Anti-androgenic properties of Compound A, an analog of a non-steroidal plant compound.

We investigated the interactions between Compound A (CpdA), an analog of a hydroxyphenyl aziridine precursor found in an African shrub, and the androgen receptor (AR). CpdA represses androgen-induced activation of both specific and non-specific androgen DNA response elements. While a similar effect was obtained for the progesterone receptor (PR) via a non-specific hormone response element, CpdA had no effect on the actions of the glucocorticoid and mineralocorticoid receptors. CpdA represses the ligand-dependent interaction between the NH(2)- and COOH-terminal domains of the AR, similar to well-characterised anti-androgens. CpdA also interferes with the interaction of steroid receptor co-activator 1 (SRC1) with the activation domain AF2 but not with AF1. However, CpdA does not compete with androgen for binding to the AR. These results demonstrate that CpdA elicits anti-androgenic actions by a mechanism other than competitive binding for the AR.

Tissue-specific androgen responses in primary cultures of lacrimal epithelial cells studied by adenoviral gene transfer.

The lacrimal gland secretes most of the water and many proteins present in tear fluid. The composition of the tear fluid is affected dramatically by androgens, an observation which has been linked to the fact that more than 90% of the patients with Sjögren syndrome are female. Although the presence of androgen receptors in the lacrimal gland has been established, the molecular biology of the protective effects of androgens remains largely unknown. Here, we report the use of primary cultures of the lacrimal gland which express endogenous proteins under androgen control, as a more homologous test system for tissue-specific transcription studies. Infection with recombinant adenoviral vectors was the most efficient method to introduce foreign gene constructs in these cultures. A thus introduced mouse mammary tumor virus promoter was inducible with androgens and this effect was independent of the sexual genotype of the infected cells. By use of two recombinant adenoviral vectors containing genomic fragments of the SC gene, which is androgen responsive in the lacrimal gland, we could demonstrate the functionality of the sc promoter as well as its androgen regulation in this culture system.

Effects and characterization of paracrine factors produced by human prostate stromal cells in bioassays using rat Sertoli cells, LNCaP tumor cells, and cultured prostate epithelial cells.

BACKGROUND: Prostatic stroma affects both proliferation and differentiation of epithelial cells but the factors involved remain poorly understood. In order to identify and characterize potential paracrine mediators, we studied the effects of human prostate fibroblast-conditioned media (PFCM) in three bioassay systems. METHODS: The first bioassay uses transferrin secretion by cultured rat Sertoli cells as an endpoint for differentiating activity. Factors active in this (heterologous) assay were compared to PModS, a mediator of mesenchymal-epithelial interactions in the testis, also produced by rat prostate stromal cells. The two other (homologous) bioassays use LNCaP tumor cells or subcultured human prostate epithelial cells (PEC) as targets. Differentiation is evaluated by prostate-specific antigen (PSA) secretion and reverse transcriptase-polymerase chain reaction (RT-PCR) for a number of markers of epithelial function. Proliferation is assayed by measurements of DNA and thymidine incorporation. RESULTS: PFCM markedly stimulates transferrin production by Sertoli cells. The main factor(s) involved are acid stable and bind to heparin. However, both their size (approximately 37 kDa) and their behavior on reversed-phase chromatography differ from that of PModS. Although PFCM increases total RNA content of LNCaP, it does not increase or restore differentiated function of LNCaP or PEC. Proliferative effects are observed in LNCaP and these effects cannot be neutralized by an antiserum directed against basic fibroblast growth factor (bFGF). Antiproliferative effects are observed in PEC and these effects are largely due to transforming growth factor-beta (TGF-beta). CONCLUSIONS: PFCM provokes differentiating effects in a Sertoli cell bioassay, but unlike with rat stromal cells, the factor(s) involved differ from PModS. In the two homologous systems studied, differentiating effects could not be demonstrated and discordant effects were noted on proliferation. Various bioassay systems will be required to identify the spectrum of mediators present in PFCM.

Selective DNA binding by the androgen receptor as a mechanism for hormone-specific gene regulation.

Steroid hormones control many physiological processes by activating specific receptors that act as transcription factors. In vivo, each of these receptors has a specific set of target genes, but in vitro the glucocorticoid, progesterone, mineralocorticoid and androgen receptors (class I receptors) all recognise response elements which are organised as inverted repeats of 5'-TGTTCT-3'-like sequences with a three nucleotide spacer. This poses the question how the in vivo specificity of the different steroid responses is mediated. To unravel the mechanisms involved, we have compared the structural features of the androgen-selective enhancers of the probasin, the secretory component and the sex-limited protein genes with those of non-selective enhancers in the mouse mammary tumour viral promoter and the C3(1) gene. The probasin promoter contains an androgen response element which is recognised with high affinity by the androgen receptor, but not by the other class I receptors. Swapping experiments between the DNA-binding domains of the androgen and glucocorticoid receptor revealed that it is not the first zinc finger, but rather the second zinc finger and part of the hinge region which contribute to this specificity. Three AR-specific aminoacids are involved in the probasin ARE recognition, but not in the C3(1) ARE binding by the AR. The location of these residues strongly suggests that an alternative dimerisation interface is involved in the probasin ARE binding. We could subsequently demonstrate that the AR binds direct repeats of 5'-TGTTCT-3'-like sequences in gel retardation assays as well as in transfection experiments. Moreover, the androgen-specific enhancers all contain direct repeats, and point mutations that change the nature of these elements into inverted repeats result in a change of specificity. It seems, therefore, that direct repeat elements can be the determinants of the AR-specificity. It will be exciting to learn how such DNA elements will affect the properties of the receptor dimer with respect to ligand binding, interactions between the aminoterminal domain and the ligand-binding domain, the recruitement of co-activators and cooperativity with other transcription factors.

Androgen-receptor-specific DNA binding to an element in the first exon of the human secretory component gene.

Androgens and glucocorticoids are steroid hormones, which exert their effects in vivo by binding and activating their cognate receptors. These intracellular receptors are transcription factors that can bind specific DNA sequences, called hormone response elements, located near the target genes. Although the androgen receptor (AR) and the glucocorticoid receptor (GR) bind the same consensus DNA sequence, androgen-specific responses can be achieved by non-conventional androgen response elements (AREs). Here we determine the specificity mechanism of such a selective element recently identified in the first exon of the human gene for secretory component (sc ARE). This sc ARE consists of two receptor-binding hexamers separated by three nucleotides. The DNA-binding domains of the AR and GR both bind the sc ARE, but, although the AR fragment dimerizes on the element, the GR fragment does not. Comparing the affinities of the DNA-binding domains for mutant forms of the sc ARE revealed that dimeric GR binding is actively excluded by the left hexamer and more precisely by the presence of a G residue at position -3, relative to the central spacer nucleotide. Inserting a G at this position changed a non-selective element into an androgen-selective one. We postulate that the AR recognizes the sc ARE as a direct repeat of two 5'-TGTTCT-3'-like core sequences instead of the classical inverted repeat. Direct repeat binding is not possible for the GR, thus explaining the selectivity of the sc ARE. This alternative dimerization by the AR on the sc ARE is also indicated by the DNA-binding characteristics of receptor fragments in which the dimerization interfaces were swapped. In addition, the flanking and spacer sequences seem to affect the functionality of the sc ARE.

Differences in DNA binding characteristics of the androgen and glucocorticoid receptors can determine hormone-specific responses.

The basis for specificity of gene regulation by steroid hormone receptors remains an important problem in the study of steroid hormone action. One possible mechanism for steroid specificity is the difference in DNA binding characteristics of the receptors, although they share a high homology in their DNA-binding domains. Indeed, the androgen-specific expression of, for example, the probasin (PB) gene can be explained by the presence of an androgen response element (ARE) in its promoter (PB-ARE-2), specifically recognized by the androgen and not by the glucocorticoid receptor. Three residues in the DNA-binding domain of the AR were identified as main determinants for its high affinity for the PB-ARE-2. In addition, the direct repeat nature of this ARE seems to prohibit high affinity binding by the glucocorticoid receptor. This is confirmed by the fact that several imperfect direct repeats of the 5'-TGTTCT-3' core recognition sequence are recognized by the androgen receptor and not by the glucocorticoid receptor. Up to now, only differences between the androgen and glucocorticoid receptor in the transcription activation functions were invoked to explain the specificity of their genomic actions. In the present study, we describe the influence of the DNA-binding domain on the specificity of androgen action. The novelty of our working hypothesis resides in the demonstration of the capacity of the AR-DNA-binding domain to recognize elements with a direct repeat structure.

Change of specificity mutations in androgen-selective enhancers. Evidence for a role of differential DNA binding by the androgen receptor.

The androgen and glucocorticoid receptors recognize identical DNA motifs, leaving unanswered the question of how steroid specificity of transcriptional regulation is established in cells containing both receptors. Here, we provide evidence that subtle differences in low affinity DNA recognition might be a crucial element in the generation of steroid-specific responses. Here we identify simple hormone response elements in the mouse sex-limited protein enhancer and the human secretory component androgen response unit to be essential for the androgen specificity of both enhancers. We describe specific in vitro binding to these motifs by the DNA-binding domain of the androgen but not the glucocorticoid receptor. Both elements can be considered partial direct repeats of the 5'-TGTTCT-3' core binding motif. In addition, we show that specific point mutations in their left half-sites, essentially changing the nature of the repeats, strongly enhance the glucocorticoid sensitivity of the respective enhancers, whereas they have no effect on their androgen responsiveness. Accordingly, these mutations allow specific binding of the glucocorticoid receptor DNA-binding domain to both elements in vitro. With these experiments, we demonstrate that differential recognition by the androgen receptor of nonconventional steroid response elements is, at least in some cases, an important mechanism in androgen-specific transcriptional regulation.

Androgens decrease and retinoids increase the expression of insulin-like growth factor-binding protein-3 in LNcaP prostatic adenocarcinoma cells.

Changes in circulating levels of insulin-like growth factors (IGF) and IGF-binding proteins (IGFBP) have been related to prostate cancer, but the nature and the significance of this relationship remains elusive. Recent reports suggest that modulation of the production of IGFBP-3 by retinoids may affect growth of breast and prostate tumor cells. In the present study we explored whether androgens (R1881), retinoids (all-trans- and 9-cis-retinoic acid: atRA and 9cRA), deltanoids (1alpha,25-dihydroxycholecalciferol: VD3) and thyroid hormone (triiodothyronine: T3) influence the production of IGFBPs by LNCaP prostatic adenocarcinoma cells and whether the observed changes affect tumor cell growth. Northern blot experiments demonstrated that LNCaP cells express IGFBP-2, -3, -4 and (to a small extent) -5. IGFBP-4 and -5 were not measurably affected by the mentioned agonists. At a growth promoting concentration (10(-10) M), R1881 increased IGFBP-2 transcript levels two- to three-fold and this effect was neutralized by atRA and VD3. Similar effects could not be demonstrated, however, at the protein level using Western ligand blotting. R1881 decreased and atRA increased the mRNA levels of IGFBP-3 and these effects were confirmed by Western ligand blotting and by radioimmunoassay. The effects of atRA were mimicked by 9cRA and by a specific RAR agonist but not by a RXR agonist. VD3 and T3 had no significant effect on IGFBP-3 secretion but respectively enhanced or decreased the effect of 9cRA. The effects of retinoids required high concentrations (10(-6)-10(-5) M) that also induced growth inhibition. R1881, however, decreased IGFBP-3 at growth promoting (10(-10) M) as well as at growth inhibitory (10(-8) M) concentrations. Moreover, under serum-free conditions, we were unable to demonstrate any growth modulating effect of IGFBP-3. It is concluded that several agonists acting by nuclear receptors affect IGFBP-3 secretion by LNCaP cells but that the functional significance of these changes warrants further investigation.

Androgen specificity of a response unit upstream of the human secretory component gene is mediated by differential receptor binding to an essential androgen response element.

The expression of secretory component (SC), the epithelial receptor for poly-immunoglobulins, is regulated in a highly tissue-specific manner. In several tissues, e.g. lacrimal gland and prostate, SC synthesis is enhanced by androgens at the transcriptional level. In this study, we describe the presence of an androgen response unit, located 3.3 kb upstream of the sc transcription initiation site and containing several 5'-TGTTCT-3'-like motifs. Although each of these elements is implicated in the enhancer function, one element, the ARE1.2 motif, is found to be the main interaction site for the androgen receptor as demonstrated in in vitro binding assays as well as in transient transfection assays. A high-affinity binding site for nuclear factor I, adjacent to this ARE, is also involved in the correct functioning of the sc upstream enhancer. The ARE1.2 motif consists of an imperfect direct repeat of two core binding elements with a three-nucleotide spacer and therefore constitutes a nonconventional ARE. We demonstrate that this element displays selectivity for the androgen receptor as opposed to glucocorticoid receptor both in in vitro binding assays and in transfection experiments. Mutational analysis suggests that the direct nature of the half-site repeat is responsible for this selectivity. We have thus determined a complex and androgen-specific response unit in the far upstream region of the human SC gene, which we believe to be involved in its androgen responsiveness in epithelial cells of different organs such as prostate and lacrimal gland. We were also able to demonstrate that the primary sequence of a single nonconventional ARE motif within the enhancer is responsible for its androgen specificity.

The first exon of the human sc gene contains an androgen responsive unit and an interferon regulatory factor element.

Secretory component (SC) plays a key role in the transport of IgA and IgM to the lumina of many glands. The gene is constitutively expressed, but can be modulated by hormonal and immunological stimuli. Recently, the promoter and the first exon of the human sc gene have been cloned. The first exon contains a putative androgen/glucocorticoid response element (ARE/GRE) and an Interferon Regulatory Factor Element (IRF-E). Here we show that the ARE/GRE can bind the DNA-binding domain (DBD) of both the androgen (AR) and glucocorticoid receptor (GR) with a preference for the AR-DBD. In transient transfection experiments, this element confers higher responsiveness to androgens than to glucocorticoids. The IRF-E can function as an IRF-2, but surprisingly not as an IRF-I responsive element. We postulate that these two regulatory elements play a key role in the complex regulation of the sc gene in vivo.

The androgen receptor amino-terminal domain plays a key role in p160 coactivator-stimulated gene transcription.

Steroid receptors are conditional transcription factors that, upon binding to their response elements, regulate the expression of target genes via direct protein interactions with transcriptional coactivators. We have analyzed the functional interactions between the androgen receptor (AR) and 160-kDa nuclear receptor coactivators. Upon overexpression in mammalian cells, these coactivators enhance the transcriptional activity of both the amino-terminal domain (NTD) and the ligand-binding domain (LBD) of the AR. The coactivator activity for the LBD is strictly ligand-controlled and depends on the nature of the DNA-binding domain to which it is fused. We demonstrate that the NTD physically interacts with coactivators and with the LBD and that this interaction, like the functional interaction between the LBD and p160 coactivators, relies on the activation function 2 (AF2) core domain. The mutation of a highly conserved lysine residue in the predicted helix 3 of the LBD (K720A), however, blunts the functional interaction with coactivators but not with the NTD. Moreover, this mutation does not affect the transcriptional activity of the full-size AR. A mutation in the NTD of activation function AF1a (I182A/L183A), which dramatically impairs the activity of the AR, has no effect on the intrinsic transcriptional activity of the NTD but interferes with the cooperation between the NTD and the LBD. Finally, p160 proteins in which the three LXXLL motifs are mutated retain most of their coactivator activity for the full-size AR, although they are no longer functional for the isolated LBD. Together, these data suggest that in the native AR the efficient recruitment of coactivators requires a functional association of the NTD with the LBD and that the binding of coactivators occurs primarily through the NTD.

Differential DNA binding by the androgen and glucocorticoid receptors involves the second Zn-finger and a C-terminal extension of the DNA-binding domains.

The androgen and glucocorticoid hormones evoke specific in vivo responses by activating different sets of responsive genes. Although the consensus sequences of the glucocorticoid and androgen response elements are very similar, this in vivo specificity can in some cases be explained by differences in DNA recognition between both receptors. This has clearly been demonstrated for the androgen response element PB-ARE-2 described in the promoter of the rat probasin gene. Swapping of different fragments between the androgen- and glucocorticoid-receptor DNA-binding domains demonstrates that (i) the first Zn-finger module is not involved in this sequence selectivity and (ii) that residues in the second Zn-finger as well as a C-terminal extension of the DNA-binding domain from the androgen receptor are required. For specific and high-affinity binding to response elements, the DNA-binding domains of the androgen and glucocorticoid receptors need a different C-terminal extension. The glucocorticoid receptor requires 12 C-terminal amino acids for high affinity DNA binding, while the androgen receptor only involves four residues. However, for specific recognition of the PB-ARE-2, the androgen receptor also requires 12 C-terminal residues. Our data demonstrate that the mechanism by which the androgen receptor binds selectively to the PB-ARE-2 is different from that used by the glucocorticoid receptor to bind a consensus response element. We would like to suggest that the androgen receptor recognizes response elements as a direct repeat rather than the classical inverted repeat.

Interaction of the putative androgen receptor-specific coactivator ARA70/ELE1alpha with multiple steroid receptors and identification of an internally deleted ELE1beta isoform.

Steroid-regulated gene transcription requires the coordinate physical and functional interaction of hormone receptors, basal transcription factors, and transcriptional coactivators. In this context ARA70, previously called RFG and ELE1, has been described as a putative coactivator that specifically enhances the activity of the androgen receptor (AR) but not that of the glucocorticoid receptor (GR), the progesterone receptor, or the estrogen receptor (ER). Here we describe the cloning of the cDNA for ELE1/ARA70 by RT-PCR from RNA derived from different cell lines (HeLa, DU-145, and LNCaP). In accordance with the previously described sequence, we obtained a 1845-bp PCR product for the HeLa and the LNCaP RNA. Starting from T-47D RNA, however, an 860-bp PCR product was obtained. This shorter variant results from an internal 985-bp deletion and is called ELE1beta; accordingly, the longer isoform is referred to as ELE1alpha. The deduced amino acid sequence of ELE1alpha, but not that of ELE1beta, differs at specific positions from the one previously published by others, suggesting that these two proteins are encoded by different nonallelic genes. ELE1alpha is expressed in the three prostate-derived cell lines examined (PC-3, DU-145, and LNCaP), and this expression is not altered by androgen treatment. Of all rat tissues examined, ELE1alpha expression is highest in the testis. This is also the only tissue in which we could demonstrate ELE1beta expression. Both ELE1alpha and ELE1beta interact in vitro with the AR, but also with the GR and the ER, in a ligand-independent way. Overexpression of either ELE1 isoform in DU-145, HeLa, or COS cells had only minor effects on the transcriptional activity of the human AR. ELE1alpha has no intrinsic transcription activation domain or histone acetyltransferase activity, but it does interact with another histone acetyltransferase, p/CAF, and the basal transcription factor TFIIB. The interaction with the AR occurs through the ligand-binding domain and involves the region corresponding to the predicted helix 3. Mutation in this domain of leucine 712 to arginine greatly reduces the affinity of the AR for ELE1alpha but has only moderate effects on its transcriptional activity. Taken together, we have identified two isoforms of the putative coactivator ARA70/ELE1 that may act as a bridging factor between steroid receptors and components of the transcription initiation complex but which lack some fundamental properties of a classic nuclear receptor coactivator. Further experiments will be required to highlight the in vivo role of ELE1 in nuclear receptor functioning.

Primary rat lacrimal cells undergo acinar-like morphogenesis on reconstituted basement membrane and express secretory component under androgen stimulation.

Single cells or small cell clusters, isolated from the rat lacrimal gland, were incubated on reconstituted basement membrane (matrigel) in a well-defined serum-free medium. During the first days of culture, cells reassociated and reorganized in structures resembling acini. These multicellular structures, maintained in culture for 2 weeks, consisted of well-polarized cuboidal cells surrounding a central lumen and exhibiting apically located microvilli. Myoepithelial cells were observed at the periphery of the acinar structures. Both in the native lacrimal and in the cultured aggregates, epithelial cells displayed strong immunoreactivity for cytokeratin 8, while myoepithelial cells were immunoreactive for vimentin and alpha-smooth muscle isoactin. These data indicate that the cultured aggregates closely mimic the in vivo architecture of lacrimal glands both by morphology and immunohistochemistry. We further demonstrated the presence of an intact androgen receptor and the ability of the cultured aggregates to respond to androgens with increased secretion of the secretory component. Comparable androgen responses were observed in lacrimal gland cultures of 5-week-old male and female rats. In conclusion, we report a morphologically and functionally differentiated culture system of primary rat lacrimal cells, in which androgen-regulated gene expression was observed. This culture model provides a unique experimental paradigm for studying the effects of hormones, cytokines, and growth factors on the morphogenesis, growth, and functional differentiation of lacrimal glands.

Ubiquitous transcription factors NF1 and Sp1 are involved in the androgen activation of the mouse vas deferens protein promoter.

Transcription of the mouse vas deferens protein (MVDP) gene is stimulated by androgens and we have previously shown that in a 162 bp fragment, located at position -121 to +41, a TGAAGTtccTGTTCT sequence functions as an androgen-dependent enhancer. To determine which factors are involved in the hormonally regulated MVDP gene transcription, we have used DNase I footprinting and band-shift assays to examine in vitro binding of proteins to the enhancer and promoter sequences and have determined the functional significance of the recognized DNA sequences in transient transfection assays. Studies using recombinant proteins such as the DNA binding domain of the androgen receptor (AR-DBD) and Sp1, and crude cellular extracts from T47D and vas deferens epithelial cells (VDEC) showed that in addition to AR-DBD, the transcriptional activators NF1 and Sp1 interact with the -121/+41 fragment in a specific manner. Transient transfection assays revealed that site-directed mutations in the NF1 and Sp1 binding elements strongly reduced (NF1) or abolished (Sp1) androgen induced expression. The results demonstrate that the -121/+41 sequence is a composite site for the androgen receptor mediated transactivation of the MVDP gene.

Identification of a functional androgen-response element in the exon 1-coding sequence of the cystatin-related protein gene crp2.

Two hormone-responsive segments, one in the region of the promoter and one in intron 1, are identified in two homologous androgen-regulated and differentially expressed rat genes encoding the cystatin-related proteins (CRPs). Footprint analysis with the androgen receptor (AR) DNA-binding domain on the promoter-containing fragments reveals an AR-binding site downstream of the transcription start point in the crp2 gene (ARBSd/crp2, +40/+63). It displays an androgen response element-like sequence motif 5'-AGAAGAaaaTGTACA-3' and overlaps with the ATG translation start codon. A double-stranded oligonucleotide containing this sequence forms a DNA-protein complex with the full-length AR synthesized by vaccinia, as seen in band shift assays. Additional AR-binding sites, ARBSu/crp1 and ARBSu/crp2, occur 5' upstream of the transcription start point and are located at an identical position (-142/ -120) in crp1 and crp2. The AR affinity for these two slightly different sequence motifs is relatively weak. The biological function of all three AR-binding sites as transcription control elements has been studied. The ARBSd/crp2 element clearly shows androgen-response element characteristics. The contribution of the common upstream element to the androgen-dependent control of reporter gene transcription is less clear. The transcription of a reporter gene construct containing the crp2 footprint fragment crp2F (-273/+88) is hormonally regulated as determined by transfection into the human breast cancer cell line T-47D. Androgens, but also glucocorticoids, efficiently stimulate steroid-dependent transcription of the chloramphenicol acetyltransferase gene. Mutation of the 5'-TGTACA-3' sequence in ARBSd/crp2 destroys the AR binding and abolishes the androgen-dependent synthesis of chloramphenicol acetyltransferase. A large fragment derived from intron 1 of the crp1 and crp2 gene can also provide the androgen-dependent transcription of chimeric constructs in T-47D cells. However, the induction measured is less than the one observed with crp2F (-273/+88), and this activity seems to reside in several subfragments that each display a low but consistent androgen responsiveness.

Characterization of an androgen response element within the promoter of the epididymis-specific murine glutathione peroxidase 5 gene.

We have shown in earlier studies, using a mouse model, that the expression of the glutathione peroxidase 5 protein (GPX5) is restricted to the epididymis and that the accumulation of its corresponding mRNA is hormonally, spatially and temporally regulated throughout postnatal development. We report here, using run-on assays, transient expression experiments as well as gel-shift and footprinting analyses on the findings that at least part of the androgenic control of the GPX5 expression is exerted at the transcriptional level via an androgen response element localized in the distal promoter region of the GPX5 gene. The gpx5 androgen response element (ARE) is found to be consistent with the consensus palindromic steroid-receptor target sequence 5'-AGWACWnnnTGTYCT-3' but exhibits a quite weak conservation in the left half site. The data presented here further expand the diversity of sequence able to confer androgen responsiveness.