Matriptase and HAI-1 are expressed by normal and malignant epithelial cells in vitro and in vivo - PubMed (original) (raw)

Matriptase and HAI-1 are expressed by normal and malignant epithelial cells in vitro and in vivo

M Oberst et al. Am J Pathol. 2001 Apr.

Abstract

Matriptase and its cognate, Kunitz-type serine protease inhibitor, HAI-1, comprise a newly characterized extracellular matrix-degrading protease system that may function as an epithelial membrane activator for other proteases and latent growth factors. Both enzyme and inhibitor have been detected in breast cancer cells, immortalized mammary epithelial cells, and human milk, but not in cultured fibroblasts nor in fibrosarcoma cells. To test the hypothesis that this system is expressed by normal breast epithelium, invasive breast cancers, and other cancers of an epithelial origin (carcinomas) but not in cancers of a mesenchymal origin, we have expanded our expression analysis of matriptase and HAI-1 in vitro and in vivo. Matriptase and HAI-1 were detected at the protein and mRNA levels both in hormone-dependent and hormone-independent cultured breast cancer cells, and this expression correlated with the expression of the epithelial markers E-cadherin or ZO-1. However, none of the breast cancer cell lines tested that express the mesenchymal marker vimentin express matriptase or HAI-1, consistent with an epithelial-selective expression of this system. Expression of matriptase, as determined by Western blot analysis, was observed in primary human breast, gynecological, and colon carcinomas, but not in stromal-derived ovarian tumors and human sarcomas of various origins and histological grades. The epithelial-selective expression of matriptase and HAI-1 was further confirmed in human breast cancers by immunohistochemistry and in situ hybridization, where the expression of the protease and the inhibitor were found in the carcinoma cells and in surrounding normal breast epithelia. The expression of the matriptase/HAI-1 system by malignant epithelial cells in vivo suggests a possible role for this protease in multiple aspects of the pathophysiology of epithelial malignancy, including invasion and metastasis.

PubMed Disclaimer

Figures

Figure 1.

Figure 1.

Expression analysis of matriptase and HAI-1 in immortalized human breast epithelial, human breast cancer, and ovarian cancer cell lines. Ten micrograms of total RNA were examined for each breast or ovarian cell line by Northern blot analysis using matriptase-specific (A) or HAI-1-specific (B) riboprobes. Analysis of cell lines included two immortalized breast epithelial lines (MCF-10A and A1N4), four ER+ breast cancer cell lines (MCF-7, ZR-75-1, T47D, BT474), nine ER− breast cancer cell lines (SKBR3, MDA-MB-468, -453, -436, -435, -157, and -231, BT549, and Hs578t), and three ovarian cancer cell lines (SKOV3, PA-1, and OVCAR-3). Matriptase expression always correlated with HAI-1 expression, and both were found in two of two immortalized breast epithelial cell lines, four of four ER+ breast cancer cell lines, three of nine ER− breast cancer cell lines, and in one of three ovarian cancer cell lines.

Figure 2.

Figure 2.

Expression analysis of matriptase in mammary tissues. Samples of proteins were extracted using RIPA buffer from normal breast tissue surrounding the breast tumor of three different patients (lanes 13 to 15) and tumors of 10 different patients (lanes 3 to 12). Proteins (50 μg per lane) were separated by SDS-polyacrylamide gel electrophoresis, transferred to polyvinylidene fluoride membrane, and probed by anti-matriptase mAb 21-9 (A) and anti-HAI-1 mAb M19 (B). The positions of matriptase (70 kd), HAI-1 (55-kd membrane-bound form and 50-kd fragment), and the 95-kd matriptase/HAI-1 complex were indicated according to the samples from the cell-conditioned medium (lanes 1) and the membrane fractions (lanes 2) of T-47D breast cancer cells.

Figure 3.

Figure 3.

Expression analysis of matriptase in gynecological tumors. A and B: Samples of proteins (50 μg per lane), which were extracted by RIPA buffer from nine ovarian carcinomas (lanes 3 to 11) and three stromal-derived tumors, including two fibrothecomas (lanes 12 and 13) and one granulosa cell tumor (lane 14), were analyzed by immunoblot using anti-matriptase mAb 21-9 and anti-HAI-1 mAb M19. The positions of matriptase, HAI-1 and the 95-kd matriptase/HAI-1 complex were indicated according to the samples from the cell-conditioned medium (lanes 1) and the membrane fractions (lanes 2) of T-47D cells. C and D: Samples of proteins from four uterine carcinomas (lanes 3, 5, 7, and 8) and two patient-matched normal tissues surrounding tumors (lanes 4 and 6) were probed by immunoblot using anti-matriptase mAb 21-9 and anti-HAI-1 mAb M19. The positions of matriptase, HAI-1, and the 95-kd matriptase/HAI-1 complex are indicated, as described above.

Figure 4.

Figure 4.

Expression analysis of matriptase in human colon tumors. Protein samples (50 μg per lane) that were extracted by RIPA buffer from nine colon carcinomas (lanes 3 to 10) and four normal colon specimens (lanes 11 to 14) were examined by Western blot using anti-matriptase mAb 21-9 (A) and anti-HAI-1 mAb M19 (B). The positions of matriptase, HAI-1, and its 95-kd matriptase/HAI-1 complex, were indicated according to the samples from the cell-conditioned medium (lanes 1) and the membrane fractions (lanes 2) of T-47D cells.

Figure 5.

Figure 5.

Analysis of matriptase and HAI-1 protein expression in human breast carcinomas by immunohistochemistry. Human breast carcinomas were stained by immunohistochemistry using mAbs directed specifically against matriptase (S5) or HAI-1 (M58). Positive staining for matriptase and HAI-1 are observed as a brown precipitate (diaminobenzidine) within the sections, and nuclei were counterstained with hematoxylin. A metastatic breast adenocarcinoma shows both cytoplasmic and membranous staining for matriptase [original magnifications, ×100 (A) and ×400 (B)] and HAI-1 [original magnifications, ×100 (D) and ×400 (E)] in the breast epithelial cells. No staining is noted in stromal components of the tumor. A colloid breast carcinoma likewise shows a similar staining pattern for matriptase [original magnification, ×400 (C)] and HAI-1 [original magnification, ×400 (F)].

Figure 6.

Figure 6.

Analysis of matriptase and HAI-1 protein expression in normal and hyperplastic human breast epithelium by immunohistochemistry. Intense staining for matriptase is seen in the duct and mild or no staining in the surrounding terminal duct lobular units (TDLU) of this area of normal breast epithelium surrounding a breast carcinoma [original magnification, ×20 (A)]. A duct with usual ductal hyperplasia shows intense staining for matriptase, and the surrounding TDLUs show mild staining [original magnification, ×40 (B)]. Focal staining for HAI-1 is noted in the TDLU, and no staining is seen in the surrounding duct [original magnification, ×40 (C)]. A high-power view of the same lobule shows preferential staining of the lobular cells, and no staining is seen in the myoepithelial cells [original magnification, ×200 (D)].

Figure 7.

Figure 7.

Analysis of the expression of matriptase in invasive primary breast tumors by in situ hybridization. A matriptase-specific antisense probe (A and C) and the corresponding control sense probe (B and D, respectively) were hybridized to paraffin-embedded sections of primary breast tumors as described in Materials and Methods. The matriptase antisense probe shows reactivity with the cancer cells within the sections and lack of reactivity with stromal elements such as fibroblasts and adipocytes (A and C). Control sense probes do not show any reactivity with the breast cancer sections (B and D), demonstrating the specificity of the labeled antisense probe.

Figure 8.

Figure 8.

The tissue concentration of matriptase in human breast tumors and the surrounding tissues. The tissue concentration of matriptase was determined by immunoblot. The concentration of a purified matriptase standard was determined by comparison with a bovine serum albumin standard curve resolved by SDS-polyacrylamide gel electrophoresis and stained with Coomassie Blue. Different amounts of purified matriptase were then used (80 pg, 200 pg, 400 pg, and 800 pg; lanes 1 to 4) to generate a standard curve for the immunoblot. Protein samples (50 μg) from six human breast tumors (lanes 5 to 10) and two surrounding tissues (lanes 11 and 12) were examined by immunoblot and compared to the standard curve for matriptase. The final concentration of each specimen was calculated and indicated.

Similar articles

Cited by

References

    1. Mignatti P, Rifkin DB: Biology and biochemistry of proteinases in tumor invasion. Physiol Rev 1993, 73:161-195 - PubMed
    1. Behrendt N, Ronne E, Dano K: The structure and function of the urokinase receptor, a membrane protein governing plasminogen activation on the cell surface. Biol Chem Hoppe Seyler 1995, 376:269-279 - PubMed
    1. Conese M, Blasi F: Urokinase/urokinase receptor system: internalization/degradation of urokinase-serpin complexes: mechanism and regulation. Biol Chem Hoppe Seyler 1995, 376:143-155 - PubMed
    1. Holst-Hansen C, Johannessen B, Hoyer-Hansen G, Romer J, Ellis V, Brünner N: Urokinase-type plasminogen activation in three human breast cancer cell lines correlates with their in vitro invasiveness. Clin Exp Metastasis 1996, 14:297-307 - PubMed
    1. Hofmann R, Lehmer A, Buresch M, Hartung R, Ulm K: Clinical relevance of urokinase plasminogen activator, its receptor, and its inhibitor in patients with renal cell carcinoma. Cancer 1996, 78:487-492 - PubMed

Publication types

MeSH terms

Substances

LinkOut - more resources