Inhibition of histone deacetylase activity causes cell type-specific induction of the PDGF-B promoter only in the absence of activation by its enhancer.

There is a strong correlation between the acetylation status of nucleosomal histones and transcriptional activity. Here we show that the histone deacetylase inhibitor trichostatin A (TSA) activates reporter gene constructs driven by the human platelet-derived growth factor B (PDGF-B) gene promoter. This activation showed an inverse correlation with the cell type-specific transcriptional activities of the promoter. The TSA response was minimal in three tumor cell lines that exhibit high-level promoter activity. In JEG-3 choriocarcinoma cells, however, where the basal promoter activity is considerably lower, there was a strong response to TSA. This was in contrast to constructs that included a PDGF-B enhancer, which were refractory to TSA effects, indicating a possible function of the enhancer in modulating acetylation status. Analysis of PDGF-B promoter mutants with respect to TSA induction revealed no specific TSA-responsive element, but suggested that association of nonacetylated histones to the PDGF-B promoter may be a default process in the absence of enhancer activation. TSA treatment of JEG-3 cells, either alone or in combination with the demethylating agent 5-azacytidine, failed to activate the silenced endogenous PDGF-B transcript, however, which appears to be repressed by additional mechanisms.

The development of pharmacogenomic models to predict drug response.

The potential of pharmacogenomic research to improve general healthcare, reduce morbidity and mortality resulting from treatment side effects, and to accelerate therapeutic compound discovery, design and development in the pharmaceutical industry, remains largely unfulfilled. Major contributory factors affecting progress in the field include: (i) the need for large clinical populations and control/placebo-treated cohorts to be studied; (ii) the difficulty in evaluating drug response in many instances with empirical or qualitative treatment response values often not available; (iii) interactions of underlying biochemical pathways are often not fully understood, both in terms of mode-of-action of a therapeutic compound itself and in the occurrence of side effects; (iv) population-specific frequency data for reported genetic variants and physically mapped genomic markers for evaluation in drug response studies are often not readily and/or publicly available; (v) accurate, high-throughput, cost-effective polymorphism detection and screening methods are required; and (vi) sophisticated biostatistical data analysis programs to perform the multi-parametric tests and permutation analyses necessary are still under development. With the recent human genome sequence draft release by both the public and commercial initiatives, in addition to the publication of locus and chromosome-specific polymorphism maps and TSC (The SNP Consortium) SNP map information expected in late-2001, it is hoped that at least some of the inherent difficulties in pharmacogenomics research will soon be alleviated.

Pharmacogenomics to predict drug response.

From theory to proof-of-concept, pharmacogenomics promises to improve future general healthcare in a number of ways. By identifying individuals who will respond to a particular drug treatment compared to those who have a low probability of response, pharmacogenomic test development hopes to aid the physician in prescribing the optimal medication for each patient. This approach promises faster relief from symptoms, a lowering of side effect risks and a reduction in healthcare costs. Pharmacogenomic tests used by the pharmaceutical companies themselves can be used to help identify suitable subjects for clinical trials, aid in interpretation of clinical trial results, find new markets for current products and speed up the development of new treatments and therapies. This type of approach should also see fewer compounds failing during later phases of development. The questions we are faced with as we enter the new millennium, however, are if and when the promises of pharmacogenomnics in improving healthcare will be fulfilled. Currently, there are only a handful of pharmacogenomic tests and associated products which are commercially available and it remains to be seen what impact these will have on the market and on healthcare in general.

Random monoallelic expression of the imprinted IGF2 and H19 genes in the absence of discriminative parental marks.

The IGF2 and H19 genes are genomically imprinted and expressed preferentially from the paternal and maternal alleles, respectively, during human prenatal development. The exact role of the parental imprint(s), however, is not known. To explore this issue in some detail, we have examined human androgenetic cells which by definition should be incapable of allelic discrimination given the paternal origin of both genomes. Allele-specific in situ hybridisation analysis of dispermic complete hydatidiform moles shows that IGF2 and H19 can be found to be transcriptionally active in a variegated manner, which results in the generation of random monoallelic expression patterns. This data shows that imprinted genes can be expressed monoallelically in the absence of discriminating parental marks and raises the question whether or not mechanisms underlying monoallelic expression preceded the acquisition of parental imprints during evolution.

A novel type of regulatory element is required for promoter-specific activity of the PDGF-B intronic enhancer region.

We have previously described a non-classical, promoter-specific enhancer for the human Platelet-Derived Growth Factor B (PDGF-B) gene. In JEG-3 choriocarcinoma cells the activity of the enhancer depends upon co-operation with a sequence (the Enhancer-Dependent cis Co-activator "EDC" element) within the promoter. The PDGF-B enhancer fails to activate heterologous promoters, indicating that promoter-specificity depends on an element within the enhancer that can recognise a target sequence within the promoter. Here we identify a sequence within the enhancer of the PDGF-B gene which directs activation of the PDGF-B promoter by distal cis-acting elements. This specifies the wild-type PDGF-B promoter as the target for the enhancer and has been designated the EDC specificity element (EDCse). The cell-type specific nature of this interaction is extended by the observation that the EDCse is also dispensable for enhancer activity in breast-cancer cells (ZR-75). Concomitant to this observation, JEG-3 and ZR-75 cells differ in the binding of nuclear factors to the EDCse. We discuss the relevance of the EDC/EDCse system in regulation of gene expression.

Cell-type-specific modulation of PDGF-B regulatory elements via viral enhancer competition: a caveat for the use of reference plasmids in transient transfection assays.

The human platelet-derived growth factor-B (PDGF-B) gene has been shown to display a wide range of levels of mRNA transcription in a variety of cell types. Functional analyses of PDGF-B gene expression have begun to reveal intricate, cell-type-specific regulatory mechanisms involving multiple control elements. We have previously isolated and characterised several elements involved in the control of human PDGF-B gene expression in the JEG-3 choriocarcinoma cell line and in the breast cancer-derived cell line, ZR-75. Assessment of the positive or negative regulatory contributions of these elements was carried out using transient transfection assays. Such studies routinely require the inclusion of a reference plasmid in order to determine transfection efficiency. Here we show that competition for regulatory factors occurs in transfected cells between viral enhancers and elements regulating PDGF-B gene transcription. A frequently used reference plasmid which utilises the SV40 promoter and enhancer region to drive expression of a beta-galactosidase reporter gene was found to severely repress the activity of a co-transfected reporter construct containing the PDGF-B promoter and its intronic enhancer in JEG-3 cells. This competition was localised to the enhancer region of the SV40 regulatory sequences and surprisingly, the effect was reversed in ZR-75 cells; where increasing the amount of reference plasmid strongly stimulated the activity of the PDGF-B construct. These results imply that the same intronic region which functions equally well as an enhancer in two distinct cell-types, may operate in response to different transcription factor complements. Furthermore, this data demonstrates that the choice of reference plasmid and its quantitative use can be a crucial factor when examining putative regulatory elements by transient transfection methods.

Allele-specific in situ hybridization (ASISH) analysis: a novel technique which resolves differential allelic usage of H19 within the same cell lineage during human placental development.

Precursory studies of H19 transcription during human foetal development have demonstrated maternally derived monoallelic expression. Analyses in extra-embryonic tissues, however, have been more equivocal, with discernible levels of expression of the paternal allele of H19 documented in the first trimester placenta. By refining the in situ hybridization technique we have developed an assay to enable the functional imprinting status of H19 to be determined at the cellular level. This assay involves the use of oligonucleotide DNA probes that are able to discriminate between allelic RNA transcripts containing sequence polymorphisms. Biallelic expression of H19 is confined to a subpopulation of cells of the trophoblast lineage, the extravillous cytotrophoblast, while the mesenchymal stroma cells maintain the imprinted pattern of monoallelic expression of H19 throughout placental development. This data demonstrates that the low level of paternal H19 expression previously detected in normal human placenta is not due to a random loss of functional imprinting, but appears to result from a developmentally regulated cell type-specific activation of the paternal allele. In addition, biallelic expression of H19 does not seem to affect the functional imprinting of the insulin-like growth factor II gene, which is monoallelically expressed at relatively high levels in the extra-villous cytotrophoblasts. These results imply that the allelic usage of these two genes in normal human placental development may not be directly analogous to the situation previously documented in the mouse embryo.

Genomic imprinting and mammalian development.

Genomic imprinting, which results in the mono-allelic expression of certain genes in a parent of origin-dependent manner, represents a specialized form of gene regulation which may be vitally important for mammalian development. The mechanisms which underlie imprinting and the molecular nature of the imprint itself remain elusive but most likely include epigenetic modifications of DNA, such as methylation and chromatin structure changes. It is clear, however, that many of the known imprinted genes play important developmental roles and that changes in the functional imprinting of some of these genes may have important pathological consequences, including placental abnormalities.

An Inr-containing sequence flanking the TATA box of the human c-sis (PDGF-B) proto-oncogene promoter functions in cis as a co-activator for its intronic enhancer.

High-level activity of the human PDGF-B promoter in choriocarcinoma cell lines depends upon an atypical, intronic enhancer-like element which does not function with heterologous promoters tested. An extensive series of mutant PDGF-B promoter-driven constructs identified a sequence flanking the TATA box which is required specifically for enhancer-mediated transcription in human choriocarcinoma cell lines. This element, which we here term an enhancer-dependent cis co-activator (EDC) contains an Inr (initiator) consensus sequence upstream of the TATA box which is required, but not sufficient for its function. Requirement for the EDC is cell type-specific, since it was dispensable for enhancer-mediated transcription in a human breast cancer cell line. Although it lies within the region defined, the TATA box itself is not required for EDC function, or for basal promoter function which may derive from a second Inr-like sequence situated at the transcriptional start site. These observations indicate that interactions between some promoter and enhancer elements may be more complex than that generally described for 'classical' enhancer systems and may suggest an additional function for the initiator motif.

Expression of the human PDGF-B gene is regulated by both positively and negatively acting cell type-specific regulatory elements located in the first intron.

Potential cis-acting regulatory elements of the human platelet derived growth factor-B (PDGF-B) gene were identified by DNase I hypersensitive site mapping. The transcription unit was examined for the presence of hypersensitive sites in chromatin DNA isolated from human term placental cytotrophoblasts, human placental fibroblasts, the JEG-3 choriocarcinoma cell line and the U2-OS osteosarcoma cell line. A number of cell type-specific hypersensitive sites were identified, all within the 1st intron. Transient transfection of JEG-3 cells with CAT constructs containing regions of the c-sis 1st intron linked to the basal c-sis promoter identified a cell type-specific positive regulatory activity within the intron, composed of at least two distinct elements. One element appeared to be specific for JEG-3 cells, while the other was also active in U2-OS cells. The overall positive regulatory activity of the 1st intron region was specific for JEG-3 cells, but did not function as a classically defined enhancer, as it was orientation-dependent (unless stably integrated into chromatin DNA). In addition, the activator appears to require interaction with the c-sis promoter, as little or no activation was seen when either the SV40 or human beta-globin promoters were substituted for the c-sis promoter. The 1st intron also contained a negative regulatory element, which was specific for U2-OS cells and silenced an abnormally high basal c-sis promoter activity in these cells. The complexity of the transcriptional control of the PDGF-B gene is discussed.