7031

TFF1

trefoil factor 1;TFF1;trefoil factor 1

breast cancer estrogen-inducible protein|breast cancer estrogen-inducible sequence|gastrointestinal trefoil protein pS2|polypeptide P1.A|protein pS2

21q22.3

protein-coding

AIM: To determine allelic imbalance on chromosomal region 21q22-qter including trefoil factor family genes (TFF) in cholangiocarcinoma (CCA) patients and analyze the correlation between allelic imbalances and clinicopathological parameters. METHODS: Quantitative PCR amplification was performed on four microsatellite markers and trefoil factor family genes (TFF1, TFF2, and TFF3) using a standard curve and SYBR Green I dye method. The relative copy number was determined by DNA copy number of tested locus to reference locus. The relative copy number was interpreted as deletion or amplification by comparison with normal reference range. Associations between allelic imbalance and clinicopathological parameters of CCA patients were evaluated by chi(2)-tests. Kaplan-Meier method was used to analyze survival. RESULTS: The frequencies of amplification at D21S1890, D21S1893, and TFF3 were 32.5%, 30.0%, and 28.7%, respectively. Patients who had amplification at regions covering D21S1893, D21S1890, and TFF showed poor prognosis, whereas patients who had deletion showed favorable prognosis (mean: 51.7 wk vs 124.82 wk, P = 0.012). Multivariate Cox regression analysis revealed that amplification of D21S1893, D21S1890 and TFF, blood vessel invasion, and staging were associated with poor prognosis. CONCLUSION: D21S1893-D21S1890 region may harbor candidate genes especially TFF and serine protease family, which might be involved in tumor invasion and metastasis contributing to poor survival. The amplification in this region may be used as a prognostic marker in the treatment of CCA patients.

The gastric tumour suppressor trefoil protein TFF1 is present as a covalently bound heterodimer with a previously uncharacterised protein, TFIZ1, in normal human gastric mucosa. The purpose of this research was firstly to examine the molecular forms of TFIZ1 present, secondly to determine if TFIZ1 binds other proteins apart form TFF1 in vivo, thirdly to investigate if TFIZ1 and TFF1 are co-regulated in normal gastric mucosa and fourthly to determine if their co-regulation is maintained or disrupted in gastric cancer. We demonstrate that almost all human TFIZ1 is present as a heterodimer with TFF1 and that TFIZ1 is not bound to either of the other two trefoil proteins, TFF2 and TFF3. TFIZ1 and TFF1 are co-expressed by the surface mucus secretory cells throughout the stomach and the molecular forms of each protein are affected by the relative abundance of the other. TFIZ1 expression is lost consistently, early and permanently in gastric tumour cells. In contrast, TFF1 is sometimes expressed in the absence of TFIZ1 in gastric cancer cells and this expression is associated with metastasis (lymph node involvement: p=0.007). In conclusion, formation of the heterodimer between TFIZ1 and TFF1 is a specific interaction that occurs uniquely in the mucus secretory cells of the stomach, co-expression of the two proteins is disrupted in gastric cancer and expression of TFF1 in the absence of TFIZ1 is associated with a more invasive and metastatic phenotype. This indicates that TFF1 expression in the absence of TFIZ1 expression has potentially deleterious consequences in gastric cancer.

pS2 or TFF1 is a member of the trefoil factor family, which is distributed throughout the gastrointestinal tract in both normal and diseased tissues. It is also considered to be one of the major estrogen-regulated proteins and an indicator of estrogen receptor (ER) functionality. pS2 has previously been investigated in benign and malignant prostate lesions with little information about its relationship to steroid receptor status. Our purpose was to correlate pS2 expression with steroid receptor status (ER alpha and progesterone receptor (PR)) and other pathologic variables in prostate carcinoma. 15 benign prostate hyperplasia (BPH) and 47 prostate carcinoma cases were investigated by means of immunohistochemistry for pS2, ER and PR expression. 80% of BPH showed pS2 cytoplasmic immunoreactivity in hyperplastic acini and about half of these cases also exhibited nuclear staining decorating basal or both basal and luminal nuclei. pS2 was highly expressed in prostate carcinoma (91.4%) with both cytoplasmic and nuclear patterns of staining. The latter pattern was significantly associated with carcinoma having a low Gleason score (p=0.02). pS2 lacked any significant correlation with steroid receptor status, stage or grade. Univariate survival analysis revealed a significant impact of stage (p=0.03) and nodal status (p<0.0001) on patient outcome. The diagnostic value of pS2 expression in prostate carcinoma validated 74.19% accuracy, 91.48% sensitivity and 78.18% positive predictive value. The high sensitivity of pS2 expression in prostate carcinoma could make it a suitable marker for diagnosis of prostate carcinoma, especially in metastatic cases of unknown origin. The absence of correlation and dissimilarity in immunolocalization between pS2 and ER alpha leads to the assumption that ER alpha could not be the regulatory protein for pS2 and may raise questions about the functionality of ER alpha in prostate. The nuclear pattern of pS2 immunoreactivity either in benign or malignant prostatic lesions is similar to the published data on ER beta distribution and could also identify a subset of carcinoma patients with a favorable prognosis.