Study Of Expression And Correlation Of MMP-26, TIMP-4 And ER In Malignant Endometrial Tissue

BackgroudThe incidence of endometrial cancer is rising gradually in recent twenty years. Within worldwide, endometrial cancer has been one of the three main tumors threatening women's physically and mentally healthy. Although the etiology of endometrial cancer is unknown, but it is generally believed that endometrial cancer is an estrogen - dependent disease. Endometrial hyperplasia results from prolonged stimulation of the endometrium with unopposed estrogen, and classified as either simple or complex depending on glandular histology. In extension , complex hyperplasia may develop atypia of the glandular architecture as well as nuclear atypia. Almost one third of atypical lesions will develop into endometrial carcinoma. About 75% of cases are diagnosed in stage . In this early stage, it is a highly curable tumor, with a 5 - yaer overall survival of 83%. His-tological differentiation of the tumors is an important risk factor for postoperative prognosis.The matrix metalloproteinases ( MMPs) are a family of zinc containing pro-eases with a capability to degrade most component of the extracellular matrix ( ECM) and basement membranes ( BM). To date, approx, 20 human MMPs have been cloned and partially characterized. The activity of MMPs is regulated locally within tissues by specific inhibitors, belonging to the tissue inhibitor of metalloproteinase (TIMP) family. Currently, four main mammalian TIMPs (TIMP-1, - 2, -3, -4) have been identified, they form tight 1:1 complexes with activated MMPs inhibiting their enzymatic activity. There exists dynamic e-quilibrium between MMPs and TIMPs to secure cell motility and ECM remodelling in physiological environment.Matrix metalloproteinase - 26 ( MMP - 26) and tissue inhibitor of metallo-proteinase -4 (TIMP -4) are novel members of the MMP and TIMP families, initially cloned from the human endometrial tumor and adults heart cDNA libraries , respectively. Kinetic studies of the inhibitory capacity of TIMP -4 for MMP -1,-2, -3, - 7, -9 and -26 revealed that TIMP -4 had highest affinity for MMP - 26 among these MMPs. Previous studies have established that MMPs and TIMPs levels are abnormally up - regulated in metastatic tumors and are associated with invasive lesions and poor clinical prognosis in cancer patients. And then, what are the expression patterns of MMP - 26 and TIMP - 4 in human en-dometrium and their regulation mechanism? Here we report the cellular distribution and expression of MMP - 26, TIMP - 4, ER and their correlations in normal endomerium, hyperplastic and malignant tissues using immunohistochemis-try and RT - PCR technique.Materials and methods1. Sampling and processing of tissue: All samples were collected between January 2003 and August 2005 at the Department of Obstetrics and Gynecology, the Second Hospital Affiliated to China Medical University. Samples from patients receiving steroid treatment were excluded from the study. We collected 37 cases of normal endometrial tissue. All specimens were classified according to an ideal 28 - day menstrual cycle. Normal endometrial samples were grouped in the early (days 4 ~ 10, n = 12 ) , the middle ( days 11 ~ 22, n = 14 ) and the late parts (days 23 -28 and 1-3, n = ll) of the cycle. Endometrial samples with hyperplasia were 23 cases and were classified as simple (n = 13) and complex with atypia(n = 10). All the malignant samples had pathological diagnoses and were divided into different groups according to their diverse clinical pathological characteristics.One part of every tissue sample was immediately stored at - 801 until further processed for RT - PCR. The other part was formalin - fixed and paraffin embedded for histopathological examination and immunohistochemistry.2. Methods;We performed immunohistochemistry to identify MMP-26, TIMP - 4 and ER protein cellular location. Reverse transcription - polymerasechain reaction ( RT - PCR) and immunohistochemistry were performed to detect MMP - 26, TIMP - 4 and ER expression in normal endomerium, hyperplasitc and malignant tissue. Furthermore we sought to determine any correlation between the expression of MMP -26, TIMP -4 and ER with different pathological characteristics and also tried to find any correlation among them three.3. Statistical methods: Results are presented as box plots showing 0. 5 th, 25 th, 50 th, 75 th, 99.5 th percentiles. SpsslO.O statistical software was used to evaluate the significance of differences between the groups and to analyse their correlations.Results1. Immunostaining localized MMP - 26 predominantly in epithelial cells and tumor cells. No general stromal staining was seen. Semi - quantifications with immunohistochemistry and RT - PCR revealed that MMP - 26 expression increased from the early to the middle part of the cycle, and was then low in the late part of the cycle in normal endometrial tissue. Samples with endometrial hy-perplasia had similarly high intensity of the signal as samples from the middle part of the cycle. Furthermore, expression level was not different between hyper-plastic samples without atypia and those with atypia. In contrast much lower signal intensity was found in the malignant samples, and within this group, expression level decreased gradually with loss of histological differentiation to finally be absent in the poorly differentiated tumors.2. Immunostaining for TIMP -4 in above endometrial samples was present mainly in epithelial cells and tumor cells, stroma had only weak staining. Semi-quantitative results obtained with RT - PCR were very similar to semiquantitative estimates of immunostaining signal. Peak expression in normal endometrial samples was found at midcycle and the transcript was virtually absent in the late part of the cycle. Samples of hyperplastic tissue, both with and without atypia, had high levels of MMP - 26 mRNA, similar to those found at midcycle. Significantly lower levels were found in the malignant tissues, and TIMP levels decreased from well to poorly differentiated tumor specimens.3. The ratio of MMP - 26mRNA/TIMP - 4mRNA was higher at midcycle than in the early and late parts of the normal cycle. There was no difference between the midcycle, hyperplasia, and atypical hyperplasia groups. The level of MMP-26mRNA/TIMP -4mRNA altered progressively with loss of histological differentiation, myometrial invasion and lymph node metastases in malignant samples, which was different from the expression pattern of MMP - 26 and TIMP -4.4. ER protein was localized in cell nucleus of epithelial cells, stromal cells and smooth muscle cells, in vessel walls, and in tumor cells. ER immunostain-ing decreased from maximal levels in the early part of the menstrual cycle to lowest level in the late part of the cycle. High ER content, comparable to that found in the early part of the cycle, was also present in samples with hyperplasi-a, both with and without atypia. ER staining declined in tumor specimens with loss of histological differentiation. The signal intensity was lower in poorly differentiated than in intermediately and well differentiated samples.5. The positive expression rates of MMP - 26 and TIMP - 4 in the ER positive group were much higher than those in the ER negative group which demonstrated that the expression of ER is closely related to the expression of MMP - 26 and TIMP - 4. Thus we searched the promoter regions of the MMP - 26 gene and TIMP -4 gene and found that sequence 130 to 116 of the MMP -26 promoter and sequence 930 to 916 of the TIMP -4 promoter have high homology with the consensus sequence of estrogen response elements (ERE). Conclusions:1. The expression pattern of endometrial MMP -26 mimics that of TIMP -4 in normal, hyperplastic, and malignant tissues. Their maximal levels in normal tissue at midcycle suggest a role for MMP - 26 and TIMP - 4 in the process of embryo implantation.2. Samples of hyperplastic tissue, both with and without atypia, have high levels of MMP -26 and TIMP -4 , similar to those found at midcycle. Expression of MMP - 26 and TIMP - 4 in normal, hyperplastic, and malignant tissues are significantly different.3. We found expression of MMP -26 and TIMP -4 to be as high in atypical hyperplasia as they were in non - atypical hyperplasia, so decreased expres-sion of MMP -26 and TTMP -4 do not seem to be an early event in the progression to endometrial malignancy.4. Downregulation of MMP -26 and TTMP -4 in malignant tissue is not related to different clinical stages, histological subtypes and lymph node metasta-ses except histological differentiations which suggests the more malignant the tumor is, the less MMP -26 and TIMP -4 express, the worse the patients'prognoses are. Thus MMP -26 and TIMP -4 may be prospective molecular biological markers to predict the prognosis of endometrial carcinoma.5. The level of MMP - 26mRNA/TIMP -4mRNA alters progressively with loss of histological differentiation, myometrial invasion and lymph node metasta-ses in malignant samples which is opposite to the expression pattern of MMP - 26 and TIMP -4. It shows that the absent expression of TIMP -4 is more obvious than that of MMP - 26 in malignant samples. Since the overall amount of MMP-26 is low in endometrial cancer, the comparatively higher level of MMP -26mRNA/TIMP - 4mRNA may not be a prominent factor that contributes to in-vasiveness and metastases of the malignant tumor.6. The finding of high expression levels of MMP -26 and TIMP -4 at mid-cycle and in endometrial hyperplasia and low levels in the late part of the cycle and in endometrial carcinomas suggests co - variation with the estrogen receptor. Expression of ER is closely related to expression of MMP -26 and TIMP -4 not only in normal menstrual cycle but also in endometrial carcinoma. So we further analysed the promoter regions of MMP - 26 and TIMP - 4 and found that sequence 130 to 116 o( the MMP -26 promoter and sequence 930 to 916 of the TIMP - 4 promoter have high homology with the consensus sequence of estrogen response elements ( ERE). These data suggest that the identified sequences in the MMP - 26 and TIMP - 4 promoters have high affinity for ER and may function as an ERE. Thus we draw a conclusion that ER may be involved in the gene regulation of MMP - 26 and TIMP - 4 which implies the distinct expression pattern of MMP - 26 and TIMP -4 from other members of MMP and TIMP families.