Tissue expression and serum levels of periostin during pregnancy: a new biomarker of embryo-endometrial cross talk at implantation.

OBJECTIVE: The molecular aspects involved in human implantation include many elements that were first discovered due to their role in cancer cell metastasis. Periostin, a cell adhesion protein that allows the maintenance of cancer stem cells, may influence implantation. The objective of this experimental case-control study was to investigate tissue and serum expression of periostin during pregnancy, and evaluate the potential role of periostin in endometrial receptivity and embryo implantation. STUDY DESIGN: Forty-five subjects were included in the final analysis: 15 women who had experienced spontaneous pregnancy loss were enrolled as cases, and 30 healthy pregnant women awaiting voluntary pregnancy termination were enrolled as controls. For both cases and controls, trophoblastic and decidual tissues were collected at 12 weeks of gestation. Periostin expression in decidual and trophoblastic tissues was evaluated by immunohistochemical staining and reverse transcription polymerase chain reaction in cases and controls, and periostin serum levels was analyzed by Western blotting assays in cases, controls and non-pregnant female volunteers. RESULTS: Periostin mRNA and protein levels were higher in decidual and trophoblastic tissues from women undergoing voluntary pregnancy termination compared with women who had experienced spontaneous pregnancy loss. This finding was also reflected at serum level. CONCLUSIONS: Periostin may be a serum marker of good endometrial receptivity and embryo quality, and predictive of pregnancy evolution.

Negative transcriptional regulation of the human periostin gene by YingYang-1 transcription factor.

Periostin (POSTN), an osteoblast-specific secreted protein known to be associated with cell adhesion activity for bone formation and development by the epithelial cell-derived tumors, leads to a significant enhancement in angiogenesis and tumorigenesis. At present, little is known about the mechanisms underlying its transcriptional control either in physiological or neoplastic conditions. In this study we demonstrate that the ability of the human POSTN promoter to drive transcription mostly depends on the activity of YingYang-1 (YY1) zinc finger transcription factor. YY1, whose regulatory role in biology includes, besides transcriptional control, also chromatin remodeling, DNA damage repair and tumorigenesis, acts as a strong negative modulator of the POSTN expression. We retain that the identification of the functional role of YY1 in the transcriptional control of the human POSTN gene adds new insights in the studies focused on gene expression in normal and transformed cells.

BRCA1 expression modulates chemosensitivity of BRCA1-defective HCC1937 human breast cancer cells.

Germline mutations of the tumour suppressor gene BRCA1 are involved in the predisposition and development of breast cancer and account for 20-45% of all hereditary cases. There is an increasing evidence that these tumours are characterised by a specific phenotype and pattern of gene expression. We have hypothesised that differences in chemosensitivity might parallel molecular heterogeneity of hereditary and sporadic breast tumours. To this end, we have investigated the chemosensitivity of the BRCA1-defective HCC1937 breast cancer cell line, and the BRCA1-competent MCF-7 (hormone-sensitive) and MDA-MB231 (hormone-insensitive) breast cancer cell lines using the MTT assay. The 50% inhibitory concentration (IC(50)) for the individual compounds were derived by interpolate plot analysis of the logarithmic scalar concentration curve after a 48 h exposure. HCC1937 cells were significantly (P<0.005) more sensitive to cisplatin (CDDP) (IC(50) : 30-40 microM) compared with MCF-7 (IC(50) : 60-70 microM) and MDA-MB231 (IC(50) : 90-100 microM) cells. On the other hand, BRCA1-defective breast cancer cells were significantly less sensitive to doxorubicin (Dox) (IC(50) : 45-50 microM) compared with MCF-7 (IC(50) : 1-5 microM) and MDA-MB231 (IC(50) : 5-10 microM) (P<0.02), as well as to paclitaxel (Tax) (IC(50) : >2 microM for HCC1937, 0.1-0.2 microM for MCF-7 and 0.01-0.02 microM for MDA-MB231) (P<0.001). Full-length BRCA1 cDNA transfection of BRCA1-defective HCC1937 cells led to the reconstituted expression of BRCA1 protein in HCC1937/(WT)BRCA1-derived cell clone, but did not reduce tumour cell growth in soft agar. BRCA1 reconstitution reverted the hypersensitivity to CDDP (P<0.02), and restored the sensitivity to Dox (P<0.05) and Tax (P<0.001), compared with parental HCC1937 cells. Taken together, our findings suggest a specific chemosensitivity profile of BRCA1-defective cells in vitro, which is dependent on BRCA1 protein expression, and suggest prospective preclinical and clinical investigation for the development of tailored therapeutical approaches in this setting.

Transcriptional regulation of the mismatch repair gene hMLH1.

We have characterized the promoter region of the human gene coding for the MLH1 mismatch repair protein. The total transcriptional activity of the hMLH1 promoter is driven by two positive cis-elements included between nucleotides -300 and -220. The upstream element is a canonical CCAAT box, and it is recognized by the heterotrimeric transcription factor NF-Y. On the other hand, the downstream element is recognized by a nuclear factor of about 120 kDa. Variations in hMLH1 intracellular levels may influence the surveillance of the genome integrity. The identification of the two elements may shad some light on the regulation of the transcriptional regulation of hMLH1 expression.

Evidence of a founder mutation of BRCA1 in a highly homogeneous population from southern Italy with breast/ovarian cancer.

Several genes have been involved in the pathogenesis of hereditary breast/ovarian cancer (BOC), but mutations in the BRCA1 gene are by far the most recurrent. In this study, we report the identification of a founder mutation in a geographically and historically homogeneous population from Calabria, a south Italian region. A screening performed on 24 patients from unrelated families highlighted the high prevalence of a 5083del19 alteration in the BRCA1 gene, which accounts for 33% of the overall gene mutations. The same mutation was also detected in 4 patients, all of Calabrian origin, referred to us by research centres from the north of Italy. Allelotype analysis, performed on probands and unaffected family members revealed the presence a common allele, therefore suggesting a founder effect due to a common ancestor. Our findings underscore the importance of ethnic background homogeneity in patients' selection and highlight the usefulness of founder mutations as a potential tool for optimisation of preclinical diagnosis in gene carriers and therapeutic approaches in affected individuals.

Hereditary nonpolyposis coloretal cancer: identification of novel germline mutations in two kindreds not fulfulling the Amsterdam criteria. Mutations in brief no. 203. Online.

Hereditary nonpolyposis colon cancer results from heritable defects in the MLH1, MSH2, PMS1 and PMS2 genes, which encode proteins involved in the mismatch repair process. In this work we report the identification of two novel germline mutations in the MLH1 gene from two unrelated HNPCC families. The two affected families do not fulfill the Amsterdam criteria. In family 1 we found a missense S93G mutation, which lies in a MLH1 domain critical for its MMR functions. In family 2 we found a two nucleotide insertion (AG) in position 523 from the AUG which determines an early stop codon at position 606 (codon 203). In both families the mutant alleles cosegregate with the cancer phenotype.

A common mechanism underlying the E1A repression and the cAMP stimulation of the H ferritin transcription.

Transcription of the H ferritin gene in vivo is stimulated by cAMP and repressed by the E1A oncoprotein. We report here the identification of the cis-element in the human promoter responsive to both cAMP- and E1A-mediated signals. This promoter region is included between positions -62 to -45 and binds a approximate 120-kDa transcription factor called Bbf. Bbf forms a complex in vivo with the coactivator molecules p300 and CBP. Recombinant E1A protein reduces the formation of these complexes. In vivo overexpression of p300 in HeLa cells reverses the E1A-mediated inhibition of the ferritin promoter transcription driven by Bbf. These data suggest the existence of a common mechanism for the cAMP activation and the E1A-mediated repression of H ferritin transcription.

Transcriptional activation of the H-ferritin gene in differentiated Caco-2 cells parallels a change in the activity of the nuclear factor Bbf.

In this paper, we examine the mechanisms that regulate the expression of the heavy (H) ferritin subunit in the colon carcinoma Caco-2 cell line allowed to differentiate spontaneously in vitro. The differentiation process of these cells in continuous culture is accompanied by an accumulation of the mRNA coding for the apoferritin H chain. The analysis of Caco-2 subclones stably transfected with an H-chain promoter-chloramphenicol acetyltransferase (CAT) construct revealed that the mRNA increase is paralleled by an enhanced transcription of the H gene, driven by the -100 to +4 region of the H promoter. The H gene transcriptional activation seems to be a specific feature of differentiated Caco-2 cells, since the activity of other promoters did not change upon differentiation. The -100 to +4 region of the H promoter binds a transcription factor called Bbf (B-box binding factor); electrophoretic-mobility-shift-assay analyses showed that the retarded complex due to Bbf-H promoter interaction is significantly increased in the differentiated cells. We propose that the activation of H-ferritin gene expression may be associated with the establishment of a differentiated phenotype in Caco-2 cells, and that the H-ferritin gene transcriptional up-regulation is accompanied by a modification in the activity of the transcription factor Bbf.

Negative and positive elements in the promoter region of the human apoferritin L gene.

We have characterized the promoter of the human gene coding for the apoferritin L subunit. Transient transfections of 5' and 3' deletion mutants indicate that the efficiency of the L promoter depends on both negative and positive cis-elements, located upstream and downstream of the transcription start point. DNaseI footprinting analysis of this DNA region revealed the presence of five protected segments. The most upstream one (element 1) corresponds to the negative cis-element and is recognized by factor(s) sharing a GC-sequence specificity. Three positive elements are in the region upstream of the start of transcription; a fifth positive cis-element (element 5) is localized in the first exon of the L gene.

PCR analysis of the H ferritin multigene family reveals the existence of two classes of processed pseudogenes.

The human gene coding for the apoferritin H subunit belongs to a complex multigene family constituted by the expressed gene and by an undefined number of pseudogenes. We have used a strategy based on PCR to amplify specifically the H pseudogenes from a sample of human genomic DNA. With this approach, three new H pseudogenes have been cloned and characterized by DNA sequence analysis. In addition, we have identified a new type of pseudogene, the size of which (700 bp) is caused by multiple detection events in the putative coding region.

Deletion polymorphism in the gene for angiotensin converting enzyme is associated with elevated fasting blood glucose levels.

The frequency and distribution of angiotensin converting enzyme insertion/deletion (ACE I/D) polymorphism, and its association with other known risk factors for coronary atherosclerosis, has been studied, in a normal south Italian population. Subjects homozygous for deletion showed elevated fasting blood glucose levels when compared with subjects homozygous for insertion. The difference was consistent with an increased number of type 2 diabetics among the former group of subjects.

Linkage disequilibrium of three polymorphic RFLP markers in the apolipoprotein AI-CIII gene cluster on chromosome 11.

We analysed the allelic and genotypic frequencies of three restriction fragment length polymorphisms in the region of chromosome 11 encoding apolipoprotein AI and CIII genes in a free-living population from South Italy (Calabria). These markers are located at -2500 and -78 bp from the transcription start site of apolipoprotein AI gene (XmnI and MspI, respectively), and in the 3' untranslated region of apolipoprotein CIII gene (SstI). XmnI and SstI label rare alleles (X2 and S2 indicate the presence of the site), whereas the absence of the MspI site (because of a G to A transition) marks the rare allele, M2. Pairwise linkage disequilibrium analysis was determined. Two significant non-random associations were found: a positive disequilibrium between ApoA1/XmnI and ApoA1/MspI markers (P < 0.0001), and a negative disequilibrium between ApoA1/XmnI and ApoC3/SstI markers (P < 0.05). Statistical analysis showed a significant difference in the S2-M2 haplotype frequency between the group of subjects with serum cholesterol levels in the highest decile (P < 0.005) and the group with serum cholesterol levels below the highest decile. The allelic frequency for each locus showed no significant difference between the two groups for all other metabolic parameters, included total cholesterol serum levels. These haplotypes are a more precise measure of genetic variations in the apolipoprotein cluster and their use should allow the mapping of mutations responsible for high serum cholesterol levels.