| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Carcinogenesis |
Finsen Laboratory [B. S. N., L. R. L., K. D.] and the Department of Pathology [F. R.], Rigshospitalet, Copenhagen University Hospital, 2100 Copenhagen, Denmark; Departamento de Morfología y Biologia Celular [J. M. L.] and Departamento de Bioquímica y Biología Molecular [M. B., C. L-O.], Instituto Universitario de Oncología, Universidad de Oviedo, 33006 Oviedo, Spain; and Servicio de Cirugia, Hospital de Jove, 33290 Gijon, Spain [F. V.]
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
-smooth muscle actin immunoreactivity and MMP-13 mRNA in myofibroblasts. In contrast, cytokeratin-positive cancer cells, -smooth muscle actin-positive vascular smooth muscle cells, CD68-positive macrophages, and CD31-positive endothelial cells were all MMP-13 mRNA negative. In situ hybridization for MMP-13 in 17 DCIS lesions revealed expression in 10 cases. Immunohistochemical analysis of all DCIS cases identified microinvasion in 8 of the 17 lesions. Seven of the eight lesions with microinvasion were MMP-13 positive. Further analysis showed that MMP-13 expression was often associated with the microinvasive events. This particular expression pattern was unique for MMP-13 among other MMPs analyzed, including MMP-2, -11, and -14. We conclude that MMP-13 is primarily expressed by myofibroblasts in human breast carcinoma and that expression in DCIS lesions often is associated with microinvasive events. On the basis of these data, we propose that MMP-13 may play an essential role during transition of DCIS lesions to invasive ductal carcinomas. |
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Human MMP-13 was initially identified in breast cancer (9) and later was found overexpressed in other malignant tumors, including chondrosarcoma (11) , head and neck carcinoma (12 , 13) , basal cell carcinoma of the skin (14) , malignant melanoma (15) , vulvar carcinoma (16) , and esophageal carcinoma (17) . Furthermore, MMP-13 has been detected in a variety of inflammatory conditions (18, 19, 20) , in atherosclerosis and aortic aneurysms (21 , 22) , in skin wound healing (23) , and in destructive joint diseases such as osteoarthritis and rheumatoid arthritis (24, 25, 26) . Biochemical characterization of MMP-13 has revealed that it efficiently degrades the native helix of fibrillar collagens with preferential activity on type II collagen (24 , 27) . However, MMP-13 is also able to degrade several other extracellular matrix proteins in vitro, including collagens type IV, X, and XIV; fibronectin; tenascin; and fibrillin (28 , 29) . Biochemical studies have also revealed that pro-MMP-13 is activated at the cell surface by a proteolytic cascade involving MMP-14 and -2 (30) , and other studies have indicated that it can also be activated by plasmin (31) .
The finding that MMP-13 is overexpressed in several malignant tumors, together with its potent proteolytic activity against numerous extracellular matrix proteins, suggest that it plays a major role in the tissue-remodeling events accompanying tumor progression. However, to date there have been few studies on the nature and functional characteristics of MMP-13-producing cells in human malignancies, and in some of these studies the results are controversial (32) . Several studies have indicated that MMP-13 in head and neck squamous cell carcinomas is produced by epithelial tumor cells located at the invading front of the tumors as well as in a subset of the carcinomas in stromal fibroblasts surrounding the tumor cells (12 , 13) . For invasive breast carcinoma, the picture appears to be more complicated. Some studies have suggested that MMP-13 is produced by fibroblast-like cells located in the stromal compartment of the breast cancer tissue (33) , whereas other studies have indicated that MMP-13 is synthesized predominantly by epithelial tumor cells (9 , 34) . In an attempt to clarify this issue, we performed a detailed study of its expression in a series of invasive ductal and lobular breast carcinomas, and to evaluate the role of MMP-13 in transition to invasive carcinoma, we studied MMP-13 expression during microinvasion, through analysis of DCIS lesions.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1 x 1 x 0.3 cm specimens, fixed overnight at room temperature in Lillie’s neutral-buffered formalin, and embedded in paraffin. Among the patients with invasive ductal carcinoma, seven were diagnosed with histological malignancy grade I, six with grade II, and eight with grade III (35)
. The group of invasive lobular carcinomas consisted of classical (three patients), nonclassical (two patients), and mixed classical/nonclassical types (four patients). Malignancy grading of DCIS was according to Holland et al. (36)
. Five DCIS lesions were grade I, four were grade II, and eight were grade III. All DCIS lesions were selected for being extensive lesions. To identify microinvasion in the DCIS lesions, we immunohistochemically stained three adjacent sections for pan cytokeratin (detection of all epithelial cells), CK-14 (detection of myoepithelial cells), and -sm-actin (detection of myoepithelium), respectively. Microinvasion was then defined as foci of (cytokeratin-positive) carcinoma cells located in the stroma not measuring more than 1 mm in diameter and not being accompanied by (-sm-actin- and/or CK-14-positive) myoepithelium in accordance with the definition by the National Coordinating Group for Breast Screening Pathology (37)
. The study was in accordance with permission provided by The Regional Scientific-Ethical Committee for Copenhagen and Frederiksberg, Denmark (journal numbers KF 01-456/93).
Antibodies.
Antibodies directed against -sm-actin (clone 1A4), CD68 (clone PGM1), CD31 (clone JC/70A), and cytokeratins (clone MNF116, CK-5, -6, -8, and -17; clone AE1/AE3, CK-2, -4, -5, -6, -8, -10, -14, -15, -16, and -19) as well as biotinylated secondary antibodies and peroxidase-conjugated streptavidin were all purchased from Dako (Glostrup, Denmark). Rabbit polyclonal antibodies against CK-14 were from Babco, Covance (Richmond, CA).
cDNA Subclones and Generation of Radiolabeled RNA Probes.
The full-length MMP-13 cDNA cloned in pEMBL (pNot3A; Ref. 9
) was used to obtain two nonoverlapping subclones: A 730-bp fragment corresponding to the second half of the open reading frame plus 100 bp of the 3'-untranslated region was isolated by digestion with BamHI and HindIII and ligated into BamHI/HindIII-digested pBluescript KS (plasmid phMMP13-730). A 790-bp fragment containing the first half of the open reading frame was isolated upon BamHI digestion and inserted into BamHI-digested pBluescript KS (plasmid phMMP13-BB). Plasmids containing MMP-2 cDNA (pCol7201; bp 647-1284), MMP-14 cDNA (pMMP-14/02' bp 1060–1484), or MMP-11 cDNA (pBSII-SK-ZIV; bp 347-2105) have been described (38, 39, 40)
.
In Vitro Transcription.
Antisense and sense riboprobes were labeled with 35S-labeled UTP (NEN, Boston, MA) by in vitro transcription using Sp6, T3, and T7 RNA polymerases (Roche, Basel, Switzerland) as described (41)
. The DNA template was digested with DNase (Promega, Madison, WI). Unincorporated 35S-labeled UTP and DNA were removed by column chromatography with S-200HR microspin columns (Amersham Pharmacia Biotech Inc., Piscataway, NJ). The 35S activity was adjusted for every probe by dilution to 500,000 cpm/µl.
In Situ Hybridization.
In situ hybridization was performed essentially as described previously (41)
with minor modifications. Three-µm paraffin sections were deparaffinized in xylene, hydrated through graded ethanol solutions, and digested with proteinase K (5 µg/ml) for 5 min at 44°C. Sections were then dehydrated, and the 35S-labeled probes (2 x 106 cpm/slide) were incubated overnight at 55°C in a humidified chamber. Sections were washed in Hellendahl chambers with SSC buffer containing 0.1% SDS and 10 mM DTT at 150 rpm and 55°C on a Bühler incubation shaker (Johanna Otto GmbH, Hechingen, Germany) for 10 min in 2x SSC, for 10 min in 0.5x SSC, and for 10 min in 0.2x SSC. Sections were then treated with RNase A [20 µg/ml in 0.5 M NaCl, 1 mM EDTA, 10 mM Tris-HCl (pH 7.2)] for 10 min at 44°C to remove nonspecifically bound riboprobe. The subsequent washing was performed in 0.2x SSC as specified above. Sections were dehydrated in graded ethanol solutions containing 300 mM ammonium acetate, soaked in an autoradiographic emulsion (Ilford), exposed for 7–10 days, and finally developed. Sections were counterstained with H&E.
Combined in Situ Hybridization and Immunohistochemistry.
Three-µm sections were hydrated as described above and boiled in a microwave oven for 10–12 min in 10 mM citric acid (pH 6.0). Sections were allowed to chill for 20 min at room temperature and then were transferred to Shandon incubation chambers (Shandon Inc., Pittsburgh, PA) in TBS. Sections were incubated with primary antibodies for 2 h at room temperature at the following dilutions: -sm-actin, 1:100; CK-14, 1:2000; pan cytokeratin (MNF116 or AE1/AE3), 1:100; CD68, 1:100; and CD31, 1:30. Cytokeratin staining of DCIS samples was performed with a mixture of antibodies: MNF116 (1:300), AE1/AE3 (1:300), and CK-14 (1:2000). The primary antibodies were detected with biotinylated goat antimouse/rabbit (K492; Dako), followed by peroxidase-conjugated streptavidin, each incubated for 30 min at room temperature. All antibodies were diluted in TBS-BSA containing 1% RNase inhibitor (Roche, Basel, Switzerland). All incubation steps were followed by washes with autoclaved TBS. Sections were developed with diaminobenzidine for 7–10 min and immediately dehydrated for in situ hybridization, which was performed as described above with the antisense probe phMMP13-BB. Finally, the sections were counterstained with hematoxylin.
Semiquantitative Histological Evaluation.
All breast carcinomas were evaluated with respect to their MMP-13 mRNA signal. No signal in the entire section was scored as "0." A section with a single positive focus (defined as between 1 and 20 positive cells) was scored "1." A section with two or more individual positive foci was scored "2." A section with many positive cells distributed within the whole tumor sample was scored "3" (see Table 1
).
|
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. There was no difference in the signal intensity observed with the two probes. No hybridization signal was seen with two MMP-13 sense probes obtained from the same plasmids (Fig. 1)
. We therefore concluded that the positive hybridization signal detected in the invasive ductal breast carcinoma depicted in Fig. 1
represents the genuine MMP-13 mRNA. To extend these results, we performed in situ hybridization on a series of 21 invasive ductal carcinomas. These studies revealed that MMP-13 mRNA was detected in all analyzed invasive ductal carcinomas, although with a pronounced variation in the number and size of positive foci (Table 1)
. There was no obvious relationship between the extent of MMP-13 mRNA expression in the tumor sections and the histological malignancy grade of the individual tumors. The MMP-13 mRNA signal was seen most often in cells with a fibroblast-like morphology, which were located in the tumor stroma. No signal could be ascribed with certainty to the epithelial cancer cells in any of the carcinomas. The expression was most often focal and was found in the tumor periphery (13 carcinomas), in the paracentral areas (18 carcinomas), and within central sclerotic areas (10 of 19 carcinomas in which central sclerosis was present; Fig. 2, a–f
). No signal was detected in the normal, unaffected breast tissue present around the malignant lesion. In contrast to the ductal carcinomas, MMP-13 mRNA was detected in only 4 of 9 invasive lobular breast carcinomas (Table 1)
. The four positive lobular carcinomas were either nonclassical lesions or mixed classical/nonclassical lesions, whereas three of the five negative lobular carcinomas were classical lesions and the other two mixed classical/nonclassical lesions. The expression pattern in the four positive lobular carcinomas (Fig. 2, g and h)
was similar to that described above for the invasive ductal carcinomas, i.e., expression was found in stromal cells located within foci in the malignant area but not in any cancer cells and not in the normal, unaffected breast tissue.
|
|
-sm-actin for myofibroblasts, CD68 for macrophages, and CD31 for endothelial cells. With the anticytokeratin antibodies, we could substantiate our preliminary observation, that no cancer cells were MMP-13 mRNA positive in any of the six carcinomas studied (Fig. 3, a and b)
. However, this double-labeling technique allowed us to detect coexpression of -sm-actin immunoreactivity and MMP-13 mRNA in all carcinomas studied (Fig. 3, c and d)
. It was evident that virtually all MMP-13 mRNA-positive cells were -sm-actin-positive, whereas the majority of the -sm-actin-positive cells were MMP-13-negative. These -sm-actin-positive stromal cells could be either myofibroblasts or vascular smooth muscle cells (42)
. To differentiate myofibroblasts from vascular smooth muscle cells, we considered the association of the vascular smooth muscle cells to vessels, a characteristic morphological parameter (42)
. Because all MMP-13 mRNA- and -sm-actin-positive cells were not directly associated with vascular structures, we considered the MMP-13-producing cells as myofibroblasts. In addition, we did not detect the macrophage marker CD68 (Fig. 3, e and f)
or the endothelial cell marker CD31 (not shown) in any of the MMP-13-positive cells.
|
. Although MMP-13 expression was scarce, it was always seen in stromal fibroblasts close to the DCIS foci (Fig. 4, a and b)
. Because the expression of MMP-13 mRNA in the DCIS lesions was either absent or limited to single foci and because mainly grade II and III lesions were positive, we decided to study the expression pattern with respect to the presence of microinvasion. We analyzed the DCIS lesions for the presence of microinvasive foci by comparing three adjacent sections immunohistochemically stained for cytokeratins (all epithelial cells), CK-14 (myoepithelium), or -sm-actin (here for myoepithelium), respectively. DCIS lesions were categorized as microinvasive when foci of cytokeratin-positive carcinoma cells (1 mm or less) were identified in the stroma and when, in addition, these carcinoma cells were not associated with myoepithelium, i.e., they were negative for both CK-14 and -sm-actin. In 8 of the 17 cases, we identified microinvasion (Table 1)
. Microinvasion was not observed in any of 5 grade I DCIS lesions, but was observed in 8 of the 12 grade II/III DCIS lesions. Comparison of these data with the initial MMP-13 expression study showed that seven of the eight lesions of DCIS with microinvasion had MMP-13-positive cells, whereas only three of the nine lesions of DCIS without microinvasion had MMP-13-positive cells (Table 1)
.
|
. From each of the 17 lesions, three adjacent sections were immunohistochemically stained for pan cytokeratin, CK-14, or -sm-actin, and subsequently hybridized with an antisense probe for MMP-13. As expected, seven of the eight DCIS lesions with microinvasion were MMP-13-positive, and interestingly, MMP-13-positive stromal cells were found in virtually all of the microinvasive foci. This pattern was observed in both grade II (Fig. 4, d and e)
and grade III (Fig. 4, f–i)
DCIS. Among the DCIS lesions without microinvasion, a few stromal cells located in the vicinity of a DCIS structure were MMP-13 mRNA-positive in 3 of the 9 cases (Fig. 4, j–l)
. These studies also confirmed the expression of MMP-13 mRNA in stromal fibroblast-like cells, most of which were -sm-actin-positive myofibroblasts, and showed the presence of myoepithelial cells positive for both -sm-actin and CK-14 but negative for MMP-13 mRNA. We therefore conclude that DCIS grade II and III can induce MMP-13 expression in neighboring myofibroblasts, and because microinvasion is usually accompanied by MMP-13 mRNA-positive myofibroblasts, we suggest that MMP-13 enzymatic degradation of extracellular matrix induces microinvasion in these DCIS foci.
Finally, we examined whether the observed association of MMP-13 expression with microinvasion was a characteristic feature of this protease or whether it could be extended to other MMP family members. We analyzed the expression of MMP-2, -11, and -14 mRNAs in six of the DCIS lesions without microinvasion and in four of the DCIS lesions with microinvasion by in situ hybridization. Two of the six DCIS lesions without microinvasion were MMP-13 mRNA-positive, but all six were found to be MMP-2 and -11-positive, and five were MMP-14-positive (see Fig. 5
). Three of the four DCIS lesions with microinvasion were MMP-13-positive, but all four were MMP-2, -11, and -14-positive. These four MMPs were all expressed by fibroblast-like stromal cells, but with different expression patterns. MMP-2 showed the most extensive expression pattern, and MMP-13 the most restricted expression pattern. MMP-11 showed increased expression around noninvasive DCIS outgrowths, as has been reported previously by others (43)
. MMP-2 and -14 are considered the main activators of pro-MMP-13, and by studying their putative colocalization in consecutive adjacent sections, we found that both MMP-2 and -14 were expressed at, but not restricted to, microinvasive areas (see Fig. 5
).
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The first novel finding in this study was that all of 21 cases of invasive ductal carcinoma investigated contained MMP-13 mRNA-positive cells, in contrast to only 4 of 9 cases of invasive lobular carcinoma. This is interesting because most invasive lobular carcinomas have better prognoses than most invasive ductal carcinomas (44) . However, a larger clinical study is required to clarify whether the subgroup of MMP-13-negative invasive lobular carcinomas represents a group of carcinomas of particularly favorable prognosis.
Another significant finding was the characterization of the MMP-13 mRNA-positive cells as myofibroblasts. On the basis of an immunohistochemistry and in situ hybridization double-labeling technique, we concluded that virtually all of the MMP-13 mRNA-positive cells were -sm-actin-positive fibroblast-like cells not directly associated with vascular structures; therefore, they can be considered as myofibroblasts. Myofibroblasts were originally described in healing wounds and are believed to be involved in wound contraction (45
, 46)
. It is a predominant cell type in the stroma of several cancer types (46)
, including breast cancer (42
, 47)
, where it has been shown to originate primarily from fibroblasts and vascular smooth muscle cells (48)
. The possibility that cancer cells also can convert into myofibroblasts on epithelial to mesenchymal transition has recently been discussed (49)
. Studies of myo- (differentiated) fibroblasts cocultured with breast cancer cells indicated that they can facilitate the migration of cancer cells and thereby promote tumor progression (50)
. We have previously shown that the myofibroblast is the predominant cell type in human breast cancer expressing uPA mRNA (51)
. This is noteworthy for at least two reasons: (a) high uPA levels in breast cancer tissue are associated with short survival (52
, 53)
, suggesting that myofibroblasts also have a tumor-promoting role in vivo; and (b) plasmin is one of the few proteases that have been reported to process pro-MMP-13 into active MMP-13 (see below), potentially making the myofibroblasts capable of activating pro-MMP-13 on their own. It does, however, remain to be clarified whether MMP-13 and uPA are expressed by the very same myofibroblasts in human breast cancer.
Our findings are in accordance with a previous study showing that stromal cells adjacent to epithelial tumor cells are the main source of this enzyme (33)
. In that study, MMP-13 mRNA was detected only in 30% of invasive ductal carcinomas, whereas we detected MMP-13 expression in all 21 cases studied, although with a pronounced variation in the number of positive cells identified in the individual samples. We believe that this difference is caused by differences either in tissue fixation (54
, 55)
or by the selection of tissue samples chosen for analysis because MMP-13 expression usually is focal and in some cases limited to one or a few foci within a single sample. A discrepancy with the study by Heppner et al. (34)
, who found MMP-13 mRNA signal confined to some isolated breast cancer cells, is more difficult to reconcile. However, in that study, no attempts to specifically characterize the positive cells were reported. A preliminary immunohistochemical analysis suggested that MMP-13 immunostaining is associated with breast cancer cells (9)
. Although such localization may be explained by the uptake of extracellular MMP-13 protein, produced by stromal cells, by cancer cells, it should be noted that these immunohistochemical data were based on the use of a single preparation of polyclonal antibodies. Our attempts to identify MMP-13 by immunohistochemistry using recently described monoclonal antibodies (56)
have failed to provide definitive information on the identity of MMP-13-containing cells, probably because of cross-reactivity with other MMPs, and further studies will be required to clarify the localization of MMP-13 protein in breast cancer.
Probably the most interesting finding of this work came from the studies of DCIS lesions. A consistent diagnostic difficulty in DCIS is to assess whether microinvasion is present. However, the diagnosis is decisive for the treatment. Patients with DCIS without microinvasion generally receive local excision only, whereas patients with microinvasive DCIS are treated as patients with invasive cancer, including axillary dissection and possibly adjuvant drug treatment, depending on the disease stage. Microinvasive events are characterized by carcinoma cells that traverse or already have traversed the continuous layer of basal cells, i.e., the myoepithelium that surrounds DCIS foci (36)
. In this situation, the myoepithelial cell layer is interrupted by the invasive carcinoma cells, which may appear as single cells or cell clusters in the stroma without any associated myoepithelial cells. The diagnosis of microinvasive carcinoma according to the definition "a tumor in which the dominant lesion is DCIS but in which there are one or more clearly separate foci of infiltration of fibrous or adipose tissue, none measuring more than 1 mm in diameter" (37)
is generally based on standard H&E staining, which often is supplemented with identification of myoepithelial cells by immunohistochemical staining for markers such as the basal cell cytokeratins CK-5, -14, and -17 (47
, 57, 58, 59, 60, 61)
as well as muscle-specific proteins such as -sm-actin, smooth muscle myosin heavy chain, and calponin (47
, 62)
. However, the basal cell cytokeratins are also expressed by a subpopulation of cancer cells in 5–30% of invasive carcinomas (59
, 61)
, and the muscle-specific proteins are also expressed by myofibroblasts and vascular smooth muscle cells (62
, 63)
. Therefore, the use of a single myoepithelial cell marker in the clarification of the presence of intact myoepithelium or microinvasion may not be sufficient, but would require additional supplementary myoepithelial cell markers.
A general method to assess microinvasion in DCIS with certainty has not been established, but the use of different combinations of markers has been reported, including the combination of CK-17 and the tumor cell marker CK-8 (60)
, pan cytokeratin and -sm-actin (64)
, -sm-actin and calponin (65)
, and smooth muscle myosin heavy chain and calponin (63)
. In this study we chose to stain three adjacent sections with one of the following markers: a pan cytokeratin for identification of all epithelial cells, and two supplementary myoepithelial cell markers, one specific for a myoepithelium-specific cytokeratin (CK-14), and one specific for a muscle-specific protein (-sm-actin). The pan cytokeratin-stained section facilitated the identification of foci suspicious of microinvasion, and the myoepithelial cell markers were used for the final assessment of microinvasion. Thus, microinvasion was defined as foci of (cytokeratin-positive) carcinoma cells traversing the myoepithelial cell layer or foci of carcinoma cells, measuring <1 mm in diameter, that were not accompanied by (CK-14- and/or -sm-actin-positive) myoepithelium, in accordance with the above definition.
Using the immunohistochemistry and in situ hybridization double-labeling technique to compare the MMP-13 expression with the staining of the three markers, we observed in seven of eight DCIS lesions with microinvasion an association of microinvasive events with MMP-13 expression in surrounding myofibroblasts, whereas little or no MMP-13 signal was observed in nine DCIS lesions without microinvasion. The association of MMP-13 expression with the microinvasive events was very distinct when compared with other MMPs, including MMP-2, -11, and -14, which all showed a broader expression pattern not selectively associated with microinvasion. These findings indicate that MMP-13-catalyzed extracellular matrix degradation may be directly involved in early cancer cell invasion occurring in DCIS lesions.
The induction of MMP-13 expression in the fibroblasts associated with microinvasion may be caused by a specific combination of factors located in the microenvironment of microinvasive foci. In vitro studies have shown that particularly IL-1, IL-1ß, and transforming growth factor-ß1 can induce MMP-13 mRNA expression in fibroblasts (33)
. In breast tumor tissue, it has been reported that IL-1 and IL-1ß are located in cancer cells (66)
, and transforming growth factor-ß1 mRNA is expressed in both cancer and stromal cells (67)
. Taken together, these observations suggest intercellular paracrine mechanisms in association with microinvasion in vivo. Regardless of the mechanism involved in the generation of microinvasive lesions, it seems clear that several extracellular matrix-degrading proteases are involved in the development of the invasive process (5
, 68)
. Because MMP-13 is synthesized as an inactive precursor, it must be activated by other enzymes, such as plasmin, as indicated above. Preliminary data indicate that indeed the plasminogen activator uPA is expressed by stromal fibroblast-like cells associated with DCIS in a pattern partially overlapping that of MMP-13.4
It is therefore noteworthy that recent data (69)
suggest that pro-MMP-13 can bind to a type C lectin called endo-180 (70)
, which is identical with a novel interaction partner for uPA and its receptor, termed uPARAP (71)
. Whether this interaction can promote pro-MMP-13 activation through plasmin remains to be clarified. MMP-14 and MMP-2 in a membrane-bound complex with MMP-14 are powerful activators of pro-MMP-13 in vitro (30)
. The finding that both MMP-14 and -2 are expressed by fibroblast-like cells similar to those producing MMP-13 mRNA in DCIS raises the possibility that pro-MMP-13 is secreted from the fibroblasts and subsequently bound and activated at the cell surface by the MMP-14/MMP-2 complex.
It is evident from the present study that MMP-13 is one of several extracellular matrix-degrading proteases expressed by stromal cells in breast cancer. In other types of adenocarcinomas, such as colon adenocarcinoma, similar stromal cells express several different proteases (72) . However, this pattern appears to be different in squamous cell carcinoma, where at least some proteases, including MMP-13, MMP-9, and uPA, are expressed by the cancer cells themselves (14 , 16 , 38 , 51 , 73) . Expression of MMP-13 by stromal cells in association with cancer cell invasion, as reported here in microinvasive ductal carcinoma, suggests that nonepithelial cells can modify particular steps in carcinogenesis (74) . Such a hypothesis recently was studied experimentally in MMP-9-deficient mice that develop skin cancer, where MMP-9 supplied by transplanted normal bone marrow cells was found to promote squamous cell skin carcinogenesis (75) .
Thus, histochemical studies on the expression of extracellular matrix-degrading proteases in human breast cancer may represent a valuable tool to illuminate steps in breast carcinogenesis. Taken together, our observations show that MMP-13 expression is induced in myofibroblasts during early stromal invasion of breast carcinoma cells in DCIS lesions and that MMP-13 expression is more restricted to microinvasive events than other MMPs, including MMP-2, -11, and -14. According to these data, analysis of MMP-13 expression should be explored to identify DCIS lesions with an increased risk of invasive local recurrence (76) , thereby facilitating decisions on the therapeutic strategy to be followed after diagnosis of these early lesions in the multistage process of tumor invasion and metastasis.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
1 This work was supported by The Danish Cancer Society, The Danish Cancer Research Foundation, and Plan Feder-Spain. 
2 To whom requests for reprints should be addressed, at Finsen Laboratory, Rigshospitalet Copenhagen University Hospital, Strandboulevarden 49, 7.2, 2100 Copenhagen, Denmark. Phone: 45 3545 5727; Fax: 45 3538 5450; E-mail: schnack{at}finsenlab.dk
3 The abbreviations used are: MMP, matrix metalloprotease; MT-MMP, membrane-type MMP; DCIS, ductal carcinoma in situ; CK, cytokeratin; -sm-actin, -smooth muscle actin; TBS, Tris-buffered saline; uPA, urokinase-type plasminogen activator; IL, interleukin.
4 B. S. Nielsen, unpublished observation.
Received 4/30/01. Accepted 7/25/01.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
and IL-1ß expression. Oncol. Rep., 6: 65-70, 1999.[Medline]
This article has been cited by other articles:
|
|
 |
|
  N. Rocks, G. Paulissen, F. Quesada-Calvo, C. Munaut, M.-L. A. Gonzalez, M. Gueders, J. Hacha, C. Gilles, J.-M. Foidart, A. Noel, et al. ADAMTS-1 Metalloproteinase Promotes Tumor Development through the Induction of a Stromal Reaction In vivo Cancer Res., November 15, 2008; 68(22): 9541 - 9550. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Almholt, A. Juncker-Jensen, O. D. Laerum, K. Dano, M. Johnsen, L. R. Lund, and J. Romer Metastasis is strongly reduced by the matrix metalloproteinase inhibitor Galardin in the MMTV-PymT transgenic breast cancer model Mol. Cancer Ther., September 1, 2008; 7(9): 2758 - 2767. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. A. Alyahya, M. Kolko, J. U. Prause, B. S. Nielsen, J. Wang, and S. Heegaard Matrix Metalloproteinase-2 Is Expressed in Melanoma-Associated Spongiform Scleropathy Invest. Ophthalmol. Vis. Sci., July 1, 2008; 49(7): 2806 - 2811. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Huet, B. Vallee, D. Szul, F. Verrecchia, S. Mourah, J. V. Jester, T. Hoang-Xuan, S. Menashi, and E. E. Gabison Extracellular matrix metalloproteinase inducer/CD147 promotes myofibroblast differentiation by inducing {alpha}-smooth muscle actin expression and collagen gel contraction: implications in tissue remodeling FASEB J, April 1, 2008; 22(4): 1144 - 1154. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Rizki, V. M. Weaver, S.-Y. Lee, G. I. Rozenberg, K. Chin, C. A. Myers, J. L. Bascom, J. D. Mott, J. R. Semeiks, L. R. Grate, et al. A Human Breast Cell Model of Preinvasive to Invasive Transition Cancer Res., March 1, 2008; 68(5): 1378 - 1387. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. X. Pedersen, C. J. Pennington, K. Almholt, I. J. Christensen, B. S. Nielsen, D. R. Edwards, J. Romer, K. Dano, and M. Johnsen Extracellular protease mRNAs are predominantly expressed in the stromal areas of microdissected mouse breast carcinomas Carcinogenesis, July 1, 2005; 26(7): 1233 - 1240. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. EGEBLAD, L.E. LITTLEPAGE, and Z. WERB The Fibroblastic Coconspirator in Cancer Progression Cold Spring Harb Symp Quant Biol, January 1, 2005; 70(0): 383 - 388. [Abstract] [PDF]
|
|
|
|
 |
|
 B. V. Offersen, B. S. Nielsen, G. Hoyer-Hansen, F. Rank, S. Hamilton-Dutoit, J. Overgaard, and P. A. Andreasen The Myofibroblast Is the Predominant Plasminogen Activator Inhibitor-1-Expressing Cell Type in Human Breast Carcinomas Am. J. Pathol., November 1, 2003; 163(5): 1887 - 1899. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
