Interleukin-6, interleukin-10 and heat shock protein-90 expression in renal epithelial neoplasias and surrounding normal-appearing renal parenchyma.

Cytokines, notably the interleukins IL-6 and IL-10, have an important role in the development and progression of renal-cell carcinomas, acting in the host-tumor interaction and in tumor bulk. Heat shock proteins (HSP), in particular HSP-90, may have a regulatory role in cytokine biosynthesis and prognostic implication in some tumors. To define the roles of the cytokines IL-6 and IL-10 and HSP-90 in the progression of renal-cell carcinoma we analyzed immunohistochemical expression of these proteins in human renal-cell carcinomas from 95 total nephrectomies. IL-6, IL-10 and HSP-90 proteins were more strongly expressed in epithelium and stroma of the renal tumoral compartment than in adjacent normal peritumoral tissue. But the difference reached significance only for HSP-90 protein. The percentage of cells expressing IL-6, IL-10 and HSP-90 immunoreactivity was higher in benign epithelial tumors, than in normal peritumoral tissue, but lower than in renal-cell carcinomas. Whereas HSP-90 immunoreactivity seemed higher in more aggressive histological phenotypes (collecting-duct carcinoma) of renal-cell carcinomas, IL-10 protein levels were higher in more advanced TNM stage (pT3) tumors. Our observation suggests that IL-6 and IL-10 and HSP-90 may be useful markers associated with the development and progression of renal-cell carcinomas and have independent functional roles in this malignant condition.

Quantitative immunohistochemical analysis of matrilysin 1 (MMP-7) in various renal cell carcinoma subtypes.

The role of matrilysin 1 or matrix metalloproteinase-7 (MMP-7) in cancer is extremely complex and poorly understood. In this study we investigated differential expression of MMP-7 in the epithelium and stroma of 95 paraffin-embedded renal tumor samples by immunohistochemistry and compared tumoral with normal peritumoral renal tissue. We also determined a possible correlation of the immunohistochemical findings with histological subtype, tumor grade and stage of RCC. In all areas examined MMP-7 protein expression was significantly higher in epithelium than in stroma (P less than 0.01). MMP-7 was more less expressed in peritumoral normal areas than in benign epithelial neoplasias (renal papillary and oncocytomas) and RCC carcinomas, reaching the highest immunopositive reaction in chromophobe RCC subtypes, followed by conventional clear-cell and chromophilic-papillary RCC histological subtypes and the lowest levels in more aggressive RCC histotypes (spindle-cell and collecting-duct RCCs). MMP-7 reached their highest levels in high-grade and high-stage RCCs. Our observation suggests an important role of MMP-7 in the development and progression of renal cancer. The differential expression of MMP-7 in the various histological RCC subtypes may reflect the malignant phenotype and more aggressive behavior of RCCs.

Prostate--specific G protein couple receptor genes and STAG1/PMEPA1 in peripheral blood from patients with prostatic cancer.

We investigated whether prostate - specific G protein couple receptor genes and STAG1/PMEPA1 gene expression in peripheral- blood could be useful as a diagnostic or prognostic marker of prostate cancer. Circulating cells were identified by reverse transcription-polymerase chain reaction (RT-PCR) to detect PSGR and STAG1/PMEPA1 mRNA in peripheral blood (PB) from 11 patients with treated prostate cancer (CaP), 11 with newly-diagnosed untreated CaP and 20 with benign prostatic hyperplasia (BPH) (controls). RT-PCR amplified PSGR in 8 of 11 untreated and in 9 of 11 treated patients with CaP and in 16 of 20 with BPH; whereas it amplified PMEPA1 in 1 of 11 untreated and in 7 of 11 treated patients with CaP and in 4 of 20 with BPH. In our control tissues and cell lines nearly all the prostatic and non- prostatic tissues and cell lines expressed PSGR mRNA, whereas only one prostatic neoplastic tissue and the androgen-responsive (LNCaP) and androgen non-responsive (PC3) prostatic cell lines expressed PMEPA1. These findings suggest that the investigated genes are poorly specific and probably of little use as diagnostic or prognostic markers in peripheral blood for monitoring prostate cancer progression and recurrence.

Detection of human papillomavirus DNA, p53 and ki67 expression in penile carcinomas.

Our study is aimed at evaluating the presence of p53 and Ki67 expression by immunohistochemistry in a series of 11 paraffin-embedded penile carcinomas. We also investigated the presence of Human Papillomavirus (HPV) DNA in these tumours and performed an accurate typing by DNA sequencing on positive samples. Immunohistochemistry (IHC) was performed with the anti-p53 and Ki67 mouse monoclonal antibodies. DNA extracted from small sections of each specimen was submitted to amplification with HPV specific general primers; PCR products of the proper length were purified and sequenced. IHC demonstrated nuclear accumulation of mutated p53 and Ki 67 expression in 10/11 tumour samples (90.9%). The prevalence of HPV DNA was 72.7%; the most prevalent type was HPV16. Sequencing analysis revealed the presence of HPV53 (12.5%), HPV18 (25%) and HPV16 (62.5%). Out of the p53 or Ki67 positive carcinomas the percentage of HPV positives was 80% and 70% respectively. Our results indicate that penile carcinoma is frequently associated to high risk HPV and with diffuse p53 and Ki67 expression.

Transforming growth factor-beta pathway in human renal cell carcinoma and surrounding normal-appearing renal parenchyma.

OBJECTIVE: To analyze the role of the transforming growth factor (TGF)-beta pathway in renal tumors and to verify whether alterations in TGF-beta 1 pathway expression are associated with the grade of tumor differentiation and pathologic stage in renal cell carcinomas. STUDY DESIGN: The expression of TGF-beta 1 and TGF-beta receptors (T beta RI and T beta RII), SMAD-2 and SMAD-4 was investigated by immunohistochemistry in normal peritumoral and tumoral tissue from 53 renal cell carcinomas (clear cell type). The gene expression of SMAD-2 and SMAD-4 was also studied by reverse transcription polymerase chain reaction (RT-PCR) in normal peritumoral and tumoral tissue from 6 of 56 primary tumors. RESULTS: TGF-beta 1, T beta RI and T beta RII immunoreactivity was more frequent in tumoral than in normal peritumoral renal tissue (96.22%, 79.25% and 75.41% vs. 88.37%, 69.76% and 62.69%), whereas SMAD-2 and SMAD-4 immunoreactivity was more frequent in normal peritumoral than in tumoral tissue (23.25% and 30.23% vs. 15.09% and 7.54%). In tumor areas, immunohistochemical scores were lower for T beta RII than for T beta RI and TGF-beta 1 and higher than SMAD-4 and SMAD-2 scores. TGF-beta 1, T beta RI, T beta RII and SMAD-4 histologic scores correlated with neither the histologic grade of malignancy nor TNM clinical stage, whereas SMAD-2 protein levels were significantly lower in grade 3 than in grade 1 tumors. In the samples of normal kidney and carcinoma studied, RT-PCR detected the correct transcripts for SMAD-2 and SMAD-4, indicating that the RNA of the samples analyzed contained RNA sequences coding for these genes. CONCLUSION: Our data support the concept that the reduction of T beta RII and SMAD proteins in renal cell carcinomas is involved in tumor development and suggest an altered TGF-beta/SMAD signaling pathway in kidney neoplasia.

Effects of long-term treatment with the anti-androgen bicalutamide on human testis: an ultrastructural and morphometric study.

AIMS: To assess the effects of more than 4 years' treatment with the anti-androgen bicalutamide on human testis by clinical, ultrastructural and morphometric analysis. METHODS AND RESULTS: Two patients (aged 74 and 69 years) with prostate cancer were treated for more than 4 years with bicalutamide 50 mg daily. Clinical characterization and follow-up included luteinizing hormone (LH), follicle-stimulating hormone (FSH), testosterone and prostate-specific antigen (PSA) measurements and clinical response of the tumours. Due to progression of the disease, patients underwent surgical orchidectomy as a further androgen withdrawal therapy. Testis biopsies were studied by light and electron microscopy and analysed by morphometry. Control samples were obtained from the normal testis of two patients with testicular cancer who underwent orchidectomy. Clinical follow-up showed a good response in the control of tumour growth and serum PSA decreased to < 4 ng/mL; concentrations of serum LH, FSH and testosterone were within the normal range. Testicular morphology of treated patients was unexpectedly well preserved; the organization of seminiferous tubules was normal with all the germ line elements and mature spermatozoa present. In some areas, a net increase of peritubular connective tissue was evident which may be a consequence of the age of the patients. CONCLUSIONS: Long-term bicalutamide (50 mg) treatment appears to have very little impact on testis ultrastructure and sperm maturation, while it is effective in the control of androgen-dependent prostatic tumours.

Human herpesvirus 6 infection in neoplastic and normal brain tissue.

The presence and variant distribution of human herpesvirus 6 (HHV-6) was investigated by a nested polymerase chain reaction (PCR) in 118 biopsies from patients affected by nervous tissue tumor (115 primary tumors and 3 metastasis) and in 31 autopsy samples from the brain of healthy individuals. HHV-6 DNA sequences were detected in normal and neoplastic nervous tissue at a frequency of 32% and 37%, respectively. In both tissues, variant A was three times more frequent than the variant B. Peripheral blood lymphocytes (PBLs) derived from seven tumor affected patients contained the same variant as their respective brain sample, as judged by PCR. The expression of HHV-6 encoded immediate early protein p41 was detected by immunohistochemistry in neoplastic but not in normal brain. This may reflect viral reactivation from latency in immunocompromised patients. The seroepidemiological data indicated a frequency distribution of anti-HHV-6 antibodies in patients with brain tumors similar to that found in healthy donors.

Transforming growth factor beta 1 and androgen receptors in prostate neoplasia.

OBJECTIVE: To investigate the interplay between transforming growth factor (TGF) beta 1, androgen receptors and stromal-epithelial interactions in benign prostatic hyperplasia (BPH), prostate intraepithelial neoplasia (PIN) and prostate carcinoma areas of prostate neoplasia. STUDY DESIGN: In this immunohistochemical study we investigated staining patterns and then determined the correlation between TGF-beta 1 expression and androgen receptor status in the epithelium and stroma of 60 paraffin-embedded tissues from radical prostatectomies. RESULTS: Staining patterns differed in the epithelium and stroma of tumor and peritumor prostatic tissue. TGF-beta 1 immunostaining (H-scores) in the epithelium and stroma increased significantly from BPH to PIN and from BPH to prostate carcinoma in the epithelium (P < .05), whereas androgen receptor (AR) immunoreactivity significantly (P < .05) increased from BPH to PIN to prostatic carcinoma in epithelium and stroma. TGF-beta 1 did not correlate with histologic grade of differentiation, whereas AR proteins were more strongly expressed in Gleason score 5 and 6 than score 7 tumors (P < .05). Nonlinear regression showed a significant correlation (P < .01) between TGF-beta 1 and AR expression only in the stromal compartment of PIN. CONCLUSION: These findings argue in favor of an interaction between TGF-beta 1 and AR in the early stages of prostate carcinogenesis and suggest that TGF-beta 1 plays a central role in stromal-epithelial interactions during the early stages of malignant transformation.

Epidermal growth factor receptor, MUC-1 and MUC-2 in bladder cancer.

The aim of this study was to investigate the immunohistochemical expression of epidermal growth factor receptor (EGFR), mucin-1 (MUC-1) and mucin-2 (MUC-2) proteins in primary bladder carcinomas and to compare EGFR and MUC staining patterns with the histological findings, grade and stage of bladder carcinoma. Fifty-six surgical specimens obtained from superficial and deeply invasive bladder carcinomas were studied. Of the 56 bladder tumors 42 (75%) expressed EGFR, 34 (60.71%) MUC-1 and 15 (26.78%) MUC-2; while 7 tumors (12.5%) coexpressed MUC-1 and MUC-2 proteins. Immunohistochemical scores showed higher levels of EGFR than of MUC-1 (P <0.05) and MUC-2 (P = 0.000) and higher levels of MUC-1 than MUC-2 (P = 0.0010). EGFR and MUC-1 expression was stronger in high-grade tumors (grade 2/3) than in low-grade (grade 1/2) ones (P <0.05) and stronger in muscle invasive tumors (T2-T4) than in superficial (Ta-T1) ones. Linear regression showed a significant (P <0.05) correlation between EGFR and MUC-1 proteins, but no correlation between EGFR and MUC-2 or between MUC-1 and MUC-2. Immunohistochemical expression of EGFR, MUC-1 and MUC-2 increases as primary bladder carcinomas acquire a more aggressive phenotype. Differences in the distribution of EGFR and mucins within the urothelium may be of diagnostic and prognostic value. These antigens may be useful as markers for bladder malignancy.

Effects of triptorelin, a gonadotropin-releasing hormone agonist, on the human prostatic cell lines PC3 and LNCaP.

Some analogues of gonadotropin-releasing hormone (GnRH) influence the in vitro proliferation of cultured human cells by complex interactions that are only partially understood. This study explored the effect of Triptorelin, a GnRH agonist, on the LNCaP and PC3 prostatic cell lines, which are, respectively, responsive and unresponsive to androgen stimulation. The toxicity and cell cycle modifications induced by the drug were investigated by FACScan analysis; the effect on cell proliferation in different culture conditions was determined by counting in a Burker chamber; and the expression of binding sites for 125I-Triptorelin was revealed by displacement experiments. PC3 cell growth was completely unaffected by Triptorelin. The drug caused a double stimulatory-inhibitory action on the growth of actively proliferating LNCaP cells, depending upon the dose and environment. A significant inhibitory effect on proliferation, ranging from 25% to 65% compared with controls, was observed at a high dose (10(-4) M) according to the culture conditions; and a proliferative effect (42% compared with controls) was observed at a lower dose (10(-7) M) only in fetal bovine serum-supplemented medium. Displacement experiments revealed the expression of moderately high affinity and low affinity binding sites in LNCaP cells (Kd = 2.6 x 10(-8) and 7.7 x 10(-6) M) but only low affinity binding sites in PC3 cells (Kd = 2.7 x 10(-6) M), which suggests that the expression of binding sites with different affinity could be associated with a biological response to the drug. Proliferation studies in the presence of Cetrorelix, a GnRH antagonist, confirmed the different sensitivity of the 2 cell lines to GnRH analogues and showed that the proliferative effect of Triptorelin on LNCaP cells can be inhibited by the antagonist. Data confirm the cell specificity of Triptorelin's action and the peculiarity of its effects on prostatic cell proliferation in our experimental conditions.

Identification of an estrogen-mediated deoxyribonucleic acid-binding independent transactivation pathway on the epidermal growth factor receptor gene promoter.

To investigate the estrogenic effects on the transcriptional regulation of the epidermal growth factor (EGF) receptor (EGFR) gene, we assayed its promoter ability to direct transcription of the luciferase reporter gene after transfection into HeLa cells. Our studies demonstrated a dose-dependent activation of the EGFR gene transcription by ligand-bound estrogen receptor alpha (ERalpha). This action was retained by the 36-bp core promoter fragment and did not require the receptor DNA binding domain, as demonstrated by analyzing the role of ERalpha deletion mutants on EGFR gene promoter-derived constructs. The 36-bp promoter fragment does not contain an estrogen response element but an imperfect thyroid hormone response element half-site that overlaps the Sp1 binding site. ERalpha does not bind this imperfect thyroid hormone response element half-site but is able to enhance binding of Sp1 to its site, in gel mobility shift assays, suggesting that the mechanism by which the receptor stimulated the transcription involved protein-protein interactions that replaced DNA binding. To explain this action, we propose a model in which induction of the EGFR gene expression by estrogens in HeLa cells is dependent upon the formation of a transcriptionally active ERalpha-Sp1 complex that binds to the GC-rich (Sp1) region of the minimal promoter.

Transforming growth factor-beta expression in prostate neoplasia.

OBJECTIVE: To understand the role of transforming growth factor (TGF) -beta 1, -beta 2 and -beta 3 proteins and TGF-beta type I and II receptors in prostate neoplasia; to determine the correlation between expression of TGF-beta s and their relative receptors in the epithelial and stromal compartments of benign prostatic hyperplasia (BPH), prostatic intraepithelial neoplasia (PIN) and prostate carcinoma; and to determine whether TGF-beta and TGF-beta receptor expression is associated with the grade of tumor differentiation. STUDY DESIGN: Sixty prostate neoplasms were analyzed by immunohistochemistry using anti-TGF-beta 1, -beta 2, -beta 3, -beta RI and -beta RII antibodies. RESULTS: TGF-beta and TGF-beta receptor immunoreactivity was more strongly expressed in prostate carcinoma than in PIN and BPH, and TGF-beta type I and type II receptors were less strongly expressed than TGF-beta 1-3 proteins. The difference between epithelial and stromal compartments reached significance (P < .05) for all TGF-beta isoforms and related receptors only in BPH, whereas a significant difference was found for TGF-beta protein in all grades of PIN but not for prostate carcinoma tissue. Luminal epithelial cells of BPH and PIN coexpressed all three TGF-beta isoforms and preferentially TGF-beta RII. Conversely, basal epithelial cells stained strongly for TGF-beta 1, -beta 3 and -beta RI but not for TGF-beta 2 and more strongly for TGF-beta RI than -beta RII. Linear regression showed a positive correlation between TGF-beta 1 and -beta 2, between TGF-beta 2 and -beta 3 and between TGF-beta RI and -beta RII proteins in all areas. The epithelium of Gleason score 7 tumors contained significantly higher TGF-beta 2 protein levels than Gleason score 3 and 4, and 5 and 6 tumors (P < .05). CONCLUSION: Stromal and epithelial cells of malignant and nonmalignant prostatic tumors express all three TGF-beta isoforms and their related receptors. These may act as both paracrine and autocrine factors to influence prostate function and the stromal-epithelial cell interaction. TGF-beta and -beta R immunoreactivity noted in basal cells indicates that in BPH and PIN, TGF-beta Rs and signaling pathways remain intact. The overexpression of TGF-beta proteins and underexpression of TGF-beta receptors in prostate cancer could suggest a mechanism for prostate cancer cells to escape the growth inhibitory effect of TGF-beta, thus leading to a more malignant phenotype.

Heat shock protein-90, IL-6 and IL-10 in bladder cancer.

The progression of transitional-cell carcinomas of the bladder is associated with changes in general and local immune status. To understand the factors involved in the progression of transitional cell carcinoma and in the maintenance of an efficient anti-tumoural response, in this study we investigated by immunohistochemistry expression of HSP-90, IL-6 and IL-10 proteins in 56 surgical specimens obtained from superficial and deeply invasive bladder carcinomas. Of the 56 bladder carcinoma 52 (92.9%) expressed HSP-90, 48 (85.7%) IL-6 and 45 (80.3%) IL-10. High-grade and muscle-invasive tumours contained significantly higher levels of HSP-90 and IL-6 antibodies than low-grade and superficial tumours (p < 0.05). Linear regression showed a significant correlation between HSP-90 and IL-10 (p = 0.022) but not between HSP-90 and IL-6, or IL-6 and IL-10 expression. The variable quantities of HSP-90, IL-6 and IL-10 in the high-grade bladder carcinomas studied suggest that these proteins have independent functional roles and may be the immunogenic targets for an anti-tumoural response.

Transforming growth factor-beta1 expression in human acoustic neuroma.

HYPOTHESIS: The differing clinical behavior of acoustic neuroma (AN) may be explained by the presence of specific biological features involved in tumorigenesis and growth. BACKGROUND: Transforming growth factor (TGF) beta1 is known to participate in the regulation of peripheral nerve tumors, modulating cell proliferation and differentiation with mechanisms different from those of glial growth factors (GGF) and fibroblastic growth factors (FGF), which are responsible for Schwann cells' mitogen activity. METHODS: Surgically removed human AN specimens were fixed in formalin and embedded in paraffin for immunohistochemistry studies. Expression and localization of TGF-beta1 in different tumor regions were assessed after incubation of paraffin sections with a mouse monoclonal anti-TGF beta1 antibody (DBA, Milan, Italy). Clinically, the time elapsed between the beginning of symptomatology and AN size as shown by preoperative computed tomography, magnetic resonance imaging, or both was calculated as rough value of growth rate, which enabled slow-growing and fast-growing ANs to be distinguished. RESULTS: Eighty-four percent of AN specimens expressed TGF-beta1 positivity at the level of the cytoplasm of the Schwann cells. TGF-beta1 reactivity was also shown in the blood vessel walls (96.15%) and the tumor capsule (80.86%). TGF-beta1 reaction appeared higher in Antoni A regions than in Antoni B regions. No significant relationship was found between TGF-beta1 positivity and AN growth rate in the two groups. CONCLUSIONS: TGF-beta1 could participate in the biological behavior of AN, particularly as an important factor of tumor growth prediction by allowing rapidly progressive or potentially recurrent tumors to be differentiated from slow-growing tumors that are unlikely to recur. The clinical course of patients with AN is currently still of little help in predicting the rate of AN growth.

Transforming growth factor-beta expression in human testicular neoplasms.

OBJECTIVE: To investigate the localization of transforming growth factors (TGF-beta 1, -beta 2 and -beta 3) and their receptors (TGF-beta RI and RII). STUDY DESIGN: The study included 26 paraffin-embedded tissues from human testicular neoplasms: 15 seminomas, 2 embryonal carcinomas, 1 immature teratoma, 4 immature teratomas with embryonal carcinoma, 1 immature teratoma with seminoma, 1 seminoma with embryonal carcinoma and 2 gonadal stromal tumors (Leydig cell tumors). RESULTS: TGF-beta 1 immunoreactivity was cytoplasmic and was expressed in 22 (84.6%), TGF-beta 2 in 20 (77%), TGF-beta 3 in 11 (42.3%), TGF-beta-RI in 21 (80.8%) and TGF-beta-RII in 18 (69.2%) of the 26 neoplasms. The percentage of positive immunostained cells and the intensity of staining were significantly higher in tumor than in peritumor nonneoplastic testis. In the peritumor nonneoplastic testis, Leydig, Sertoli and germ cells coexpressed both the three TGF-beta isoforms and TGF-beta-RI and RII. The myoepithelial cells of the seminiferous tubules showed immunoreactivity for TGF-beta RI and RII but not for TGF-beta s. In tumor testis areas the pattern of TGF-beta and TGF-beta receptor expression and distribution varied according to the histologic type of testicular tumor. Seminomas showed a diffuse pattern of TGF-beta immunoreactivity, whereas immature teratomas had focal and patchy distribution. In teratomas, differentiated structures contained more TGF-beta s than undifferentiated structures.

Androgen receptors in ovarian tumors: correlation with oestrogen and progesterone receptors in an immunohistochemical and semiquantitative image analysis study.

The expression and distribution of androgen, estrogen and progesterone receptors was examined by immunohistochemical staining in 31 paraffin-embedded sections from ovarian tumors and the results were assessed by semiquantitative image analysis. Immunohistochemical staining showed heterogeneous patterns of steroid receptor distribution, with mainly nuclear immunoreactivity. Eighty-four percent of benign and malignant ovarian tumors expressed androgen receptors (AR), 74.19% estrogen receptors (ER) and 41.16% progesterone receptors (PR). All benign tumors showed immunoreactivity for the three steroid receptors. Malignant tumors expressed higher AR and ER histochemical scores (H-scores) than PR (82% vs 71% vs 39%). The incidence and expression levels of the steroid receptors varied widely in the different histological types of malignant tumors. Spearman rank analysis showed a positive significant (P < 0.05) correlation between AR- and ER and between ER- and PR-H-scores. In malignant ovarian tumors, neither AR, ER nor PR immunohistochemical scores correlated with tumor FIGO stage. Densitrometric analysis of immunostained steroid receptors is a valid method for assessing the steroid status, because it reduces subjective elements in scoring sections and increases the reliability of results. The high incidence of AR expression confirms the functional role of AR in ovarian tumors and suggests that the determination of AR content in ovarian cancer could have prognostic value.

TGF-beta1 in colonic neoplasia: a genetic molecular and immunohistochemical study.

Transforming growth factor-beta proteins are key regulators of cellular growth and differentiation. To understand the role of TGF-beta in colonic tumour progression, 47 paraffin embedded samples from colonic tumours (13 adenomas, and 34 adenocarcinomas) were studied. Gene mutations in the region coding for the active protein were studied by PCR SSCP analysis of exons 5, 6, and 7. TGF-beta1 mRNA expression and localization were studied by NISH using cDNA probes generated by RT-PCR. Protein distribution was investigated by immunohistochemistry using antibodies against both intracellular and extracellular forms. Three mutations were found: one in exon 5 (Dukes C) and two in exon 7 (Dukes C and A). TGF-beta1 mRNA was observed in 9 (69%) of 13 adenomas and in 30 (88%) of 34 adenocarcinomas. Levels of TGF-beta1 mRNA and proteins were higher in colorectal carcinomas than in colorectal adenomas. Tubular adenomas associated with dysplasia showed greater expression of TGF-beta1 than adenomas without dysplasia and than non-neoplastic colonic mucosa. In dysplasia and cancer, epithelial neoplastic cells and stromal cells were positive for TGF-beta1. In normal tissue endothelial cells and granulocytes sporadically showed immunoreactivity for TGF-beta1, whereas epithelial cells were all negative. The three mutations in TGF-beta1 exons 5, 6 and 7 found in colorectal adenocarcinomas suggest that TGF-beta1 mutation may play a role in colorectal carcinogenesis and that the presence of the mutant form contributes to the transformed state. Our two other findings, the loss of staining gradient in normal colonic mucosa and the higher levels of TGF-beta1 mRNA and proteins in colorectal carcinomas than in colorectal adenomas, indicate that the de-regulation of TGF-beta1 expression may occur as an early event in colorectal carcinogenesis. Although TGF-beta1 expression is generally a reflection of cellular differentiation, tumour grade is not associated with TGF-beta1 mRNA expression. Beside being present in the epithelial cells of the colonic tumours, TGF-beta1 mRNA also occurred in the stroma: its localization in the macrophages adjacent to tumour strongly suggests that TGF-beta1 could be secreted by macrophages. This hypothesis should lead to new therapeutic strategies for colorectal carcinoma.

Molecular genetic analysis of TGF-beta1 in ovarian neoplasia.

Malignant ovarian tumours have been associated with a loss of autocrine growth inhibition by transforming growth factor-beta. This study aimed to detect abnormalities in the gene structure, expression and localization of TGF-beta1, in paraffin-embedded samples from 31 ovarian neoplasias (21 malignant, 5 borderline and 5 benign). Gene mutations in the region coding for the active protein were detected by PCR-SSCP analysis of exons 5, 6 and 7. mRNA expression and localization was studied by nonisotopic in situ hybridization (NISH) using cDNA probes generated by the reverse transcriptase polymerase chain reaction (RT-PCR), and immunohistochemistry, using antibodies against both intracellular and extracellular (matrix-associated) forms of TGF-beta1. Four mutations were found: one in exon 6 (serous adenocarcinoma), one in exon 7 (Mullerian tumor), and two in exons 5 and 6 from a serous cystoadenoma. TGF-beta1 mRNA was expressed in 87% and proteins in 90% of ovarian tumours. Most tumours expressing large amounts of TGF-beta1 mRNA, also contained a large number of protein binding sites. In malignant tumors, TGF beta1 was more strongly expressed in high-grade ovarian carcinomas with a cystic-papillary pattern than in tumours with a solid growth pattern. Normal ovarian tissue (follicles, granulosa cells) adjacent to tumor showed weak epithelial labeling and staining. Gene mutation did not correlate with histological type of tumor, mRNA or protein expression. TGF-beta1 mutation and abnormalities in its expression seem to occur in benign and malignant ovarian tumors, and could be involved in their pathogenesis. TGF beta1 gene mutations may act in multistage ovarian neoplasia, by reducing epithelial cell responsiveness to TGF-beta1 negative growth control.

Molecular genetic analysis of TGF beta1 in breast cancer.

Conflicting data suggest that TGF-beta1 can either inhibit or promote the progression of breast cancer. To determine the biological role of TGF beta1 in mammary carcinoma, in this study we examined the gene structure, expression and localization of TGF-beta1 using paraffin-embedded samples from 32 (27 IDC, 1 ILC, 1 DCIS, 1 ADH) breast lesions. Gene mutations in the region coding for the active protein were investigated by PCR-SSCP of exons 5, 6, and 7. mRNA -TGF-beta1 expression and distribution was examined by NISH using cDNA probes generated by RT-PCR and immunohistochemistry. We detected two mutations in exon 6 TGF-beta1 from IDC; and TGF beta1 mRNA and proteins in 28 (87%) of the tumors. Invasive breast carcinomas had more intense TGF-beta1 activity than CIS and than normal tissue adjacent to tumor. TGF beta1 mRNA and proteins were higher at the edge of the tumor than in the center and were also higher in less differentiated breast neoplasms. TGF-beta1 mRNA transcription and protein levels did not correlate either with TGF-beta1 exon 6 mutation or type and grade of differentiation of breast tumors. These observations suggest that TGF beta1 mutations in breast neoplasms might cause loss or inactivation of the growth inhibitory effects of TGF-beta1. They also support the proposed role of TGF-beta1 in the pathogenesis of breast cancer.