Immunohistochemical localization of matrix metallloproteinase 2 and its specific inhibitor timp‐2 in neoplastic tissues with monclonal antibodies

Abstract

Matrix metalloproteinase-2 (MMP-2), synthesized as a 631 amino-acid proenzyme, is activated by cleavage of the first 80 amino acids and naturally inhibited by tissue inhibitor of metalloproteinase-2 (TIMP-2). We report here the production of MAbs against MMP-2 and TIMP-2 and their use in localizing the respective antigens on tumor tissues. The anti-MMP-2 MAb recognized the latent and activated MMP-2 mutant protein (mutein) with C-terminal deletion at amino acid 425, indicating that both N- and C-terminal amino acids of MMP-2 are not important for its binding. The binding study of anti-TIMP-2 MAb, using several C-terminally truncated TIMP-2 muteins, showed that the amino acids 111-126 of TIMP-2 are essential for the binding of this antibody. Besides their respective antigens, both MAbs also recognized the MMP-2/TIMP-2 complex. On frozen sections of breast tumor, anti-MMP-2 MAb stained mainly tumor-cell cytoplasm with varying intensity, while anti-TIMP-2 MAb gave a stromal staining of varying intensity and a weak or absent staining of tumor-cell cytoplasm, suggesting different localization of the proteins in these tumors. In addition, in 1/3 of the breast cases both antibodies also localized on tumor-cell membranes. Similar cytoplasmic and stromal but not membrane staining patterns were observed in colon, gastric, endometrial, squamous-cell, prostatic and ovarian carcinoma as well. Since MMP-2 degrades type-IV collagen, the major component of basement membranes, the differences between MMP-2 and TIMP-2 levels and localization in individual tumors may relate to the invasiveness of the tumor and thus provide predictive information. However, this aspect could not be discussed in this study because no biological and clinical parameters such as lymph-node involvement or Dukes' stage of the tumors were available.

Expression of tissue inhibitor of metalloproteinases TIMP-2 in human colorectal cancer - a predictor of tumour stage

“…The TIMP-2 positivity was most prominent in the tumour stroma, a finding also reported by others (Poulsom et al, 1992;Urbanski et al, 1993;Hoyhtya et al, 1994). Studies on the closely related TIMP-1 have shown similar distribution, predominantly to the ECM of colorectal tumours (Hewitt et al, 1991;Newell et al, 1994;Kossakowska et al, 1996), but we are not aware of any studies exploring the relation between TIMP-1 expression and tumour staging and survival in colorectal cancer.…”

“…We found weak focal or weak diffuse epithelial TIMP-2 staining in as many as 93 (44%) out of the 212 tumours, while Hoyhtya et al (1994) observed strong cytoplasmic staining in 22% of tumour epithelium but, as the clinicopathological variables in these patient materials are possibly not identical, no conclusions may be drawn.…”

“…CA-4001 was raised against the amino terminus of the proenzyme of the human MMP-2 (Margulles et al, 1992); GE-213 was raised against intact human MMP-9, recognizing both the latent and active forms of the enzyme (Nikkari et al, 1996); and T2-101 was raised against intact human TIMP-2, recognizing an epitope between the amino acids 111-126 on the TIMP-2 molecule (Hoyhtya et al, 1994). The antibodies, used at concentrations of 1.5, 10 and 20 jg ml-' respectively, were diluted in phosphate-buffered saline supplemented with 5% normal horse serum and 1% bovine serum albumin and were then incubated with tissues for 60 min at room temperature.…”

“…This degradation is suggested to be independent of the Dukes' stage in colorectal cancer in a paper by Jessup (1994). A possible predictor of local tumour invasiveness has been suggested for the levels and localizations of MMP-2 and TIMP-2 (Hoyhtya et al, 1994;Jessup, 1994), as tumour progression may result from increased degradation or decreased production of basement membrane components (Havenith et al, 1988;Jessup, 1994).…”

“…Only a limited number of studies have reported on TIMP-1 and/or TIMP-2 and/or MMP-2, and/or MMP-9 in colorectal cancer (Nakajima et al, 1990;Poulsom et al, 1992;Pyke et al, 1993;Emmert-Buck et al, 1994;Hoyhtya et al, 1994;Kossakowska et al, 1996;Nielsen et al, 1996;Swallow et al, 1996) and the possible correlation with tumour stage (van der Stappen et al, 1990;Levy et al, 1991;Urbanski et al, 1993;Newell et al, 1994;Liabakk et al, 1996) and survival (Liabakk et al, 1996). 805 The aim of the present study was to search for differences in the expressions of the type IV collagenases MMP-2 and MMP-9 and the inhibitor TIMP-2 in various Dukes' stages and their possible prognostic impact, thus possibly enabling improved selection of patients for additional therapy and surveillance.…”