerning the effect of anti­TB drugs on the cells of the intervertebral discs (IVDs), the predominant location of the osteoarticular form of TB (OATB). Cells from the nucleus pulposus (NP), which are located in the center of the IVDs, were obtained from 12 adolescent patients who underwent surgery due to idiopathic scoliosis. The NP cells were incubated for 24 h with transforming growth factor ß1 (TGF­ß1) and each anti­TB drug (INH, RIF, ETH and PYR), separately. Incubation with 2.5 ng/ml TGF­ß1 resulted in an 80% decrease in ACAN mRNA levels; while 5 µg/ml INH led to a 2.3­fold increase in COL2A1 and a 2.9­fold increase in ACAN mRNA levels. Treatment with 10 µg/ml RIF initiated a 2.2­fold increase in COL1A1 mRNA levels and 5 µg/ml PYR resulted in an 8­fold increase in SOX9 mRNA levels. Following 192 h of treatment with INH and RIF, NP cell viability was diminished; however, no drugs modified the concentrations of glycosaminoglycans (GAGs). This study aimed to determine the effect of anti­TB drugs on the expression of chondrocyte marker genes in human IVD cells. Anti­TB drugs increased the expression of chondrocyte marker genes and diminished the viability of IVD cells. This study demonstrated that in addition to the common side effects of anti­TB drugs, these drugs also have an effect on IVD cells.

Co­culture of human nucleus pulposus cells with multipotent mesenchymal stromal cells from human bone marrow reveals formation of tunnelling nanotubes.

Degeneration of the intervertebral disc (IVD) is the main cause of age-related damage of spinal tissues. Using multipotent mesenchymal stromal cells (MSCs) regenerative medicine intends to restore the IVD components of annulus fibrosus (AF) and nucleus pulposus (NP). In the present study NP cells (NPCs) and MSCs obtained from adolescent patients suffering from scoliosis were used. IVDs and vertebrae were obtained during surgery and subsequently processed in order to establish cultures of NPCs and MSCs. The two cell types were co-cultured in 1-µm pore size insert system (indirect co-culture) or on one surface (direct co-culture). Prior to co-culture in these systems one of the cell types was stained by lipophilic fluorescent dye DiD (red). The results demonstrated that regardless of the cell type, the flow of DiD from stained to non-stained cells was more efficient in the direct co-culture in comparison with the insert system. Moreover, in the direct system the DiD flow was more efficient from MSCs towards NPCs compared with that in the opposite direction. These data indicated that the membrane interchange between the two cell types was asymmetric. To discriminate the subpopulation of cells that underwent membrane interchange, cells were double stained with DiD and DiO (green). In the first part of the experiment NPCs were stained by DiO and MSCs by DiD. In the second, NPCs were stained by DiD and MSCs by DiO. The cells were co-cultured in the direct system for 8 days and subsequently analyzed by flow cytometry and confocal microscopy. This analysis revealed that >50% of cells were stained by the DiO and DiD dyes. NPCs and MSCs formed structures similar to tunnelling nanotubes (TnT). In conclusion, the formation of TnT-like structures is able to promote, phenotypic changes during the direct co-culture of NPCs with MSCs.

rs12976445 variant in the pri-miR-125a correlates with a lower level of hsa-miR-125a and ERBB2 overexpression in breast cancer patients.

Expression of MIR125A is diminished in breast tumors, however the reason for the hsa-mir-125a decrease in the cancer is not known. HER2 is encoded by ERBB2, a target for hsa-miR-125a which interacts with the 3'UTR of ERBB2 mRNA. The present study reveals that a polymorphism (rs12976445) within the pri-miR-125a sequence correlates with the amount of mature hsa-miR-125a in breast tumor samples. miRNA, RNA and DNA were extracted from breast cancer samples obtained from 26 patients. Following immunohistological evaluation of the samples, the ERBB2, PGR and ESR1 mRNA profiles were also analyzed using real-time PCR. Genomic DNA was sequenced using MIR125A flanking primers. PCR products were analyzed using a BaeGI restriction enzyme specific to the rs12976445 variant. The rs12976445 variant (C/T and C/C) correlated with a lower level of hsa-miR-125a in comparison with the T/T variant. The expression of HER2 mRNA was increased in tumors with the rs12976445 variant (C/T and C/C) compared with T/T. We conclude that rs12976445 may be a potential prognostic marker of HER2 expression in breast cancer. Its predictive value on the efficacy of trastuzumab treatment in patients with HER2-positive breast cancer warrants further study.

The effect of mitotane on viability, steroidogenesis and gene expression in NCI­H295R adrenocortical cells.

Mitotane, also known as o,p'­DDD or (RS)­1­chl-oro­2­[2,2­dichloro­1­(4­chlorophenyl)­ethyl]­benzene, is an adrenal cortex-specific cytotoxic drug used in the therapy of adrenocortical carcinoma (ACC). The drug also inhibits steroidogenesis, however, the mechanisms of its anticancer and antisteroidogenic effects remain unknown. At present, data on the impact of mitotane on cell viability and the regulation of genes encoding proteins associated with steroids synthesis in the adrenal cortex, including cortisol and dehydroepiandrosterone sulfate (DHEAS), are limited and contradictory. In the present study, the effect of 24­h mitotane treatment on viability of the ACC cell line, NCI­H295R, was analyzed, identifying a decrease in cell viability and an increase in caspase­3 and ­7 activities. Mitotane treatment also led to decreased cortisol and DHEAS concentration in the culture media. Concomitantly, mitotane resulted in decreased mRNA levels of two cytochromes P450 (CYP11A1 and CYP17A1), mRNAs encoding proteins involved in the synthesis of cortisol and DHEAS. Mitotane did not affect mRNA levels of cyclin dependent kinase inhibitor 1A (encoding p21) and MYC (encoding cMyc). cMyc and p21 are key transcription factors associated with cell cycle regulation. However, mitotane inhibited expression of transforming growth factor ß1 gene, encoding a potent inhibitor of cell proliferation and steroidogenesis. PRKAR1A, a protein kinase A regulatory subunit, is involved in the activation of steroidogenesis. PRKAR1A mRNA levels were reduced following 24­h treatment with mitotane. Results indicate that mitotane markedly inhibited expression of genes involved in steroidogenesis, secretion of cortisol and DHEAS. Reduced expression of TGFB1 cannot account fully for the effect of mitotane on CYP11A1 and CYP17A1. We hypothesized that reduced viability of NCI­H295R cells in the presence of mitotane may be a result of apoptosis triggered by increased caspase­3 and ­7 activities. Since p21 and cMyc mRNA levels were stable in the presence of mitotane, the mechanism by which caspase­3 and ­7 are induced remains unknown.

TGF-beta inhibits CYP17 transcription in H295R cells acting via activin receptor-like kinase 5.

OBJECTIVE: Transforming growth factor beta (TGF-beta) is a potent inhibitor of 17alpha-hydroxylase/17,20 lyase activity and CYP17 gene expression. We investigated the mechanism how CYP17 is inhibited by TGF-beta in adrenocortical cells. METHODS: H295R cells were culture and incubated with TGF-beta, transcription inhibitor (DRB), activin receptor-like kinase 5 ALK5 (TbetaRII) inhibitor (SB431542), mitogen activated kinases inhibitors (PD98059 and SB203580), subsequently using reverse transcription and quantitative PCR (RT-qPCR) we determined CYP17 expression. RESULTS: TGF-beta significantly decreased the level of cytochrome P450c17 mRNA and this inhibitory effect of TGF-beta on CYP17 expression required activin receptor-like kinase 5 (ALK5) and on-going transcription. Mitogen activated kinases MEK1 and p38 MAPK are not involved it the inhibitory effect of TGF-beta on CYP17 expression. CONCLUSION: We concluded that the TGF-beta-dependent decrease of 17alpha-hydroxylase/17,20 lyase activity in the H295R cells is caused by inhibition of CYP17 transcription and is mediated by the ALK5 receptor.

SMAD3 inhibits SF-1-dependent activation of the CYP17 promoter in H295R cells.

Cytochrome P450c17, encoded by the CYP17 gene, is a component of 17alpha-hydroxylase/17,20 lyase which catalyses 17alpha-hydroxylation of pregnenolone or progesterone, required for glucocorticosteroid and androgen synthesis. It has been reported that transforming growth factor beta (TGF-beta) decreases both basal and cAMP-stimulated levels of CYP17 mRNA, but the mechanism of TGF-beta action on CYP17 expression remains unknown. We investigated an inhibitory effect of TGF-beta on CYP17 expression in H295R cells using constructs containing the CYP17 promoter region fused with the luciferase gene. In the H295R cells, TGF-beta decreased endogenous SF-1 level and inhibited activity of the 300 bp fragment of CYP17 promoter, which was stimulated by coexpression of SF-1. Overexpression of SMAD3 caused an inhibition of SF-1-stimulated CYP17 promoter activity, whereas overexpression of SMAD7 was ineffective. In conclusion, our results suggest that the inhibitory action of TGF-beta on CYP17 transcription involve at least two mechanisms: SMAD3 dependent inactivation of CYP17 promoter activity and repression of SF-1 expression.

Steroidogenic factor 1 gene transcription is inhibited by transforming growth factor beta.

The orphan nuclear receptor, steroidogenic factor 1 (SF-1), plays a major role in adrenal and gonadal development, as well as in sexual differentiation. It has been demonstrated that the expression of a number of genes regulated by SF-1 is inhibited by the transforming growth factor, (TGF-beta). To date, however, the influence of TGF-beta on the expression of SF-1 gene has not been reported. A Northern blot analysis with the use of a radiolabeled cDNA probe, and immunodetection with antibodies directed against SF-1, demonstrated that the Sf-1 transcript and the SF-1 protein levels were lowered by TGF-beta in Y-1 adrenocortical cells, both in untreated and adenylyl cyclase activator, forskolin-treated cells. An examination of the Sf-1 transcript stability in the presence of actinomycin D revealed no influence of TGF-beta on the rate of Sf-1 mRNA decay. Inhibition of Sf-1 expression by TGF-beta was abolished by cycloheximide, suggesting that the growth factor inhibitory effect requires ongoing protein synthesis. We conclude that in Y-1 cells TGF-beta inhibits the expression of SF-1 gene at a transcriptional level, and we postulate that the inhibitory effect of TGF-beta on steroid hormone synthesis in the adrenal cortex could be due to an attenuated transcription of Sf-1.

Temporal pattern of the induction of SF-1 gene expression by the signal transduction pathway involving 3',5'-cyclic adenosine monophosphate.

The objective of our study was to investigate the effect of stimulation of the cAMP-dependent pathway on the expression of an orphan nuclear receptor, SF-1/Ad4BP in mouse adrenal tumour, Y-1 cells in culture. We evaluated the temporal pattern of the effects of corticotropin (ACTH) and the adenylyl cyclase activator forskolin on the level of SF-1 mRNA, and compared the time course of induction of SF-1 with that of CYP11A1. Forskolin, corticotropin and 8-Br-cAMP significantly elevated the level of the SF-1 transcript, after 1.5 h of incubation, with a concomitant increase of SF-1 protein level, observed after 6 h. The CYP11A1 transcript increased gradually over the incubation period, and reached the maximal level after 12 to 24 h. The steady-state level of the SF-1 transcript was unaffected by forskolin when the cells were incubated with actinomycin D, indicating that stimulation of the cAMP pathway results in enhanced transcription of the gene. The effect of forskolin was augmented by cycloheximide, suggesting that an inhibitory protein, whose synthesis was inhibited by cycloheximide, could be involved in negative regulation of SF-1 expression. It is concluded that SF-1 expression is positively regulated by the cAMP pathway at the transcriptional level, and can represent the primary event in cAMP-mediated induction of steroid hormone synthesis in Y-1 cells.

Inhibition of CYP17 expression by adrenal androgens and transforming growth factor beta in adrenocortical cells.

Cytochrome P450c17, encoded by the CYP17 gene, is a component of the 17alpha-hydroxylase/17,20-lyase enzyme complex essential for production of adrenal glucocorticoids and androgens as well as gonadal androgens. The expression of CYP17 in adrenocortical cells is stimulated by corticotropin (ACTH) via the signal transduction pathway involving cAMP and protein kinase A (PKA). Thus, in addition to glucocorticoids, ACTH stimulates formation of adrenal androgens, which are known to induce transforming growth factor beta (TGF-beta) secretion. TGF-b in turn inhibits steroid hormone output by attenuating both basal and ACTH-dependent expression of CYP17. The present study revealed that treatment of bovine and human H295R adrenocortical cells with androgens resulted in a decrease in the basal level of CYP17 transcript and cortisol secretion, without affecting forskolin-stimulated levels. We also demonstrated that in H295R cells TGF-beta inhibited both basal and forskolin-stimulated accumulation of CYP17 mRNA. Determination of promoter activity, directing luciferase reporter gene expression in H295R cells transfected with deletion fragments of bovine CYP17 promoter, indicated that the -483 to -433 bp fragment of the promoter was necessary for the inhibitory action of TGF-beta on CYP17 expression. It is concluded that in bovine and human adrenocortical cells, androgens inhibit basal CYP17 expression probably at the transcriptional level and independently of the effect of TGF-beta.