The oppositely-imprinted genes insulin-like growth factor-II (IGF2) and H19, a putative tumor suppressor, often show coordinate, reciprocal regulation and are believed to play a role in carcinogenesis. To explore the possible interactions between these genes, we stably transfected diHepG2 cells with a plasmid containing either the sense or the antisense H19 cDNA sequences and verified their expression by Northern analysis and by RNase protection analysis. Levels of H19, IGF2 and g-actin mRNA were quantified by competitive RT-PCR analysis. Although H19 sense transgene overexpression (n 24 clones) did not decrease the low, basal levels of IGF2 mRNA compared to control cells, levels of IGF2 mRNA were positively correlated with the levels of H19 antisense mRNA (P , 0.0001, n 40 clones). Furthermore, the increase in IGF2 mRNA level was accompanied by an elevation of IGF-II peptide in conditioned media. To see if H19 mRNA had a specific effect on transcription, we also performed transient transfections with reporter gene constructs containing IGF2 promoter 3 in the presence of sense or antisense H19 cDNA sequences under control of a cytomegalovirus promoter. We show a lower reporter gene activity from reporter gene constructs in the presence of sense H19 cDNA than from those with antisense or neomycin. Our results suggest that H19 participates in the repression of IGF2, at least in part through effects on IGF2 transcription, an effect which may contribute to its action as a tumor suppressor.Keywords: insulin-like growth factor-II; gene regulation; H19; tumor suppressor genes.Human insulin-like growth factor (IGF-II) is an important mitogen which plays a role in normal fetal and postnatal growth, and also in tumorogenesis [1]. Regulation of the human IGF2 gene is very complex, both at the transcriptional and posttranscriptional level. Firstly, this gene is directed by four promoters which exhibit a tissue and development-specific expression pattern. Secondly, it includes sequence elements that are recognized by transcription factors (such as p53, WT-1 or C/EBPa), some of which are known to have tumor suppressor activity [2±4]. Thirdly, the minisatellite DNA polymorphism consisting of a variable number of tandem repeats (VNTR) at the human INS (insulin gene) 5 H -flanking region, has recently been shown to affect IGF2 transcription [5]. Finally, IGF-II mRNAs are subjected to differential polyadenylation, alternative splicing and site-specific endonucleolytic cleavage [4].IGF2 is also epigenetically regulated [6], so that only the paternal allele is expressed in most tissues prenatally and in some tissues during postnatal life, a phenomenon referred to as genomic imprinting. Preferential loss of maternal alleles (loss of heterozygosity) and relaxation of parental imprinting (loss of imprinting) of IGF2 occur frequently in certain pediatric tumors, such as Wilms' tumors, adrenocortical carcinomas, hepatoblastomas and rhabdomyosarcomas, and in malignant adulthood tumors (including uterine, lung and testicular tumors), sugg...

The actions of IGFs are regulated at various levels. One mechanism involves binding to IGF-binding protein-3 (IGFBP-3) for transport, thus governing bioavailability. IGFBP3 transcription is modulated by many hormones and agents that stimulate or inhibit growth. We have previously shown in pediatric and adult cohorts a correlation between IGFBP-3 serum levels and two single-nucleotide polymorphisms (SNPs) located within the minimal promoter (-202 A/C and -185 C/T). Functionality of these SNPs was further explored in hepatic adenocarcinoma-derived SK-HEP-1 cells using transient transfections of luciferase constructs driven by different haplotypes of the IGFBP3 promoter. Basal luciferase activity revealed a significant haplotype-dependent transcriptional activity (at nucleotides -202 and -185, AC > CC, P < 0.001; AC > CT, P < 0.001; AC > AT, P < 0.001). Insulin treatment produced a similar haplotype dependence of luciferase activity (AC > CC, P = 0.002; AC > CT, P < 0.001; AC > AT, P = 0.011). However, induction ratios (insulin/control) for CC and AT were significantly higher compared with AC and CT (CC > AC, P = 0.03; CC > CT, P = 0.03; AT > AC, P = 0.03; AT > CT, P = 0.04). Gel retardation assays were used to identify upstream stimulatory factor (USF-1 and USF-2) methylation-dependent binding to E-box motifs located between the SNPs. Mutation of the USF binding site resulted in a significant loss of insulin stimulation of luciferase activity in the transfection assay. Chromatin immunoprecipitation with anti-USF-1/-2 showed an enrichment of IGFBP3 promoter in insulin-treated cells compared with unstimulated cells. Bisulfite sequencing of genomic DNA revealed that CpG methylation in the region of USF binding was haplotype dependent. In summary, we report a methylation-dependent USF binding site influencing the basal and insulin-stimulated transcriptional activity of the IGFBP3 promoter.

SUMMARY:Taurine and β-amino uptake in cultured skin fibroblasts proceeds through at least two distinct amino acid transport systems. The predominant Bamino acid uptake system which we refer to as the “Beta” system, incorporates taurine in a proportion of 95%, β-alanine in a proportion of 80% and does not incorporate β-amino-isobutyric acid. A second transport system for β-alanine seems to be operative in cultured skin fibroblasts and this system shares the characteristics of system “L” for branched-chain and ringside neutral amino acids. Results of ion depletion experiments, metabolic inhibition by drugs and blocking agents and previous kinetic studies of taurine and β-alanine uptake in cultured skin fibroblasts failed to disclose any major difference in β-amino acid transport between control individuals and patients with Friedreich's ataxia.

Background: The mechanisms underlying the maintenance of normal to high rates of linear growth and plasma insulin-like growth factor I (IGF-I) levels in spite of a low growth hormone secretion in obese children remain unknown. Among the animal models of early-onset obesity, obese Zucker (fa/fa) rats (which are homozygous for an inactivating missense mutation in the leptin receptor) are particularly appropriate, because their linear growth shows this growth hormone independence. Methods: To study the regulation of IGF-I synthesis in this model, we have established primary cultures of hepatocytes derived from 12-week-old Zucker male obese and lean rats. The rat IGF-I gene contains six exons, and alternative splicing generates different mRNAs, one of which (called IGF-1B) has been shown to be decreased by fasting. We report steady state mRNA levels for IGF-I (all transcripts) and for IGF-IB in hepatocytes after 3 days in culture, in freshly isolated hepatocytes, and in whole-liver tissue. RT-PCRs using primers specific for IGF-I or IGF-IB were performed with two different internal competitors for quantification. Results: In primary cultures of hepatocytes, the IGF-IB mRNA was increased by >50-fold (p = 0.01) in cells derived from obese animals as compared with cells from lean animals. However, these transcript levels were not significantly different when measured in freshly isolated hepatocytes or in whole-liver tissue. Conclusions: Increased IGF-IB transcription could be an intrinsic characteristic of cultured hepatocytes harbouring leptin receptors that bear the fa mutation. However, the modulation of this characteristic by cell-cell interactions and by in vivo hormone and metabolic status remains to be studied.