Abstract Abstract #603 Background: Gene expression profiling has characterised breast cancers into distinct subtypes largely based on the ER, PR and HER2/neu receptor status (i.e., luminal A, luminal B, HER2/neu over-expressing and basal-like groups). This study aims to investigate relationships between expression levels of genes representing these receptors i.e., ESR1, ESR2, PGR and ERBB2, and selected estrogen receptor co-activator genes (NCOA3, CREBBP) and co-repressor genes (NCOR1 and NCOR2). We also sought to identify whether these co-regulator genes were differentially expressed across subtype, related to patient outcome or known prognostic factors.
 Materials and Methods: Using quantitative real-time PCR, we measured mRNA expression in sixty primary breast tumors and ten normal mammary glands. Relationships between gene expression levels and clinicopathological parameters, intrinsic subtype and clinical outcomes were analysed using Pearson's correlation coefficient, student t-test, ANOVA, Kaplan-Meier survival curves and Cox's proportional hazards model. P<0.05 was considered significant.
 Results: Statistically significant correlations were observed between CREBBP expression and ESR1 (r =0.324); ESR2 (r=0.387); PGR (r=0.397); and ERBB2 (r=0.306). CREBBP also correlated with NCOA3 (r=0.447), a steroid receptor co-activator previously shown to be over-expressed in breast cancers; and was differentially expressed in the luminal B subtype compared to the others (P<0.001). An increase in tumour stage and NCOA3 expression was associated with shorter overall survival (Log Rank; P=0.009 and P=0.043). Prognostic significance of these parameters remained after multivariate analysis. We found no correlation between co-regulator expression and age at diagnosis, macroscopic tumour size and lymph node status.
 Discussion: These findings demonstrate a relationship between CREBBP, breast cancer subtype, and the expression of genes known to be prognostic indicators or indicators of treatment response in breast cancer. Further investigations are required to establish how co-regulator expression influences these intrinsic groups. Citation Information: Cancer Res 2009;69(2 Suppl):Abstract nr 603.

A purpose-designed microarray platform (Stressgenes, Phase 1) was utilised to investigate the changes in gene expression within the liver of rainbow trout during exposure to a prolonged period of confinement. Tissue and blood samples were collected from trout at intervals up to 648 h after transfer to a standardised confinement stressor, together with matched samples from undisturbed control fish. Plasma ACTH, cortisol, glucose and lactate were analysed to confirm that the neuroendocrine response to confinement was consistent with previous findings and to provide a phenotypic context to assist interpretation of gene expression data. Liver samples for suppression subtractive hybridisation (SSH) library construction were selected from within the experimental groups comprising "early" stress (2-48 h) and "late" stress (96-504 h). In order to reduce redundancy within the four SSH libraries and yield a higher number of unique clones an additional subtraction was carried out. After printing of the arrays a series of 55 hybridisations were executed to cover 6 time points. At 2 h, 6 h, 24 h, 168 h and 504 h 5 individual confined fish and 5 individual control fish were used with control fish only at 0 It. A preliminary list of 314 clones considered differentially regulated over the complete time course was generated by a combination of data analysis approaches and the most significant gene expression changes were found to occur during the 24 h to 168 h time period with a general approach to control levels by 504 h. Few changes in expression were apparent over the first 6 h. The list of genes whose expression was significantly altered comprised predominantly genes belonging to the biological process category (response to stimulus) and one cellular component category (extracellular region) and were dominated by so-called acute phase proteins. Analysis of the gene expression profile in liver tissue during confinement revealed a number of significant clusters. The major patterns comprised genes that were up-regulated at 24 h and beyond, the primary examples being haptoglobin, beta-fibrinogen and EST10729. Two representative genes from each of the six k-means clusters were validated by qPCR. Correlations between microarray and qPCR expression patterns were significant for most of the genes tested. qPCR analysis revealed that haptoglobin expression was up-regulated approximately 8-fold at 24 h and over 13-fold by 168 h. (C) 2007 Elsevier Inc. All rights reserved.

Mammalian thioredoxin reductase (TRR; NADPH2:oxidized thioredoxin oxidoreductase, E.C. 1.6.4.5) is a new member of the family of selenocysteine-containing proteins. TRR activity in Se-deficient rat liver is reported to decrease to 4.5 to 15% of the activity in Se-adequate rat liver, similar to the fall in Se-dependent glutathione peroxidase-1 activity. Both glutathione peroxidase-1 enzyme activity and mRNA levels decrease dramatically in Se deficiency, whereas glutathione peroxidase-4 activity only decreases to 40% of Se-adequate levels and mRNA level is little affected by Se deficiency. The purpose of these experiments is to study the effect of Se status on TRR mRNA levels and enzyme activity in our well-characterized rat model, and to compare this regulation directly to the regulation of other Se-dependent proteins in male weanling rats fed Se-deficient diets or supplemented with dietary Se for 28 days. In two experiments, TRR activity in Se-deficient liver decreased to 15% of Se-adequate activity as compared to 2% and 40% of Se-adequate levels for GPX1 and GPX4, respectively. Using ribonuclease protection analysis, we found that TRR mRNA levels in Se-deficient rat liver decreased to 70% of Se-adequate levels. This decrease in TRR mRNA was similar to the GPX4 mRNA decrease in Se-deficient liver in these experiments, whereas GPX1 mRNA levels decreased to 23% of Se-adequate levels. This study clearly shows that TRR represents a third pattern of Se regulation with dramatic down-regulation of enzyme activity in Se deficiency but with only a modest decrease in mRNA level. The conservation of TRR mRNA in Se deficiency suggests that this is a valued enzyme; the loss of TRR activity in Se deficiency may be the cause of some signs of Se deficiency.

The introduction of real-time PCR technology has significantly improved and simplified the quantification of nucleic acids, and this technology has become an invaluable tool for many scientists working in different disciplines. Especially in the field of molecular diagnostics, real-time PCR-based assays have gained favour in the recent past. However, the wide use of real-time PCR methods has also highlighted some of the critical points and limitations of these assays. These aspects must be considered to increase the reliability of the obtained data.

Estrogen has been related to the development of hepatocellular carcinoma (HCC). In this molecular epidemiological study, we used logistic regression to compare the genotype frequencies of estrogen-metabolizing genes that are involved in estrogen biogenesis (CYP17), hydroxylation (CYP1A1) and inactivation of the reactive metabolites (catechol-O-methyltransferase, COMT) in HCC patients and control subjects, and determined their relationship with the risk of female HCC. The heterozygous or homozygous variants of high activity CYP17 (A2), high inducibility CYP1A1(m1), and low activity COMT (L) alleles were considered as high-risk genotypes. We found that the risk of HCC was elevated in women harboring either heterozygous or homozygous variants of the CYP1A1 gene and the respective OR (and 95% confidence interval) were 6.61 (1.35, 32.43) and 12.00 (1.73, 83.46). Moreover, we found that the risk of HCC was increased in the female subjects harboring higher numbers of high-risk genotypes, but not in male subjects. The OR for female HCC associated with two putative high-risk genotypes was 12.63 (1.50, 106.37), and the OR for three putative high-risk genotypes was 16.67 (1.82, 152.77). These findings strongly suggest that estrogen play a critical role in female hepatocarcinogenesis.

Increasing evidence indicates an association between the concentration of systemic progesterone during the early luteal phase of the oestrous cycle and embryo survival rate in cattle. We examined the relationship between the concentration of systemic progesterone on Days 4 to 8 post-ovulation and expression of progesterone receptor (PGR), oestrogen receptor +/- (ESR1) and retinol-binding protein (RBP) mRNA in the bovine endometrium. Heifers were blood sampled from the day of ovulation (Day 0) to Day 8 post-ovulation. On Day 4, animals were divided into low progesterone control (LC) and high progesterone control (HC) groups based on their plasma progesterone concentrations. Half of each group was supplemented with exogenous progesterone resulting in two further groups, low progesterone supplemented (LS) and high progesterone supplemented (HS). Endometrial tissues were recovered from all groups on Day 6 or Day 8 and gene expression was analysed following Northern blotting. Increasing progesterone concentrations were associated with decreased PGR and ESR1 expression. Duration-dependent effects of progesterone supplementation on ESR1 were evident and there was an effect of systemic progesterone concentrations between Day 0 and Day 4 on the expression of RBP at Days 6 and 8. Such progesterone-responsive changes in uterine gene expression are likely to affect embryo development.

Aims Our aim was to describe the commonly used molecular profiling techniques in cancer research, to examine their limitations and to discuss the challenges of bioinformatics. Methods A literature search was performed using the PubMed database to identify publications relevant to this review. Citations from these articles were also examined to yield further relevant publications. Results We describe the use of DNA microarrays, comparative genomic hybridisation, tissue microarrays and digital differential display. The limitations of these technologies, their contribution to cancer research and the challenges of bioinformatics are also discussed. Conclusions Although these high throughput technologies each have their own limitations they are rapidly developing and contributing significantly to our understanding of cancer genetics. They have also led to the emergence of bioinformatics as a rapidly developing and vital field.

Aims: Osteopontin (OPN) is an ubiquitous pro-metastasis gene associated with poor prognosis of colorectal, breast and lung cancer.Our aims were to identify any novel OPN suppressor gene and investigate its role in metastasis.Methods: Novel OPN interacting proteins were sought by bacterial and mammalian two-hybrid screens of a human cDNA library using full length rat OPN as ''bait''.Novel genes were cloned, sequenced and tested for OPN interaction by co-immunoprecipitation.OPN binding domains were mapped by sequence deletion experiments.Benign R37 mammary cells were rendered invasive by stable transfection with OPN.Effects of novel genes on OPN-mediated metastasis were investigated in vitro.Results: Of 25 different OPN interacting genes, ApoD (apolipoprotein D) showed maximal suppression of OPN expression.Transfection of invasive cells with ApoD inhibited OPN transactivation, reduced cell colony formation in soft agar from 207•8 ± 5•5 to 64•0 ± 5•5 (P < 0•001) and cell invasion from 70•5% to 47•3% (P < 0•001).Inhibition of ApoD by antisense RNA promoted OPN expression and enhanced cell invasion by parental R37 cells. Conclusion:This work shows for the first time that ApoD is a suppressor of OPN and inhibits metastasis, in a model system.These findings may provide a scientific basis for novel specific therapy to inhibit metastasis.

The relationships between the concentration of milk progesterone and early embryo survival on Days 4–8 inclusive and between the concentration of progesterone on different days from Days 0–8 inclusive following ovulation and insemination were examined in dairy cows. The relationships were examined following 77 randomly chosen artificial inseminations to cows in standing oestrus. There was a significant (P < 0.05) linear and quadratic relationship between the concentration of milk progesterone on each of Days 4–6 after ovulation and the probability of embryo survival. There was no association (P > 0.05) between milk progesterone concentration and probability of embryo survival on Days 7 and 8 after ovulation. There were no associations between milk progesterone concentration on Days 0–2 and the concentrations on Days 4–7, however, progesterone concentrations on Days 4 and 5 were highly predictive of the concentration on Days 6 and 7, respectively. Overall, the results indicate that suboptimal progesterone support during the early luteal phase is likely to deleteriously affect embryo viability and in addition, that it is possible to predict milk progesterone concentrations during the early luteal phase based on earlier stage concentrations and thus identify cows at risk of early embryo loss.