Molecular cloning of human and rat complementary DNA encoding androgen receptors.

Complementary DNAs (cDNAs) encoding androgen receptors were obtained from human testis and rat ventral prostate cDNA libraries. The amino acid sequence deduced from the nucleotide sequences of the cDNAs indicated the presence of a cysteine-rich DNA-binding domain that is highly conserved in all steroid receptors. The human cDNA was transcribed and the RNA product was translated in cell-free systems to yield a 76-kilodalton protein. The protein was immunoprecipitable by human autoimmune antibodies to the androgen receptor. The protein bound androgens specifically and with high affinity.

p21 Disrupts the interaction between cdk2 and the E2F-p130 complex.

In nonproliferating or growth-arrested cells, the transcription factor E2F remains bound to the retinoblastoma-related protein p130. Accumulation of this E2F-p130 complex correlates with an arrest of the cell cycle progression. Progression through G1 phase is associated with a cyclin-dependent binding of the cyclin-dependent kinase cdk2 to the E2F-p130 complex. By fractionating mouse L-cell extracts, we have obtained a partially purified preparation of the E2F-p130 complex that also contains cdk2. Incubation of this complex with recombinant p21 results in a disruption of the interaction between cdk2 and the E2F-p130 complex in extracts of a cell line that expresses a temperature-sensitive mutant of p53. Incubation at the permissive temperature (32 degrees C) results in an induction of p21 synthesis. An increase in the level of p21 in these cells correlates with a loss of cdk2 from the cdk2-containing E2F-p130 complex. We also show that the expression of a reporter gene containing E2F sites in the promoter region is reduced by the coexpression of p21. Since p21 is believed to be a mediator of p53, we speculated that the p21-mediated disruption of the cdk2-containing E2F-p130 complex plays a role in the growth suppression function of p53.

Increased androgen receptor activity and altered c-myc expression in prostate cancer cells after long-term androgen deprivation.

Proliferation of LNCaP 104-S cells, a clonal subline of the human prostate cancer cell line, was very slow in androgen-depleted medium but increased 10-13-fold in the presence of 0.1 nM of a synthetic androgen, R1881. This induction of proliferation was diminished at higher concentrations of R1881, indicating the biphasic nature of the androgen effect. After 20-30 passages in androgen-depleted medium, these cells progressed to 104-I cells, which exhibited much lower proliferative sensitivity to 0.1 nM R1881. After another 20-30 passages, LNCaP 104-I cells gave rise to 104-R cells, which proliferated rapidly without additional androgen. Proliferation of 104-R cells was induced 2-fold by 0.01 nM R1881 but was repressed by 0.1 nM R1881 and above. Thus, androgen induction and repression of proliferation could be seen at lower concentrations of androgen as the cells progressed. During the transition of 104-S cells to 104-R cells, the androgen receptor mRNA level increased 2.5-fold whereas the androgen receptor protein level increased 15-fold in the absence of androgen. Androgen receptor transcriptional activity, measured by androgen induction of prostate-specific antigen mRNA and chloramphenicol acetyltransferase activity in transfected cells, increased up to 20-fold during the progression. LNCaP cells, therefore, appear to be able to adapt to reduced androgen availability by increasing their sensitivity to androgen, raising questions concerning the therapeutic strategies used against prostate cancer. Androgen induction of c-myc expression in 104-R cells occurred at a 10-fold lower concentration (0.01 nM) than in 104-S cells (0.1 nM). In all stages, cell proliferation and c-myc expression were repressed by androgen at a high concentration (20 nM), but the repression of cell proliferation was blocked by retroviral overexpression of c-myc.

A transcriptional activation function of p53 is dispensable for and inhibitory of its apoptotic function.

The tumor suppressor p53 is an inducer of cell cycle arrest and programmed cell death (apoptosis). The ability of p53 to induce cell cycle arrest is linked to its ability to induce transcription of genes such as the cyclin-dependent kinase inhibitor p21. However, the dependence of p53-mediated apoptosis on transcriptional activation remains controversial. Ectopic expression of a temperature-sensitive (ts) p53 allele induced expression of p53 target genes and elicited both G1 and G2/M cell cycle arrest upon shift to the permissive temperature. Ectopic expression of the same ts p53 allele with two additional point mutations (Gln22, Ser23) that abolish p53-transcriptional activation did not induce p53 target genes and G1 nor G2/M cell cycle arrest. In HCT116 colon carcinoma cells ectopic expression of wild type p53 does not elicit apoptosis whereas p53 mutant deficient in trans-activation induces apoptosis. The ability of wild type p53 to induce apoptosis is restored in HCT116 cells that are null for p21. However, the trans-activation deficient mutant of p53 is still more potent mediator of apoptosis than wild type p53 in the p21 null cells. Although the ability of Gln22,Ser23 to trans-activate p53 target genes is diminished, it retains the ability to repress Bcl-2 expression. Thus, we conclude that while ectopic expression of wild type p53 can induce both G1 and G2/M arrest, in a p21 dependent manner, without apoptosis, a p53 mutant defective in trans-activation elicits apoptosis without inducing cell cycle arrest. Further, the anti-apoptotic function of p53 is dependent on trans-activation and is linked to cell cycle arrest. The results strongly suggest that the trans-activation deficient mutant is a more potent inducer of apoptosis because it lost its anti-apoptotic function and retains its ability to repress pro-apoptotic genes such as Bcl-2. Taken together, the results imply that employing a trans-activation deficient p53 in gene therapy approaches or the use of drugs that convert mutant p53 to a trans-activation-independent mediator of apoptosis may be much more efficient therapeutic approaches than current approaches that employ wild type p53.

Progression of LNCaP prostate tumor cells during androgen deprivation: hormone-independent growth, repression of proliferation by androgen, and role for p27Kip1 in androgen-induced cell cycle arrest.

The molecular mechanism of androgen-independent growth of prostate cancer after androgen ablation was explored in LNCaP cells. An androgen-dependent clonal subline of the LNCaP human prostate carcinoma cell line, LNCaP 104-S, progressed to a slow growing stage (104-R1) and then to a faster growing stage (104-R2) during more than 2 yr of continuous culture in the absence of androgen. Androgen-induced proliferation of 104-S cells is inhibited by the antiandrogen Casodex, while proliferation of 104-R1 and 104-R2 cells is unaffected by Casodex. This indicates that proliferation of 104-R1 and 104-R2 cells is not supported by low levels of androgen in the culture medium. Compared with LNCaP 104-S cells, both 104-R1 and 104-R2 cells express higher basal levels of androgen receptor (AR), and proliferation of these two cell lines is paradoxically repressed by androgen. After continuous passage in androgen-containing medium, 104-R1 cells reverted back to an androgen-dependent phenotype. The mechanism of androgenic repression of 104-R1 and 104-R2 sublines was further evaluated by examining the role of critical regulatory factors involved in the control of cell cycle progression. At concentrations that repressed growth, androgen transiently induced the expression of the cyclin-dependent kinase (cdk) inhibitor p21waf1/cip1 in 104-R1 cells, while expression of the cdk inhibitor p27Kip1 was persistently induced by androgen in both 104-R1 and 104-R2 cells. Induced expression of murine p27Kip1 in 104-R2 cells resulted in G1 arrest. Specific immunoprecipitates of Cdk2 but not Cdk4 from androgen-treated 104-R1 cells contained both p21waf1/cip1 and p27Kip1. This observation was confirmed by in vitro assay of histone H1 and Rb (retinoblastoma protein) phosphorylation by the proteins associated with the immune complex. Furthermore, inhibition of Cdk2 activity correlated with the accumulation of p27Kip1 and not p21waf1/cip1. From these results we conclude that androgenic repression of LNCaP 104-R1 and 104-R2 cell proliferation is due to the induction of p27Kip1, which in turn inhibits Cdk2, a factor critical for cell cycle progression and proliferation.

Fusion proteins containing androgen receptor sequences and their use in the production of poly- and monoclonal anti-androgen receptor antibodies.

Complementary DNA segments that encode different domains of human and rat androgen receptors were fused to the Escherichia coli trpE gene using pATH expression vectors. Fusion proteins expressed by the bacteria were used to immunize rats and rabbits to obtain polyclonal antibodies to androgen receptors. Spleen cells of immunized rats were fused with myeloma cells to obtain stable hybridomas that produced monoclonal antibodies. Gradient centrifugation and immuno-precipitation assays indicated that the antibodies interacted with androgen receptors specifically.

Growth inhibition and regression of human prostate and breast tumors in athymic mice by tea epigallocatechin gallate.

The human prostate cancer cell lines, PC-3 (androgen-insensitive) and LNCaP 104-R (androgen-repressed) were inoculated subcutaneously into nude mice to produce prostate tumors. Intraperitoneal injection of green tea (-)epigallocatechin-3-gallate but not structurally related catechins, such as (-)epicatechin-3-gallate, inhibited the growth and rapidly reduced the size of human prostate tumors in nude mice. (-)Epigallocatechin-3-gallate also rapidly inhibited the growth of tumor growth formed by the human mammary cancer cell line MCF-7 in nude mice. It is possible that there is a relationship between the high consumption of green tea and the low incidence of prostate and breast cancers in some Asian countries.

Identification of a new member of the steroid receptor super-family by cloning and sequence analysis.

We have isolated a human testis complementary DNA clone on the basis of homology to the DNA binding domain of steroid receptors. Expression of this complementary DNA, which we call TR2, produces a 52 kd DNA binding protein that does not bind significantly to any known steroids. Northern blot analysis has shown that TR2 mRNA is about 2.5 kilobases (kb) and is relatively abundant in androgen-sensitive organs, such as ventral prostate and seminal vesicle. Dot blot hybridization indicates that TR2 mRNA levels increased after castration of rats, and this increase is reversed by 5 alpha-dihydrotestosterone injection. This evidence suggests that TR2 mRNA levels are negatively controlled by androgen in the rat ventral prostate.

Transcriptional activation by TR3 receptor, a member of the steroid receptor superfamily.

TR3 receptor is a human homolog of mouse Nurr77 and N10 protein and the rat NGFI-B protein. A cDNA encoding a chimeric nuclear receptor composed of the N-terminal domain and C-terminal putative ligand-binding domain of the orphan receptor TR3 receptor and the DNA-binding domain of the androgen receptor was constructed. The chimeric receptor, called TR3/AR/TR3 receptor, when expressed in COS-1 monkey kidney cells or PC-3 human prostate tumor cells, cotransfected with an ARE-containing mouse mammary tumor virus long terminal repeat-linked reporter gene encoding chloramphenicol acetyltransferase (CAT), activated CAT expression in the absence of any added factor. The activation was dependent on the amount of expression vector transfected and appeared to be independent of the concentration of serum supplement. Intact TR3 receptor was not active in this system. A TR3/AR/TR3 receptor protein truncated in the putative ligand-binding domain also induced CAT activity. TR3 receptor appears to be a transcriptional factor that activates transcription independently of ligand or binds an endogenous ligand present constitutively in cultured cells.

Spontaneous and induced mitotic recombination in Ustilago violacea detected at the cellular level.

Spontaneous and induced mitotic recombination in the heterobasidiomycete Ustilago violacea was detected at the cellular level using a sporidial morphology mutation. Mitotic recombination was induced by ultraviolet light (UV), nitrogen mustard (NM) and metabolically nonactivated cyclophosphamide (CP). The effects of low (14 °C) and high (30 °C) temperature and culture age on induced mitotic recombination are reported. Low temperature after inductive treatment uniformly reduced mitotic recombination. High temperature increased UV induced recombination, had no effect on NM-induced recombination and reduced CP-induced recombination to the spontaneous level. Temperature alone had no effect on mitotic recombination. Ultraviolet light-induced recombination was correlated with the rate of cell division and cell survival as cells passed from log to stationary phase growth. Detection of mitotic recombination at the cellular level is discussed as a method to assay postreplication repair of genetic damage and as a screen for agents which induce genetic damage in eukaryotic cells.

Expression and function of normal and LNCaP androgen receptors in androgen-insensitive human prostatic cancer cells. Altered hormone and antihormone specificity in gene transactivation.

The gene for the androgen receptor (AR) in the androgen-sensitive human prostate cancer cell line LNCaP has a single-base mutation that produces a threonine to alanine change in the androgen-binding domain. Androgen-insensitive prostatic cancer (PC-3) cells were cotransfected with an expression vector encoding normal, LNCaP, or chimeric normal/LNCaP AR and a vector carrying a chloramphenicol acetyltransferase (CAT) reporter gene linked to the mouse mammary tumor virus promoter. CAT activity was specifically induced by androgens in PC-3 cells expressing normal AR. In PC-3 cells expressing LNCaP AR, however, CAT activity was also induced by progestins and the antiandrogen hydroxyflutamide, which had little activity in cells expressing normal AR. Steroid-binding competition assays using in vitro synthesized ARs showed that LNCaP AR had a higher affinity than normal AR for progestins, 17 beta-estradiol, and hydroxyflutamide. The antiprogestin and antiglucocorticoid RU 38486 induced CAT activity in PC-3 cells expressing normal AR but not LNCaP AR. These studies indicate that AR mutations may be very important in determining the appropriate method of treatment with steroid hormones or their antagonists.

Structural analysis of complementary DNA and amino acid sequences of human and rat androgen receptors.

Structural analysis of cDNAs for human and rat androgen receptors (ARs) indicates that the amino-terminal regions of ARs are rich in oligo- and poly(amino acid) motifs as in some homeotic genes. The human AR has a long stretch of repeated glycines, whereas rat AR has a long stretch of glutamines. There is a considerable sequence similarity among ARs and the receptors for glucocorticoids, progestins, and mineralocorticoids within the steroid-binding domains. The cysteine-rich DNA-binding domains are well conserved. Translation of mRNA transcribed from AR cDNAs yielded 94- and 76-kDa proteins and smaller forms that bind to DNA and have high affinity toward androgens. These rat or human ARs were recognized by human autoantibodies to natural ARs. Molecular hybridization studies, using AR cDNAs as probes, indicated that the ventral prostate and other male accessory organs are rich in AR mRNA and that the production of AR mRNA in the target organs may be autoregulated by androgens.

Isolation and characterization of human TR3 receptor: a member of steroid receptor superfamily.

Complementary DNAs (cDNAs) encoding a member of steroid receptor super-family, named TR3 receptor, were isolated from a human prostate lambda gt11 cDNA library on the basis of homology of oligonucleotide probes to the DNA-binding domain common to members of the steroid receptor super-family. Expression of TR3 receptor cDNA produced a 64 kDa DNA-binding protein in a rabbit reticulocyte lysate. Nucleotide sequence analysis showed that TR3 receptor cDNA contains two regions of sequences which correspond to the DNA- and hormone-binding domains of members of the steroid receptor super-family. The amino acid sequences in the hormone-binding domain of the TR3 receptor shares about 20% homology with estrogen receptor and less than 15% homology with other known steroid receptors. The DNA-binding domain of the TR3 receptor has about 55% homology with all other known steroid receptors. TR3 receptor had 86% nucleotide and 91% amino acid sequence homology with mouse NUR/77, suggesting that TR3 receptor may be a human homologue of mouse NUR/77 gene product.

Myc-mediated apoptosis requires wild-type p53 in a manner independent of cell cycle arrest and the ability of p53 to induce p21waf1/cip1.

Deregulated expression of the c-myc proto-oncogene can lead to apoptosis under certain physiological conditions. By introducing a conditionally active Myc allele into primary embryo fibroblasts null for p53, and into fibroblasts without endogenous p53 expression but ectopically expressing a temperature-sensitive p53 allele, we show that expression of wild-type p53 is required for susceptibility to Myc-mediated apoptosis. Although ectopic expression of wild-type p53 blocked cells in the G1 phase of the cell cycle, G1 arrest by isoleucine starvation, in a manner independent of p53, did not confer susceptibility to apoptosis. Thus, growth arrest per se is not sufficient to induce Myc-mediated apoptosis; instead, a property intrinsic to p53 is specifically required. Moreover, apoptosis did not require induction of p53 target proteins, including the cyclin-dependent kinase inhibitor p21waf1/cip1. Therefore, the role of p53 in apoptosis may be distinct from its role in cell cycle arrest.

Illegitimate recombination induced by benzo[a]pyrene diol epoxide in Escherichia coli.

Duplex DNA oligomer constructs (32 base pairs) were prepared that contained a single benzo[a]pyrene (BP) adduct at a specific deoxyadenosine or deoxyguanosine site in either one or both strands. These constructs were inserted into M13 replicative form viral DNA, and the DNA from progeny virus generated by transfection of Escherichia coli was examined by sequence analysis at the site of oligomer insertion. With nonalkylated constructs, and with constructs containing only one BP adduct, no sequence alterations were found in progeny viral DNAs. With constructs containing two BP adducts, one in each strand and closely spaced, some progeny DNAs showed the original oligomer sequence, whereas others exhibited large deletions and illegitimate (nonhomologous) recombination, both of which removed the damaged construct. Increasing the distance between BP adducts in the construct reduced the frequency of recombinant events. These sequence alterations occurred in both recA+ and recA- host cells. We speculate that the closely spaced adducts in opposite construct strands cause a rare distortion in DNA structure, which activates the recombinant machinery, and that mutagenic and carcinogenic agents other than polycyclic aromatic hydrocarbons may cause similar DNA distortions, which induce illegitimate recombination.

The use of a DNA-binding domain replacement method for the detection of a potential TR3 orphan receptor response element in the mouse mammary tumor virus long terminal repeat.

TR3 orphan receptor is a human homologue of the mouse nur77, N10 and rat NGFI-B, TIS1 genes which may represent an early response gene involved in the control of cell proliferation. We have studied potential target genes for TR3 orphan receptor using the DNA-binding domain replacement method. We found that mouse mammary tumor virus long terminal repeat-linked chloramphenicol acetyltransferase expression can be activated in transfected cells by a chimeric androgen receptor/TR3 orphan receptor/androgen receptor construct (AR/TR3/AR) in the presence of androgen. By deletion analysis, a region with 20 nucleotides in length between positions -1178 and -1159 of the mouse mammary tumor virus long terminal repeat was confirmed as a potential TR3 orphan receptor response element. These results suggest that feasibility of using the DNA-binding domain replacement method to detect target sequences of orphan receptors.

Androgen receptors: structures, mutations, antibodies and cellular dynamics.

An overview of recent studies from this and other laboratories on the structures and intracellular dynamics of androgen receptors is presented. Human and rat androgen receptors are unique in that, aside from their DNA and androgen binding domains, they have amino terminal regions rich in oligo- and poly(amino acids) motifs as in some regulatory and homeotic genes. Point mutations that cause sequence changes or deletion of regions of androgen receptors appear to be responsible for some cases of androgen-insensitivity. Monoclonal antibodies produced against specific regions of the androgen receptor bind to androgen receptors but not other major steroid receptors. Androgen receptors in the human and rat prostate, and monkey seminal vesicle were localized to the nucleus of target cells in these tissues with these antibodies; androgen receptors also were found in the cytoplasm of some target cells. Actinomycin D and 3'-deoxyadenosine, inhibitors of transcription, RNA processing and nucleo-cytoplasmic transport of RNA, interfere with the intracellular dynamics of androgen receptors, suggesting as we have proposed previously that androgen receptors may function not only at the site of transcription but also are involved in posttranscriptional regulation of mRNA stability and utilization.

Ubiquitous receptor: a receptor that modulates gene activation by retinoic acid and thyroid hormone receptors.

The cDNA for a member of the nuclear receptor family was cloned and named ubiquitous receptor (UR), since UR protein and mRNA are detected in many cell types. Rat UR/human retinoid X receptor alpha (hRXR alpha) heterodimers bound preferentially to double-stranded oligonucleotide direct repeats having the consensus half-site sequence AGGTCA and 4-nt spacing (DR-4). Coexpression of UR in COS-1 cells inhibited the stimulation of chloramphenicol acetyltransferase (CAT) reporter gene expression by hRXR alpha and human retinoic acid receptor alpha in the presence of all-trans-retinoic acid when DR-4 (but not DR-5) was present upstream of the promoter of a CAT reporter gene (DR-4-CAT). UR expression also inhibited the activation of a DR-4-CAT reporter gene by hRXR alpha and 9-cis-retinoic acid or by thyroid hormone receptor beta in the presence of thyroid hormone. However, in the absence of 9-cis-retinoic acid, UR in combination with hRXR alpha stimulation DR-4-CAT expression. Coexpression of thyroid hormone receptor markedly reduced this stimulation in the absence of thyroid hormone. UR may play an important role in normal growth and differentiation by modulating gene activation in retinoic acid and thyroid hormone signaling pathways.

Molecular cloning of new human TR2 receptors: a class of steroid receptor with multiple ligand-binding domains.

Previously we isolated a new group of cDNA clones from human testis cDNA libraries which might code for new steroid receptors. The cDNA and predicted amino acid sequences of two of these receptors, named TR2-5 and TR2-7 receptors, were determined. We report here the nucleotide and deduced amino acid structures of two other receptors that we named TR2-9 and TR2-11 receptors. The calculated MW of TR2-5 receptor, TR2-7 receptor, TR2-9 receptor and TR2-11 receptor are 52,982, 20,528, 50,849 and 67,223 respectively, which match well with the apparent MW of in vitro translated products. The 26 amino acids involved in the formation of "Zn-fingers" are conserved. The ligand-binding domain of TR2-9 receptor is 16 amino acids shorter and has 3 different amino acids compared with TR2-5 receptor. The TR2-11 receptor has a ligand-binding domain which is longer and quite different compared with the other TR2 receptors. The multiple ligand-binding domains of TR2 receptor could be the products of different genes or may be due to RNA splicing errors. So far, we have failed to find binding activity with any known steroid hormone; this promotes the possibility that an unidentified steroid hormone may be involved.