Competitive and cooperative functioning of the anterior and posterior promoter elements of an Alu family repeat.

Similar to tRNA genes and the VAI gene, the Alu family repeats are transcribed by RNA polymerase III and contain a split intragenic promoter. Results of our previous studies have shown that when the anterior, box A-containing promoter element (5'-Pu-Pu-Py-N-N-Pu-Pu-Py-G-G-3' in which Pu is any purine, Py is any pyrimidine, and N is any nucleotide) of a human Alu family repeat is deleted, the remaining box B-containing promoter element (5'-G-A/T-T-C-Pu-A-N-N-C-3') is still capable of directing weak transcriptional initiation at approximately 70 base pairs (bp) upstream from the box B sequence. This is different from the tRNA genes in which the box A-containing promoter element plays the major role in the positioning of the transcriptional initiation site(s). To account for this difference, we first carried out competition experiments in which we show that the posterior element of the Alu repeat competes with the VAI gene effectively for the transcription factor C in HeLa cell extracts. We then constructed a series of contraction and expansion mutants of the Alu repeat promoter in which the spacing between boxes A and B was systematically varied by molecular cloning. In vitro transcription of these clones in HeLa cell extracts was analyzed by RNA gel electrophoresis and primer extension mapping. We show that when the box A and box B promoter sequences are separated by 47 to 298 bp, the transcriptional initiation sites remain 4 to 5 bp upstream from box A. However, this positioning function by the box A-containing promoter element was lost when the spacing was shortened to only 26 bp or increased to longer than 600 bp. Instead, transcriptional initiation occurred approximately 70 bp upstream from box B, similar to that in the clones containing only the box B promoter element. All the mutant clones were transcribed less efficiently than was the wild type. An increase in the distance between boxes A and B also activated a second box A-like element within the Alu family repeat. We compare these results with the results of tRNA gene studies. We also discuss this comparison in terms of the positioning function of the split class III promoter elements and the evolutionary conservation of the spacing between the two promoter elements for optimum transcriptional efficiency.

Roles of fetal G gamma-globin promoter elements and the adult beta-globin 3' enhancer in the stage-specific expression of globin genes.

The human fetal G gamma-globin and adult beta-globin genes are expressed in a tissue- and developmental stage-specific pattern in transgenic mice: the G gamma gene in embryonic cells and the beta gene in fetal and adult erythroid cells. Several of the cis-acting DNA sequences thought to be responsible for these patterns of expression are located 5' to the G gamma-globin gene and 3' to the beta-globin gene. To further define the locations and functional roles of these elements, we examined the effects of 5' truncations on the expression of the G gamma-globin gene, as well as the ability of G gamma-globin upstream sequences to alter the developmental regulation of a beta-globin gene, as well as the ability of G gamma-globin upstream sequences to alter the developmental regulation of a beta-globin gene. We found that sequences between -201 and -136 are essential for expression of the G gamma-globin gene, whereas those upstream of -201 have little effect on the level or tissue or stage specificity of G gamma-globin expression. The G gamma-globin upstream sequences from -201 to -136 were, furthermore, capable of activating a linked beta-globin gene in embryonic blood cells; however, a G gamma-globin fragment from -383 to -206 was similarly active in this assay, and the complete fragment from -383 to -136 was considerably more active than either of the smaller fragments, suggesting the presence of multiple cis-acting elements for embryonic blood cells. Our data also suggested the possibility of a negative regulatory element between -201 and -136. These results are discussed in relation to several DNA elements in the G gamma-globin upstream region, which have been shown to bind nuclear factors in erythroid cells. Finally, we observed that removal of the beta-globin 3'-flanking sequences, including the 3' enhancer, from the G gamma-globin upstream-beta-globin hybrid gene resulted in a 25-fold reduction in expression in embryonic blood cells. This suggests that the beta-globin 3' enhancer is potentially active at the embryonic stage and thus cannot be solely responsible for the fetal or adult specificity of the beta-globin gene.

Prostate cancer progression, metastasis, and gene expression in transgenic mice.

We previously reported that a transgenic mouse line containing the fetal globin promoter linked to the SV40 T antigen (T Ag) viral oncogene (Ggamma/T-15) resulted in prostate tumors. In this study, we further explored tumor origin, frequency, invasiveness, androgen sensitivity, and gene expression pattern. T Ag was detected in adult but not fetal and neonatal prostates, suggesting a role for androgens in tumor progression. However, castration shortly after prostate morphogenesis did not prevent tumor development, suggesting an androgen-independent phenotype. Tumors originated within ventral or dorsal prostate lobes and involved intraepithelial neoplasia, rapid growth in the pelvic region, and metastasis to lymph nodes and distant sites. In addition, the primary cancers could be propagated in nude mice or nontransgenic mice. Seventy-five percent of hemizygous and 100% of homozygous transgenic males developed prostate tumors, suggesting a T Ag dosage effect. Biochemical characterization of advanced tumors revealed markers of both neuroendocrine and epithelial phenotypes; markers of terminal differentiation are lost early in tumorigenesis. Tumor suppressor genes (p53 and Rb), normally bound to T Ag, were up-regulated; bcl-2 proto-oncogene, which prevents apoptosis, was slightly up-regulated. Myc, a stimulus to cell cycle progression, was unchanged. We propose the Ggamma/T-15 transgenic line as a model of highly aggressive androgen-independent metastatic prostate carcinoma with features similar to end-stage prostate cancer in humans.

The microRNA-23b/-27b cluster suppresses prostate cancer metastasis via Huntingtin-interacting protein 1-related.

Deregulation of microRNAs (miRs) contributes to progression and metastasis of prostate and other cancers. miR-23b and -27b, encoded in the same miR cluster (miR-23b/-27b), are downregulated in human metastatic prostate cancer compared with primary tumors and benign tissue. Expression of miR-23b/-27b decreases prostate cancer cell migration, invasion and results in anoikis resistance. Conversely, antagomiR-mediated miR-23b and -27b silencing produces the opposite result in a more indolent prostate cancer cell line. However, neither miR-23b/-27b expression or inhibition impacts prostate cancer cell proliferation suggesting that miR-23b/-27b selectively suppresses metastasis. To examine the effects of miR-23b/-27b on prostate cancer metastasis in vivo, orthotopic prostate xenografts were established using aggressive prostate cancer cells transduced with miR-23b/-27b or non-targeting control miRNA. Although primary tumor formation was similar between miR-23b/-27b-transduced cells and controls, miR-23b/-27b expression in prostate cancer cells decreased seminal vesicle invasion and distant metastases. Gene-expression profiling identified the endocytic adaptor, Huntingtin-interacting protein 1-related (HIP1R) as being downregulated by miR-23b/-27b. Increased HIP1R expression in prostate cancer cells inversely phenocopied the effects of miR-23b/-27b overexpression on migration, invasion and anchorage-independent growth. HIP1R rescued miR-23b/-27b-mediated repression of migration in prostate cancer cells. HIP1R mRNA levels were decreased in seminal vesicle tissue from mice bearing miR-23b/-27b-transduced prostate cancer cell xenografts compared with scrambled controls, suggesting HIP1R is a key functional target of miR-23b/-27b. In addition, depletion of HIP1R led to a more rounded, less mesenchymal-like cell morphology, consistent with decreased metastatic properties. Together, these data demonstrate that the miR-23b/-27b cluster functions as a metastasis-suppressor by decreasing HIP1R levels in pre-clinical models of prostate cancer.

End-to-end transcription of an Alu family repeat. A new type of polymerase-III-dependent terminator and its evolutionary implication.

Four or more consecutive thymidine residues on the non-template strand and G + C-richness of flanking DNA are the two necessary characteristics of efficient RNA polymerase-III-dependent transcriptional terminators. We have identified, from the study of in vitro transcription of a human Alu family repeat, a new type of RNA polymerase-III-dependent transcriptional terminator. A 258 base-pair Alu repeat located on the 3' side of the human alpha 1 globin gene can be transcribed in a HeLa S-100 extract to generate three RNA species of lengths 404 to 408, 252 to 255 and 173 to 174 nucleotides, respectively. Kinetics, pulse-chase and RNA incubation experiments showed no significant internal processing of the longer transcripts into shorter ones. These data plus detailed RNA mapping demonstrated conclusively that the multiple Alu RNA species resulted from accurate initiation at the first base (5' end) of the repeat, and multiple termination downstream. The 3' end(s) of the major transcript (252 to 255 nucleotides) maps at the 3' end of the Alu repeat sequence where there are not four or more consecutive thymidine residues on the non-template strand. The functional domain of the terminator has been mapped to a 45 base-pair segment that includes 36 base-pairs of the 3' end sequence of the Alu repeat plus nine base-pairs downstream. The high efficiency of termination (greater than 90%), the lack of consecutive T residues, the richness in A + T content, and the proposed ability of the RNA to form an imperfect hairpin structure in the 3' region of the transcript, thus identify a new type of eukaryotic class III terminator. We compare the structure of this class III terminator with that of the bacterial rho-dependent terminator. We also discuss its implication in the mechanism(s) of amplification and dispersion of Alu sequences in the primate genomes.

MutT Homolog 1 (MTH1) maintains multiple KRAS-driven pro-malignant pathways.

Oncogenic RAS promotes production of reactive oxygen species (ROS), which mediate pro-malignant signaling but can also trigger DNA damage-induced tumor suppression. Thus RAS-driven tumor cells require redox-protective mechanisms to mitigate the damaging aspects of ROS. Here, we show that MutT Homolog 1 (MTH1), the mammalian 8-oxodGTPase that sanitizes oxidative damage in the nucleotide pool, is important for maintaining several KRAS-driven pro-malignant traits in a nonsmall cell lung carcinoma (NSCLC) model. MTH1 suppression in KRAS-mutant NSCLC cells impairs proliferation and xenograft tumor formation. Furthermore, MTH1 levels modulate KRAS-induced transformation of immortalized lung epithelial cells. MTH1 expression is upregulated by oncogenic KRAS and correlates positively with high KRAS levels in NSCLC human tumors. At a molecular level, in p53-competent KRAS-mutant cells, MTH1 loss provokes DNA damage and induction of oncogene-induced senescence. In p53-nonfunctional KRAS-mutant cells, MTH1 suppression does not produce DNA damage but reduces proliferation and leads to an adaptive decrease in KRAS expression levels. Thus, MTH1 not only enables evasion of oxidative DNA damage and its consequences, but can also function as a molecular rheostat for maintaining oncogene expression at optimal levels. Accordingly, our results indicate MTH1 is a novel and critical component of oncogenic KRAS-associated malignancy and its inhibition is likely to yield significant tumor-suppressive outcomes in KRAS-driven tumors.

Hepatocyte growth factor and vitamin D cooperatively inhibit androgen-unresponsive prostate cancer cell lines.

Expression of MET, the receptor for hepatocyte growth factor (HGF), has been associated with androgen-insensitive prostate cancer. In this study we evaluated MET activation by HGF and HGF action in prostate cancer cell lines. HGF causes phosphorylation (activation) of the MET receptor in three androgen-unresponsive cell lines (DU 145, PC-3, and ALVA-31) together with morphological change. Although HGF is known to stimulate the growth of normal epithelial cells, including those from prostate, we found that HGF inhibited ALVA-31 and DU 145 (hormone-refractory) cell lines. Moreover, HGF and vitamin D additively inhibited growth in each androgen-unresponsive cell line, with the greatest growth inhibition in ALVA-31 cells. Further studies in ALVA-31 cells revealed distinct cooperative actions of HGF and vitamin D. In contrast to the accumulation of cells in G1 seen during vitamin D inhibition of androgen-responsive cells (LNCaP), growth inhibition of the androgen-unresponsive ALVA-31 cell line with the HGF and vitamin D combination decreased, rather than increased, the fraction of cells in G1, with a corresponding increase in the later cell cycle phases. This cell cycle redistribution suggests that in androgen-unresponsive prostate cancer cells, HGF and vitamin D act together to slow cell cycle progression via control at sites beyond the G1/S checkpoint, the major regulatory locus of growth control in androgen-sensitive prostate cells.

Cooperative actions of hepatocyte growth factor and 1,25-dihydroxyvitamin D3 in osteoblastic differentiation of human vertebral bone marrow stromal cells.

Bone formation and remodeling require continuous generation of osteoprogenitor cells from bone marrow stromal cells (MSC), which generate and respond to a variety of growth factors with putative roles in hematopoiesis and mesenchymal differentiation. In this study we examine the interaction of two such factors on the maturation of skeletal components. We previously reported that these factors, hepatocyte growth factor (HGF) and 1,25-dihydroxyvitamin D(3) (vitD(3)), act together to increase alkaline phosphatase in chondroblasts. We now describe the cooperative effect of these agents on MSC isolated and cultured from human vertebral bone marrow. MSC (passages 3-9) isolated from bone marrow cells of human vertebrae (T1-L5) from 22-36-year-old normal donors were first expanded in vitro and then plated in the presence or absence of 10 ng/mL HGF and/or 10 nmol/L vitD(3), for 7-18 days. HGF treatment increased cell proliferation 2.5-fold, with no effect on alkaline phosphatase activity. Whereas vitD(3) treatment inhibited cell growth by 50%, alkaline phosphatase activity was stimulated eightfold, although no mineralization was observed. HGF together with vitD(3) increased cell proliferation 1.5-fold and alkaline phosphatase activity 13-fold over untreated control. Moreover, mineralization was detected only with this combination. Our findings provide evidence that HGF in concert with vitamin D may promote growth and differentiation of human MSC into osteogenic cells.

Absence of androgen-mediated transcriptional effects in osteoblastic cells despite presence of androgen receptors.

Androgen excess and deficiency affect skeletal maturation and bone cell function. Understanding the molecular basis for these androgen effects could improve therapy/prevention of short stature and osteoporosis. Androgens act through binding to androgen receptors (ARs), which modulate gene transcription via interactions with DNA response elements on target genes. Because osteoblasts contain ARs at levels just below certain androgen-sensitive tissues, we sought to define the function of AR in a number of commonly used osteoblastic cell lines. Presence and quantification of AR protein and mRNA were evaluated by ligand binding assay, western blotting, and RNAse protection assay. AR-containing osteoblastic cell lines were exposed to nonaromatizable androgens and effects on gene expression were assessed. We found no evidence for direct effects of androgen on endogenous genes nor was androgen involved in modulation of parathyroid hormone effects on early gene activation. Androgen-sensitive reporter gene constructs were stimulated by androgen only when AR cDNA expression vectors were introduced into cells by cotransfection. We conclude that, in commonly used osteoblastic cell lines, the presence of AR at the levels described here does not guarantee androgen transcriptional activity. The effects of androgen on bone in vivo may involve direct stimulation of osteoblastic cells in a different setting or stage of differentiation. Alternatively, androgen may act on bone cells other than osteoblasts, or through metabolic conversion to estrogens.

A role for GATA transcription factors in the androgen regulation of the prostate-specific antigen gene enhancer.

The transgenic mouse line Ggamma/T-15 containing the fetal globin promoter linked to SV40 T antigen unexpectedly results in androgen-independent prostate carcinomas. Given the key role of GATA-1 transcription factor in fetal globin gene promoter activity, we investigated whether specific GATA family members are expressed in the prostate and whether they can regulate prostate-specific genes. We found that GATA-2 and -3 are the predominant GATA family members expressed in human and mouse prostate and that GATA mRNA levels are not regulated by androgen. We identified six GATA sites flanking an androgen-response element located in the far-upstream enhancer of the prostate-specific antigen (PSA) gene. These GATA sites are targets for GATA factors and are essential for optimal androgen induction of transfected PSA enhancer/promoter plasmids in LNCaP, a PSA and androgen receptor expressing human prostate cancer cell line. Our results suggest that prostatic GATA-2 and -3 are involved in the androgen regulation of the PSA gene.

Id-1B, an alternatively spliced isoform of the inhibitor of differentiation-1, impairs cancer cell malignancy through inhibition of proliferation and angiogenesis.

Id-1 is a member of the helix-loop-helix family of proteins that regulates the activity of transcription factors to suppress cellular differentiation and to promote cell growth. Overexpression of Id-1 in tumor cells correlates with increased malignancy and resistance to chemotherapy and radiotherapy. Id-1B is an isoform generated by alternative splicing that differs from the classical Id-1 in the 13-C-terminal amino acids, whose function is at present unknown. We have studied the role of Id-1B in cancer and its expression in healthy/malignant lung tissues. Overexpression of Id-1B in A549 lung and PC3 prostate cancer cells reduced anchorage-dependent and independent proliferation and clonogenic potential. Moreover, it increased the proportion of cells in the G0/G1 phase of the cell cycle and p27 levels, while reduced phospho-Erk and cyclin A levels. Through microarray analysis, we identified genes involved in cell growth and proliferation that are specifically deregulated as a consequence of Id-1B overexpression, including IGF2, BMP4, Id2, GATA3, EREG and AREG. Id-1B overexpressing cells that were treated with 4Gy irradiation dose were significantly less resistant to cell death. In vivo assays demonstrated that tumors with high Id-1B levels exhibited less growth (p<0.01), metabolic activity (glucose uptake) and angiogenesis (p<0.05) compared to tumors with low Id-1B expression; mice survival was significantly extended (p<0.05). Quantification by qRT-PCR revealed that expression of Id-1B was significantly lower (p<0.01) in human lung tumors compared to their matched nonmalignant counterparts. In conclusion, our results demonstrate that Id-1B decreases the malignancy of lung and prostate cancer cells, sensitizes them to radiotherapy-induced cell death, and counteracts the protumorigenic role of the classical form of Id-1.

Estrogen and progesterone lower cyclin B1 AND D1 expression, block cell cycle in G2/M, and trigger apoptosis in human adrenal carcinoma cell cultures.

The effects of 17 beta-estradiol and progesterone were evaluated separately and in combination, on the growth, survival, and cell cycle dynamics of SW-13 human adrenal carcinoma cells in culture. Both hormones significantly decreased cell survival, with dose response curves at four days demonstrating EC (50)s estimated at 1.2 x 10 (-5) M for 17 beta-estradiol and 4.8 x 10 (-6) M for progesterone. Flow cytometry studies of these cultures indicated a strong G2/M blocking effect of both steroids, either individually or in combination; the effects of progesterone and of both agents together were substantially greater than the effect with 17 beta-estradiol alone. The sub-G1 region of the flow cytometry profile was significantly enhanced by exposure to 17 beta-estradiol and even more by progesterone. Sub-G1 "apoptosis" was confirmed by fragmented and condensed nuclear chromatin staining using a standard DAPI fluorescence assay. The expression of the critical cell cycle regulatory proteins cyclin B1 and D1 were significantly decreased by each hormone, with the influence of progesterone again predominating. These data demonstrate that high doses of 17 beta-estradiol and progesterone have inhibitory and apoptotic effects on SW-13 human adrenal carcinoma cells IN VITRO. The observed effects are associated with declines in cyclin B1 and D1 expression as well as a block in G2/M.

High dose 17 beta-estradiol and the alpha-estrogen agonist PPT trigger apoptosis in human adrenal carcinoma cells but the beta-estrogen agonist DPN does not.

Previous studies have shown that high dose 17beta-estradiol (10 (-5) M) has a G2/M blocking effect in SW-13 human adrenal carcinoma cultures and strongly enhances apoptosis. To examine the differential effects of estrogen alpha and beta-receptors in this system, we incubated SW-13 cells with specific alpha- and beta-estrogen receptor agonists, PPT [4,4',4''-(propyl-[ (1)H]-pyrazole-1,3,5-triyl) trisphenol] and DPN [2,3-bis (4-hydroxyphenyl) propionitrile], respectively (each at 10 (-5) M). Flow cytometry was used to analyze the percentages of cells in various phases of the cell cycle [sub-G1 (apoptosis), G1, S, and G2/M] in each experimental condition. Exposure to 17 beta-estradiol for 48 hours increased apoptosis more than 5-fold (from 3.6+/-0.5 to 20+/-2.2% of cells; p<0.01). The alpha-estrogen agonist PPT had a similar effect, increasing apoptosis to 22+/-1.7% (p<0.01), but the beta-agonist DPN caused no change (3.6+/-0.5 vs. 3.9+/-0.8%). While estrogen and the alpha-estrogen agonist decrease apoptosis in this system, both of these compounds decreased the percentage of cells in G1 (from 59+/-1.4% for control to 34+/-2.3% for estrogen and 40+/-2.0% for PPT; p<0.01 for both agents relative to control); the beta-agonist again had no effect. Estrogen was also found to block the cell cycle in G2/M, increasing it from 15+/-0.4 to 21+/-1.0% of cells (p<0.01), but neither the alpha- nor beta-estrogen agonists had any effect at this point in the cell cycle, indicating that the influence of estrogen was not likely to be either alpha- or beta-receptor mediated. There was no apparent effect of any of these agents on DNA synthesis, as indicated by unchanged percentages of cells in S phase. These studies suggest that induction of apoptosis by estrogen in SW-13 human adrenal cortical carcinoma cultures is mediated by the alpha-receptor, but the G2/M blocking effect of estrogen is not likely to be related to either alpha or beta mechanisms.

Distinct negative regulatory mechanisms involved in the repression of human embryonic epsilon- and fetal G gamma-globin genes in transgenic mice.

A current model for human beta-globin gene switching proposes that the stage-specific activation of embryonic and fetal globin genes requires the interaction of the beta-globin locus control region with proximal promoter elements. Subsequent repression in fetal and adult stages likely involves negative regulatory promoter elements and factors. To begin addressing these negative regulatory mechanisms, the regulation of human fetal G gamma-globin promoter fused to the SV40 T antigen gene was analyzed in transgenic mice. The results showed correct developmental expression in erythroid tissue, but lower levels of expression were also detected in non-erythroid tissue. Thus, the 5'-flanking G gamma-globin promoter sequence contains stage-specific erythroid elements but is lacking nonerythroid-specific negative elements. In contrast, the human embryonic epsilon-globin gene was only expressed in nonerythroid tissue of transgenic embryos, suggesting the presence of an erythroid-specific negative element(s). With the locus control region, complete repression of epsilon-globin RNA in fetal liver was observed in epsilon-globin genes without the previously characterized silencer, suggesting the presence of additional negative elements. Overall, this transgenic study suggests that distinct negative regulatory mechanisms function in the repression of embryonic and fetal globin genes.

Enrichment and depletion of Hela topoisomerase I recognition sites among specific types of DNA elements.

The SDS-induced nicking of DNA helix by Hela topoisomerase I in vitro has been studied by using 2.9 kb of cloned human DNA as the substrate. The frequency of nicking is increased from 1/23 (nick/nt) to 1/19 (nick/nt) when camptothecin is present in the nicking reaction. The cytotoxic drug also induces DNA nicks without the addition of SDS. Although the consensus built from DNA sequences from -20 to +20 of more than one hundred of the nicking sites only shows a preference for T at position -1, the distributions of the topoisomerae I-cleavable sites among different categories of specific DNA sequences are apparently non- random. Long stretches of tandem (CA), A, or T residues, and the GC-rich promoter region of alpha 1 globin gene are all refractory to the nicking reaction. However, the nicking frequencies of short direct repeats flanking different Alu type sequences are as high as 1/6 (nick/nt). Finally, several tandemly arranged minirepeats of the form (TxAy)z, that are usually found at the 3' ends of the primate Alu family or Kpnl family repeats, can be cleaved efficiently in a regular pattern by the enzyme. These data are discussed in terms of the mode of recognition of DNA sequences/structures by topoisomerase I, and its possible roles in the nonhomologous insertion of repetitive DNA sequences.

Distinctive sequence organization and functional programming of an Alu repeat promoter.

Plasmid clones containing a human Alu family repeat can be transcribed efficiently by RNA polymerase III in HeLa cell extract. This generated three RNA species, all of which initiated from the first base (+1) of the repeat. By studying the transcriptional properties of deletion clones, subclones, and topologically different DNA templates, we demonstrated that: supercoiled DNA templates are transcribed 3- to 5-fold more efficiently than are linear or nicked circular DNA molecules; a contiguous DNA helix in the transcription complexes that extends into the 5' flanking region of positions -30 to -85 is absolutely required for initiation to occur (this interaction does not involve recognition of specific DNA sequences); and similar to the adenovirus VAI RNA and tRNA genes, the Alu repeat 3' to the alpha 1-globin gene (designated 3'-alpha 1 Alu) contains a split intragenic promoter: an anterior element (positions +4 to +37) and a posterior element (positions +70 to +82). However, the promoter of the Alu repeat functions in distinctive ways in comparison to those of other RNA polymerase III-dependent genes. The posterior promoter element alone is sufficient and necessary for an accurate initiation to occur. The presence of the anterior promoter element, which by itself does not initiate transcription, enhances the transcriptional efficiency by a factor of 10- to 20-fold. Furthermore, the distance between the initiation sites and the posterior promoter element, but not the anterior promoter element, remains constant. These results suggest that the promoter of this Alu family repeat consists of at least two functionally different domains: a "directing element" (the posterior promoter element) that determines the accuracy of initiation and an "enhancing element" (the anterior promoter element) that is mainly responsible for the transcriptional efficiency.

Characterization of an unique RNA initiated immediately upstream from human alpha 1 globin gene in vivo and in vitro: polymerase II-dependence, tissue specificity, and subcellular location.

We have identified an abundant transcript initiated upstream from the canonical cap site of human alpha 1 globin gene in bone marrow cells and in COS-7 cells transfected with an alpha 1 globin gene-containing plasmid. Similar to the major alpha 1 globin transcript, this upstream RNA is present almost exclusively in the cytoplasm of the transfected COS-7 cells. It is also synthesized efficiently in vitro by RNA polymerase II in the nuclear extracts prepared from a Hela cell line and an erythroleukemia cell line, K562. RNAs isolated from these cell lines, however, do not contain this upstream transcript. The putative 5' end of the alpha 1 globin upstream RNA is mapped by primer extension to base -45, which is located in between the CCAAT and TATA boxes. The synthesis of this RNA in vitro and in vivo, and the close proximity of its 5' end to the promoter of the alpha 1 globin gene suggest a common mechanism regulating the transcriptional initiation of both the upstream and the major alpha 1 globin RNAs.

Prostate, adrenocortical, and brown adipose tumors in fetal globin/T antigen transgenic mice.

Targeted oncogenesis in transgenic mice has unexpectedly produced predictable tissue-specific tumors. We previously showed that hybrid gene constructs of the human fetal G gamma- or mouse embryonic beta h1-globin promoter linked to the viral simian virus 40 T antigen (G gamma/T and beta h1/T) expressed appropriately in embryonic erythroid tissue, with some unexpected expression elsewhere. Tumors arising in the G gamma/T and beta h1/T transgenic mice were identified by histology, electron microscopy, cell culture, and RNase protection analyses. In one G gamma/T transgenic line, males developed prostate tumors that showed mixed neuroendocrine and epithelial cell features, whereas females developed adrenocortical tumors. In several other G gamma/T lines, brown adipose tumors, or hibernomas, developed in the subcutaneous interscapular neck and shoulder area, as well as internally in the periadrenal and pericardial areas. Little or no expression of T antigen was detected in adult animals before visible tumor formation. In contrast, beta h1/T transgenic mice developed only choroid plexus tumors. Transient transfection assays in prostate and adrenocortical tumor-derived cell lines showed that the G gamma-globin promoter is 7-to 10-fold more active than the beta h1-globin promoter. Activity of 5' G gamma-globin promoter-deletion DNA plasmids was analyzed by transient transfection in a variety of human prostate cancer cell lines. The G gamma-globin promoter region between -140 and -201 also showed high activity in the androgen-independent human prostate cancer cell lines DU-145 and PPC-1, but low activity in the androgen-responsive human prostate cell line LNCaP. We conclude that tumor formation in the G gamma/T transgenic lines apparently results from cryptic positive DNA cis elements active in prostate and adrenocortical cells. Because G gamma-globin promoter activity is highest in embryonic tissue, tumors in adult transgenic mice may result from expression of T antigen in embryonic prostate, adrenal glands, and brown adipose tissue.

Suppression of human prostate cancer cell growth by forced expression of connexin genes.

The cell-to-cell channels in gap junctions, formed of proteins called connexins (Cxs), provide a direct intercellular pathway for the passage of small signaling molecules (< or = 1 kD) between the cytoplasmic interiors of adjoining cells. It has been proposed that alteration in the expression and function of Cxs may be one of the genetic changes involved in the initiation of neoplasia. To elucidate the role of Cxs in the pathogenesis of human prostate cancer (PCA), the pattern of expression of Cx alpha 1 (Cx43) and Cx beta 1 (Cx32) was studied by immunocytochemical analysis in normal prostate and in prostate tumors of different histological grades. While normal prostate epithelial cells expressed only Cx beta 1, both Cx alpha 1 and Cx beta 1 were detected in PCA cells. The Cxs were localized at the cell-cell contact areas in normal prostate and well-differentiated prostate tumors; however, as prostate tumors progressed to more undifferentiated stages, the Cxs were localized in the cytoplasm, followed by an eventual loss in advanced stages. Thus, epithelial cells from prostate tumors showed subtle and gross alterations with regard to expression of Cx alpha 1 and Cx beta 1 and their assembly into gap junctions during the progression of PCA. Retroviral-mediated transfer of Cx alpha 1 and Cx beta 1 into a Cx-deficient human PCA cell line, LNCaP, inhibited growth, retarded tumorigenicity, and induced differentiation, and these effects were contingent upon the formation of gap junctions. In addition, the capacity to form gap junctions in most Cx-transduced LNCaP cells was lost upon serial passage. Taken together, these findings indicate that the control of proliferation and differentiation of epithelial cells in prostate tumors may depend on the appropriate assembly of Cx beta 1 and Cx alpha 1 into gap junctions and that the development of PCA may involve the positive selection of cells with an impaired ability to form gap junctions.