help button home button Endocrine Society JCEM
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a related Letter to the Editor
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Singer, C. F.
Right arrow Articles by Cullen, K. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Singer, C. F.
Right arrow Articles by Cullen, K. J.
The Journal of Clinical Endocrinology & Metabolism Vol. 82, No. 6 1917-1922
Copyright © 1997 by The Endocrine Society

Endocrinological Oncology

Coexpression of Stromelysin-3 and Insulin-Like Growth Factor II in Tumors of Ectodermal, Mesodermal, and Endodermal Origin: Indicator of a Fetal Cell Phenotype1

Christian F. Singer, Audrey Rasmussen, Marc E. Lippman and Kevin J. Cullen

Vincent T. Lombardi Cancer Center, Georgetown University, Washington, D.C. 20007

Address all correspondence and requests for reprints to: Dr. Kevin J. Cullen, Division of Medical Oncology, Georgetown University Medical Center, 3800 Reservoir Road NW, Washington, D.C. 20007. E-mail: cullenk{at}gunet.georgetown.edu


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Stromelysin-3 (ST-3) is thought to play an important role in invasion and tumor progression. We have analyzed ST-3 expression in fibroblasts with defined topographical relations to breast cancers. We demonstrate that these fibroblasts exhibit the same distinctive pattern of messenger ribonucleic acid (mRNA) expression that we have previously shown for insulin-like growth factor II (IGF-II). Tumor-derived fibroblasts and skin fibroblasts produce abundant ST-3 mRNA. Fibroblasts from normal breast stroma distant from the malignant tumor in the same patient express considerably less ST-3 mRNA. When we analyzed ST-3 and IGF-II gene expression in sarcomas, we found a similar pattern of coexpression. Immunohistochemical analysis of IGF-II and ST-3 protein expression in sarcomas and breast tumors confirmed the mRNA data.

ST-3 mRNA expression was also seen in most colon cancer cell lines, again matching reports of IGF-II gene expression. As the two proteins are known to play an important role during fetal growth and development, their coexpression in fibroblasts from malignant tumors of ectodermal (breast cancer) and mesodermal (sarcoma) origin and in epithelial cells of endodermal origin (colon cancer) implies a more primitive cellular phenotype. The regained ability to express such developmentally regulated proteins might, therefore, be a more general marker indicating a fetal-type phenotype of cells in a malignant tumor.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
A COMMON feature of invasive processes in cancer is the degradation of extracellular matrix (ECM) adjacent to a malignant tumor cell. The ability of a neoplasm to cross anatomical barriers, such as basement membranes or basal laminae, and to invade adjacent ECM is very likely to be linked to whether it can produce matrix-degrading proteinases (1, 2, 3, 4, 5). A number of observations suggest that proteolytic enzymes, specifically matrix metalloproteinases (MMPs), might contribute to local invasion and to the progression of various malignant tumors (6, 7, 8).

MMPs comprise a family of nine highly homologous Zn2+-binding endopeptidases that collectively degrade most extracellular matrix and basement membrane components. They have been divided into three subclasses with respect to their specific substrates: collagenases, stromelysins (STs), and gelatinases (9, 10, 11). Several strong lines of evidence support the idea that MMPs play a major role in local tissue remodeling at all stages of human life: during embryonal development, in wound healing, in degenerative processes, but also in locally invasive growth of malignant tumors (12, 13, 14). The importance of MMPs in tumor invasion and metastasis has been well established in a number of in vitro and in vivo assays. Liotta et al. demonstrated that normal mouse cells are unable to degrade ECM components, whereas several cell lines derived from mouse tumors can solubilize specific substrates to varying degrees (15). Furthermore, it was shown that MMP inhibitors can actually block tumor invasion in some in vitro models (16, 17, 18).

Recently, a new member of the MMP family has been identified. ST-3 was first found in mesenchymal limb-bud cells and stroma around invasive breast tumors. It shares many features with the two previously described STs (19). Although ST-3 has been shown to be a powerful endopeptidase with a limited substrate specificity (20) and to possess the general properties of a weak metalloproteinase (21), no specific ECM-related substrate has been determined to date. Nevertheless, the ST-3 gene was expressed in almost all invasive breast carcinomas, in a number of breast cancer metastases, and in some types of in situ breast carcinomas that are known to have high tendency of becoming invasive (22). In these studies, in situ hybridization assays and immunohistochemistry were performed on tissue sections of the malignant tumor only.

Insulin-like growth factor II (IGF-II) is a 7.5-kDa peptide that is expressed in breast cancers and other malignancies. We have previously shown that, like ST-3, IGF-II is expressed in the stroma of tumors, rather than in tumor epithelium itself (23). Recently, we demonstrated that only fibroblasts derived from the stroma of malignant breast tumors [tumor-derived fibroblasts (TF)] express IGF-II, whereas the fibroblasts taken from areas of normal breast in the same patient [peripheral fibroblasts (PF)] do not (24). Both IGF-II and ST-3 are developmentally regulated proteins and are physiologically expressed during fetal development (19, 25). Both are expressed in fibroblasts immediately adjacent to the stromal-epithelial border of malignant breast tumors, and both are thought to be associated with increased tissue remodeling and invasiveness (26, 27, 28).

In this study, we investigated whether ST-3 and IGF-II are coexpressed in cancers arising in tissues derived from each of the three embryonic progenitor tissues: ectoderm, mesoderm, and endoderm.

First, we examined stromal and epithelial cells derived from breast (ectodermal) cancers. We analyzed matched sets of TF, fibroblasts from macroscopically normal breast tissue adjacent to the malignant tumor (PF), and fibroblasts from overlying breast skin (SKF) in each of six patients with invasive breast tumors. A panel of breast tumor epithelial cell lines was also examined for ST-3 messenger ribonucleic acid (mRNA) expression. IGF-II and ST-3 protein expression were analyzed by immunohistochemistry in tumor sections. Similar analyses were performed in sarcomas, which are cancers of mesodermal origin. Finally, we examined ST-3 mRNA expression in a panel of colon cancer (endodermal) cell lines.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Cell culture and tissue samples

Breast fibroblast cell lines were provided by Dr. Helene Smith, Geraldine Brush Cancer Center (San Francisco, CA), and had been established from surgical specimens of breast cancer patients as previously described (24). In each patient, fresh tissue samples were obtained from three specific locations: from the malignant tumor itself (TF), from macroscopically unaffected breast tissue adjacent to the tumor (PF), and from overlying breast skin (SKF). Sarcoma RNA samples were provided by Drs. Dennis Slamon and Peter Reissman, University of California (Los Angeles, CA). All fibroblasts were cultured in DMEM-Ham’s F-12 (Biofluids, Rockville, MD) containing 10% FCS, 2 mmol/L L-glutamine (Life Technologies, Gaithersburg, MD), and 50 µg/L gentamicin (Life Technologies). The epithelial breast cancer cell lines were purchased from the American Type Culture Collection (Rockville, MD; BT-474; MCF-7; MDA-MB-175, -231, -452, and -468; SK BR-3; T47D; and ZR-75B) or were provided by the Lombardi Cancer Center Tissue Core Facility (Washington DC). The igf-II-transfected MCF-7 breast cancer cell line clone 8 was previously established in our laboratory (28). Epithelial cell lines were maintained in FCS (10%) and L-glutamine-enriched Improved Minimal Essential Medium (Biofluids). Total RNA from epithelial colon cancer cell lines (HCT15, VACO 330, SW 48, SW 480, LS 180, Colo 201, Colo 205, WiDr, and CaCo2) and from the colon fibroblast cell line LCC-C8 were provided by Dr. S. Evans, Lombardi Cancer Center.

Extraction of total RNA

All cells were washed with iced PBS three times, harvested in GIT buffer (4 mol/L guanidine isothiocyanate, 25 mmol/L sodium acetate, pH 6.0, and 8 µL/mL ß-mercaptoethanol), and immediately frozen in liquid nitrogen until RNA extraction. The frozen cell lysates were then thawed, layered on top of a CsCl cushion (5.7 mol/L CsCl and 25 mmol/L sodium acetate, pH 6.0), and ultracentrifuged at 35,000 rpm for 16 h as previously described (29). The RNA pellet was ethanol precipitated, washed, resuspended in H2O, and stored at -20 C. All RNA samples were loaded onto denaturing 1% agarose minigels that contained 7.2% formaldehyde and subjected to electrophoresis to check quantity and integrity. RNA concentrations were determined by spectrophotometry.

Ribonuclease (RNase) protection assays

A 783-bp ST-3 complementary DNA was a gift from Dr. P. Chambon, INSERM, Institut de Chimie Biologique (Strasbourg, France). It was cloned into a pBluescript SK+ vector. Linearization with StuI resulted in a 255-bp fragment, and 234 bp were protected after RNase treatment. The IGF-II fragment, subcloned into a pBluescript SK+ plasmid, was provided by Dr. G. Bell, University of Chicago (Chicago, IL). Linearization with StuI yielded a 324-bp probe, and a 300-bp fragment was detectable in RNase protection assays. A 36B4-encoding fragment, inserted into a pGEM4 vector, was used as loading control, and a sequence of 200-bp was protected. RNase protection assays were performed as previously described (23). The gel was exposed to autoradiography film at -80 C for 2–4 days. Developed films were scanned, and the signal intensities of ST-3, IGF-II, and 36B4 bands were quantified by densitometry using NIH 1.59 software (NIH, Bethesda, MD).

Immunohistochemistry

ST-3 was immunodetected with a polyclonal rabbit antihuman antiserum (Ab 349), provided by Dr. P. Basset. The polyclonal anti-human IGF-II antibody was provided by Drs. R. Sportsman and J. Heisserman, Lilly Research Laboratories (Indianapolis, IN). The primary antibodies were detected with an alkaline phosphatase-linked secondary antibody according to the manufacturer‘s protocol (StrAvi-SuperSensitivity system, BioGenex, San Ramon, CA). In brief, tumor tissue samples were fixed in neutral buffered formalin. Five-micron sections from paraffin-embedded blocks of tumor tissues were deparaffinized and rehydrated. After blocking nonspecific binding with goat serum, the tissue sections were incubated with the primary antibody (Ab 349) at a dilution of 1:1500 or with the IGF-II antibody at a dilution of 1:1600 in PBS with 1% BSA and 0.1% acetic acid in a humidified chamber for 16 h at 4 C. The slides were then washed in PBS twice for 5 min each time. The primary antibody was detected by a 5-min incubation with a biotin-conjugated secondary antibody at 37 C. After two additional washing steps with PBS at 37 C for 5 min each, the tissue sections were covered with a film of alkaline phosphatase-labeled streptavidin and incubated at 37 C for another 5 min. After two additional washing steps, the immune reaction was visualized by the addition of Fast Red. The color reaction was controlled under the light microscope to prevent overstaining. The reaction was stopped by dipping the slides in phosphate-buffered saline. All slides were counterstained with hematoxylin and wet-mounted.

Statistical analysis

Statistical calculations were made using Blaise Statistical Software (Blaise Scientific, St. Louis, MO).


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
ST-3 mRNA expression in sets of matching fibroblasts from six breast cancer patients was studied by RNase protection assay (Fig. 1Go). Three fibroblast lines were examined from each patient: TF were grown from an area of malignant tumor, PF were derived from uninvolved breast tissue distant from the malignant lesion, and SKF were derived from overlying breast skin. Easily detectable ST-3 mRNA expression was seen in all six TF cell lines. Significant levels of ST-3 mRNA were also found in six skin-derived fibroblasts of the same breast cancer patients. In contrast to the pattern seen in TF and SKF, ST-3 mRNA expression in PF either was not detectable (in two of six cases) or was considerably weaker compared to the corresponding TF and SKF (in two of six cases). In only one set (G 60) was more ST-3 mRNA expressed in the PF samples (Fig. 1Go, upper panel, lane 9) than in both corresponding TF and SKF samples (Fig. 1Go, upper panel, lanes 8 and 10).



View larger version (59K):
[in this window]
[in a new window]
  		Figure 1. RNase protection assays for ST-3 mRNA expression in matched fibroblast sets. Each number represents one patient. 36B4, a ubiquitous human ribosomal protein, serves as the intraassay loading control. ST-3-expressing human placental tissue serves as positive control for ST-3 mRNA. The human epithelial breast cancer cell line MCF-7 serves as a negative control. Thirty micrograms of total RNA were hybridized to 32P-labeled antisense RNA probes for ST-3 and 36B4, then treated with RNase, yielding two specific bands of 234 and 220 bp (labeled ST-3 Fragment and 36B4 Fragment). All samples additionally display an artifactual band at the size of the probe for ST-3 (255 bp, labeled ST-3 Probe). These bands represent partially undigested probes, which may have hybridized to small amounts of sense transcripts from the complementary DNA template. Signal intensities of ST-3 and 36B4 gene expression were quantified by densitometry, and the relative ST-3/36B4 ratio of each matching set is shown on an arbitrary scale.

 
In Fig. 2Go we show ST-3 mRNA expression in 26 sarcomas and IGF-II mRNA expression in 30 sarcomas. With the exception of one sample (no. 107, upper panel, lane 9), all tumors analyzed for ST-3 expression yielded easily detectable message. In some cases, even very abundant ST-3 expression was seen [no. 1335 (lower panel, lane 13) and no. 1339 (lower panel, lane 15)]. In a matched tumor and normal sample from the same patient, ST3 mRNA was expressed in the malignant leiomyosarcoma (no. 103, upper panel, lane 6), but not in adjacent normal tissue (no.103N, upper panel, lane 7). When IGF-II gene expression was examined, we found a similar pattern. All 30 sarcomas expressed easily detectable IGF-II mRNA. Where both parameters were available (n = 26), we measured the signal intensities of ST-3 and IGF-II gene expression by gel densitometry and graded them on a semiquantitative scale (0 = no expression, 1+ = low expression, 2+ = intermediate expression, 3+ = high expression). For comparison of IGF-II and ST-3 expression, negative samples were grouped with low expressors, as were intermediate and high expressors. When the IGF-II and ST-3 scores were compared by densitometric measurement of the bands from the respective RNase protection assays, the level of expression for the two genes was significantly correlated (P < 0.035, by Mann-Whitney rank sum test).



View larger version (89K):
[in this window]
[in a new window]
  		Figure 2. RNase protection assays for ST-3 and IGF-II mRNA expression in sarcomas. Each number represents a tumor from a single patient. 103N, Biopsy from macroscopically unaffected tissue of patient 103. 36B4 served as an intraassay loading control. Thirty micrograms of total RNA were hybridized to 32P-labeled antisense RNA probes for ST-3, IGF-II, and 36B4, then treated with RNase, yielding specific bands of 234 bp (ST-3), 300 bp (IGF-II), and 220 bp (36B4). ST-3 and IGF-II assays were run separately.

 
Figure 3Go shows ST-3 mRNA expression in a number of epithelial breast and colon tumor cell lines. None of the 10 epithelial breast cancer cell lines investigated expressed ST-3 mRNA (Fig. 3AGo). In contrast, 5 of 8 epithelial colon cancer cell lines in Fig. 3BGo expressed ST-3 mRNA at various levels (LS 180, CaCo 2, HCT 15, SW 480, and SW 48). In addition, LCC-8, the only colon fibroblast cell line investigated (Fig. 3BGo, lane 14), and the benign colon adenoma cell line VACO 330 (Fig. 3BGo, lane 6) both expressed low, but detectable, ST-3 mRNA. The identity and significance of two strong bands of 240 bp in VACO 330 and CaCo2 (Fig. 3BGo, lanes 6 and 13) are unknown, but they might represent alternate splice products. No ST-3 gene expression was found in the two cell lines (Colo 201 and Colo 205) that had been established from ascites fluid (Fig. 3BGo, lanes 10 and 11).



View larger version (73K):
[in this window]
[in a new window]
  		Figure 3. RNase protection assays for ST-3 mRNA expression in breast cancer (A) and colon tumor cell lines (B). 36B4 serves as an intraassay loading control. Thirty micrograms of total RNA were hybridized to 32P-labeled antisense RNA probes for ST-3 and 36B4, then treated with RNase, yielding two specific bands of 234 and 220 bp (labeled ST-3 Fragment and 36B4 Fragment). The additional bands of 240 bp in lanes 6 and 13 in B might represent alternate splice products for ST-3.

 
ST-3 and IGF-II protein expression was detected in invasive breast cancer tissues and sarcomas by immunohistochemical analysis with a polyclonal antibody to the C-terminal portion of ST-3 and with a polyclonal antibody directed against human IGF-II, respectively. Figure 4Go, A and B, shows paraffin-embedded sections of invasive ductal breast cancers in which both IGF-II (A) and ST-3 (B) staining can be seen in stromal cells in the immediate vicinity of malignant epithelium exclusively. Neither malignant tumor epithelium (white arrows in A and B) nor stromal fibroblasts from microscopically uninvolved tissue at a distance from the tumor expressed IGF-II or ST-3 protein (data not shown). Figure 4CGo shows IGF-II protein expression in a cross-section of a malignant fibrous histiocytoma. Black arrows indicate cytoplasmic staining in malignant stromal cells. Normal stromal tissue in the vicinity of the malignant tumor did not express any IGF-II protein. An identical pattern of protein expression is shown for ST-3 in Fig. 4DGo, where dark cytoplasmic staining is confined to single, spindle-shaped, malignant cells in the same malignant fibrous histiocytoma as that in Fig. 4CGo.



View larger version (170K):
[in this window]
[in a new window]
  		Figure 4. ST-3 and IGF-II protein expression in invasive ductal breast carcinomas (A and B) and in a malignant fibrous histiocytoma (C and D). Brightfield photomicrographs show paraffin-embedded tissue sections stained with hematoxylin after immunohistochemical treatment with an antihuman IGF-II antibody (A and C) and an antihuman ST-3 antibody (B and D), as described in Materials and Methods. Black arrows indicate stromal staining for IGF-II (A) and ST-3 (B) in fibroblastic cells in the immediate proximity of areas of malignant breast epithelium. No staining is detectable in the malignant epithelial tumor cells (white arrows in A and B) or in stromal cells distant from the tumor (data not shown). Cytoplasmic staining for IGF-II can be seen in the malignant stromal cells in a cross-section of a fibrous histiocytoma (black arrows in C). D shows similar dark staining for ST-3 (black arrows) in fibroblast-like tumor cells of the same tumor as that in C.

 

   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Breast cancer, sarcomas, and colon cancer arise in tissues derived from embryonic ectoderm, mesoderm, and endoderm, respectively. The bilaminar germ disc comprising ectoderm and endoderm arises in the second week of fetal development, whereas the mesoderm develops by invagination of ectodermal cells in the third week of gestation (30).

In the present study we investigated coexpression of IGF-II and ST-3 in tumors that are ultimately derived from each of the three primordial germ layers. In breast cancer, IGF-II and ST-3 mRNA are confined to the stromal component of the tumor. ST-3 mRNA expression in matching fibroblast sets derived from three locations in breast cancer patients demonstrate a distinctive pattern. All cell lines derived from fibroblasts within the malignant breast tumor (TF) express easily detectable levels of ST-3 mRNA. High levels of ST-3 mRNA are also expressed by fibroblasts derived from overlying breast skin (SKF) of the same patients. In contrast, most fibroblast cell lines established from macroscopically unaffected breast tissue of the same individuals (PF) express clearly less ST-3 message. Only one of six matched sets varied from that pattern significantly. The ST-3 mRNA expression in the fibroblast sets closely matches the pattern of IGF-II expression that we previously reported in the same cells (24). There, too, IGF-II was predominantly expressed in tumor fibroblasts and skin fibroblasts, with little or no expression detected in fibroblasts from normal peripheral breast tissue. The ST-3 mRNA expression data are matched by immunohistochemical results, in which we detected ST-3 protein exclusively in single, spindle cell-shaped, stromal cells in the immediate neighborhood of invasive breast tumor epithelium (Fig. 4BGo). A strikingly similar pattern of protein expression can be seen for IGF-II in Fig. 4AGo, where detectable protein is confined to single stromal cells in the close vicinity of malignant epithelium. We did not detect any IGF-II or ST-3 protein expression in uninvolved stromal cells distant from the malignant tumor, which correspond to the PF samples in our RNase protection assays. These findings correspond well with those of our previous study that analyzed IGF-II expression in a series of 80 invasive breast carcinomas. In that study we demonstrated stromal expression of IGF-II in over 90% of the analyzed cases (31). Basset et al. published similar findings for ST-3, demonstrating expression of mRNA and protein in stromal fibroblasts immediately adjacent to malignant epithelium (19).

In agreement with previous reports, we were unable to observe ST-3 mRNA in any of the epithelial breast cancer cell lines analyzed (Fig. 3AGo). Again, this observation is analogous to the results IGF-II expression in breast cancer epithelial cell lines, which, with few exceptions, do not express IGF-II mRNA (23).

The patterns of ST-3 and IGF-II expression are not completely identical, suggesting that the two genes are not directly coregulated. Furthermore, IGF-II protein expression in igf-II-transfected IGF-II receptor-positive MCF-7 epithelial breast cancer cells (clone 8) is associated with phenotypic alterations, indicating an increased potential for malignant progression and metastasis (28). However, ST-3 mRNA expression is not seen in these cells, indicating that ST-3 expression is not directly regulated by IGF-II (MCF-7 in Fig. 1Go, top panel, lane 4, and clone 8 in Fig. 1Go, bottom panel, lane 12). These observations suggest that in breast cancer, both ST-3 and IGF-II expressions are regulated by paracrine effects from adjacent tumor epithelium, rather than by one protein (IGF-II) stimulating the expression of the other (ST-3) within the same cell through an autocrine mechanism. Basset et al. identified growth factors that are able to stimulate ST-3 gene expression in fetal fibroblasts (19). Among them, platelet-derived growth factor-B (PDGF-B) is a likely candidate for a growth factor that might be involved in the paracrine regulation of ST-3 gene expression. PDGF-B is expressed in epithelial breast cancer cell lines such as MCF-7; however, these cells lack PDGF receptors (32). Fibroblasts do not express the PDGF-B chain, but possess PDGF cell surface receptors, thus making them a possible target for paracrine PDGF action (33). Moreover, it has been demonstrated that, like ST-3, IGF-II mRNA and protein expression in fibroblasts are induced by PDGF (34, 35). Taken together, the previous experiments suggest possible mutual stimulation of ST-3 and IGF-II gene expression by PDGF via paracrine pathways. As both proteins are known to play important roles during normal fetal growth and development (19, 36, 37), the expression of ST-3 and IGF-II by adult fibroblasts might indicate a more primitive fetal-type behavior that has evolved under the paracrine influence of adjacent tumor epithelium.

The observation that ST-3 message and protein are present in sarcomas implies a different model in tumors of mesenchymal origin; in these tumors, ST-3 gene and protein expression are present in malignant stromal cells in the absence of adjacent epithelium. In agreement with previous work by Basset et al., who described ST-3 mRNA expression in fetal fibroblasts, it is possible that in sarcomas the malignant transformation of adult stromal cells is associated with regression to a fetal-type protein expression pattern. ST-3 expression in these cells would, therefore, be the result of some sort of reversion to a more primitive phenotype that is necessary for tissue remodeling and, therefore, normal and essential in fetal development. When ST-3 and IGF-II gene expressions were compared, we found that the amounts of expression correlated in most sarcomas (in 20 of 26); sarcomas that expressed high levels of IGF-II mRNA expressed similarly high levels of ST-3 mRNA, and tissues with low IGF-II mRNA expression usually expressed low levels of ST-3 mRNA (P < 0.05, by {chi}2 test).

In contrast to the finding in breast cancer and in sarcomas, where ST-3 expression is limited to stromal cells, we were able to detect ST-3 mRNA in five of eight malignant colon cancer cell lines using the sensitive and very specific RNase protection assay. An epithelial cell line derived from a benign colon adenoma (VACO 330) and the tumor-derived colon fibroblast cell line LCC-C8 both expressed low, but detectable, ST-3 message. Interestingly, two of the three ST-3 mRNA-negative cell lines (Colo 201 and Colo 205) had been established from ascites fluid of patients with colon carcinomas (Table 1Go). All other epithelial colon cancer cell lines were derived from primary tumors, where they had presumably remained under the continuous influence of paracrine stromal-epithelial interactions.


View this table:
[in this window]
[in a new window]
  		Table 1. Stromelysin-3 mRNA expression in human cell lines derived from benign and malignant colon tumors

 
No correlation was found between ST-3 gene expression and the level of cell differentiation based upon Dukes’ stage, tumorigenic potential, anchorage-independent soft agar growth, morphology, alkaline phosphatase, and carcino embryonic antigen activities (38). The detection of ST-3 mRNA in malignant epithelial colon cell lines somewhat contrasts with the finding of Newell et al., who located ST-3 message only in stromal cells surrounding malignant colorectal epithelium by in situ hybridization. No ST-3 mRNA expression was detectable in any of the sections of benign colon tissue (39).

The finding of ST-3 gene expression in colon cell lines reveals another parallel to IGF-II expression; in the normal gastrointestinal tract, IGF-II gene expression is high during fetal development and low in the adult (40). However, unlike epithelial breast cancer cell lines, virtually all colon carcinoma cell lines express significant amounts of IGF-II mRNA and protein (37, 41). This finding may indicate a different mechanism of gene regulation in these endodermally derived tumors.

Furthermore, two types of clones with a different tumorigenic potentials in nude mice have recently been isolated from a human colon carcinoma cell line (SSW 613-S); tumorigenic clones produced high levels of IGF-II, whereas nontumorigenic clones produced little IGF-II (42). We conclude that the finding of both ST-3 and IGF-II mRNA in malignant colon cell lines might infer that they represent markers of a more primitive phenotype of malignant colon epithelium. In addition, the observation of ST-3 mRNA in colon cell lines makes ST-3 a more ubiquitously expressed gene than previously described (43).

Taken together, the data presented in this study support the hypothesis that in malignant tumors the coexpression of ST-3 and IGF-II, two developmentally regulated proteins that are physiologically expressed during fetal growth and development, is associated with the regression to a more primitive, fetal-type phenotype. The combined expression of these two proteins could serve to enhance tumor growth and facilitate tumor invasion. Further studies will examine the consequence of this phenotypic change and the mechanisms by which both proteins might contribute to tumor progression.

Colon adenoma and adenocarcinoma cell lines are listed in order of decreasing level of differentiation based upon Dukes’ stage, morphology, tumorigenic potential, monolayer polarity, alkaline phosphatase expression, and carcino embryonic antigen production (35).


   Footnotes
 
1 This work was supported by NIH Grants CA-58441 and CA-55003. Back

Received January 7, 1997.

Revised March 4, 1997.

Accepted March 10, 1997.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Mignatti P, Rifkin DB. 1993 Biology and biochemistry of proteinases in tumor invasion. Physiol Rev. 73:161–195.[Free Full Text]
  2. Duffy MJ. 1992 The role of proteolytic enzymes in cancer invasion and metastasis. Clin Exp Metast. 10:145–155.[CrossRef][Medline]
  3. Liotta A, Stetler-Stevenson WG. 1991 Invasion and metastasis: an imbalance of positive and negative regulation. Cancer Res. 51(Suppl):5054s–5059s.
  4. Clark WH. 1991 Tumor progression and the nature of cancer. Br J Cancer. 64:631–644.[Medline]
  5. Fidler IJ. 1990 Critical factors in the biology of human cancer metastasis. 28th G. H. A. Clowes Memorial Award Lecture. Cancer Res. 50:6130–6138.[Abstract/Free Full Text]
  6. Nakajima M, Chop AM. 1991 Tumor invasion and extracellular matrix degradative enzymes: regulation of activity by organ factors. Semin Cancer Biol. 2:115–127.[Medline]
  7. Matrisian LM, Bowden GT. 1990 Stromelysin/transin and tumor progression. Semin Cancer Biol. 1:107–115.[Medline]
  8. Gottesman MM. 1990 Do proteases play a role in cancer? Semin Cancer Biol. 1:97–98.
  9. Birkedal-Hansen H, Moore WGI, Bodden MK, et al. 1993 Matrix metalloproteinases: a review. Crit Rev Oral Biol Med. 4:197–250.[Abstract/Free Full Text]
  10. Matrisian LM. 1992 The matrix-degrading-metalloproteinases. Bioessays 14:455–463.
  11. Nagase H, Barrett AJ, Woessner FJ. 1992 Nomenclature and glossary of the matrix metalloproteinases. Matrix. 1(Suppl):421–424.
  12. Liotta LA, Stetler-Stevenson WG. 1990 Metalloproteinases and cancer invasion. Semin Cancer Biol. 1:99–106.[Medline]
  13. Murphy G, Hembry RM, Hughes CE, Fosang AJ, Hardingham TE. 1990 Role and regulation of metalloproteinases in connective tissue turnover. Biochem Soc Transcr. 18:812–815.
  14. Woessner FJ. 1991 Matrix metalloproteinases and their inhibitors in connective tissue remodeling. FASEB J. 15:2145–2154.
  15. Liotta LA, Tyrggvason K, Garbisa S, Foltz DH, Shafie S. 1980 Metastatic potential correlates with enzymatic degradation of basement membrane collagnes. Nature. 284:67–68.[CrossRef][Medline]
  16. DeClerck YA, Perez N, Shimada H, Boone TC, Langley KE, Taylor SM. 1992 Inhibition of invasion and metastasis in cells transfected with an inhibitor of metalloproteinases. Cancer Res. 52:701–708.[Abstract/Free Full Text]
  17. Schultz RM, Silberman S, Persky B, Bajkowski AS, Carmichel DF. 1988 Inhibition by human recombinant tissue inhibitor of metalloproteinases of human amnion invasion and lung colonization by murine B16–F10 melanoma cells. Cancer Res. 48:5539–5545.[Abstract/Free Full Text]
  18. Albini A, Melchiori A, Santi L, Liotta LA, Brown PD, Stetler-Stevenson WG. 1991 Tumor cell invasion inhibited by TIMP-2. J. Natl. Cancer Inst. 83:775–779.[Abstract/Free Full Text]
  19. Basset P, Bellocq JP, Wolf C, et al. 1990 A novel metalloproteinase gene specifically expressed in stromal cells of breast carcinomas. Nature. 348:699–704.[CrossRef][Medline]
  20. Pei D, Majmudar G, Weiss SJ. 1994 Hydrolytic inactivation of a breast carcinoma cell-derived serpin by human stromelysin-3. J Biol Chem. 296:25849–25855.
  21. Murphy G, Segain J-P, O’Shea M, et al. 1993 The 28-kDa N-terminal domain of mouse stromelysin-3 has the general properties of a weak metalloproteinase. J Biol Chem. 268:15435–15441.[Abstract/Free Full Text]
  22. Wolf C, Rouyer N, Lutz Y, et al. 1993 Stromelysin 3 belongs to a subgroup of proteinases expressed in breast carcinoma fibroblastic cells and possibly implicated in tumor progression. Proc Natl Acad Sci USA. 90:1843–1847.[Abstract/Free Full Text]
  23. Yee D, Cullen KJ, Paik S, et al. 1988 Insulin-like growth factor II mRNA expression in human breast cancer. Cancer Res. 48:6691–6696.[Abstract/Free Full Text]
  24. Singer C, Rasmussen A, Smith HS, Lippman ME, Lynch HT, Cullen KJ. 1995 Malignant breast epithelium selects for insulin-like growth factor II expression in breast stroma: evidence for paracrine function. Cancer Res. 55:2448–2454.[Abstract/Free Full Text]
  25. Giudice LC, de Zegher F, Gargosky SE, et al. 1995 Insulin-like growth factors and their binding proteins in the term and the preterm human fetus and neonate with normal and extremes of intrauterine growth. J Clin Endocrinol Metab. 80:1548–1555.[Abstract/Free Full Text]
  26. Muller D, Wolf C, Abecassis J, et al. 1993 Increased stromelysin 3 gene expression is associated with increased local invasiveness in head and neck squamous cell carcinomas. Cancer Res. 53:165–169.[Abstract/Free Full Text]
  27. Rouyer N, Wolf C, Chenard MP, et al. 1994–1995 Stromelysin-3 gene expression in human cancer: an overview. Invas Metastasis. 24:269–275.
  28. Cullen KJ, Lippman ME, Chow D, Hill S, Rosen N, Zwiebel JA. 1992 Insulin-like growth factor-II overexpression in MCF-7 cells induce phenotypic changes associated with malignant progression. Mol Endocrinol. 6:91–100.[Abstract]
  29. Chirgwin JM, Przybla AE, MacDonald RJ, Rutter WJ. 1979 Isolation of biologically active ribonucleic acid from sources enriched in ribonucleases. Biochemistry. 18:5294–5299.[CrossRef][Medline]
  30. Sadler TW. 1995 Langman’s medical embryology. Baltimore: Williams and Wilkins; 41–57.
  31. Giani C, Cullen KJ, Campani D, Rasmussen A. 1996 IGF-II mRNA and protein are expressed in the stroma of invasive breast cancers: an in situ hybridization and immunohistochemistry study. Breast Cancer Res Treat. 41:43–50.[CrossRef][Medline]
  32. Bronzert DA, Pantazis P, Antoniades HN, et al. 1987 Synthesis and secretion of platelet derived growth factor by human breast cancer cell lines. Proc Natl Acad Sci USA. 84:5763–5767.[Abstract/Free Full Text]
  33. Heldin CH, Ek B, Ronnestrand L. 1983 Characterization of the receptor for platelet-derived growth factor in human fibroblasts. Demonstration of an inmate relationship with a185 :000-dalton substrate for the platelet-derived growth factor-stimulated kinase. J Biol Chem. 258:10054–10061.
  34. Cullen KJ, Smith HS, Hill S, Paik S, Rosen N, Lippman ME. 1991 Growth factor mRNA expression in human fibroblasts from benign and malignant lesions. Cancer Res. 51:4978–4985.[Abstract/Free Full Text]
  35. Clemmons DR. 1984 Multiple hormones stimulate the production of somatomedin by cultured human fibroblasts. J Clin Endocrinol Metab. 58:850–856.[Abstract]
  36. Han VKM, D’Ercole AJ, Lund PK. 1987 Cellular localization of somatomedin messenger RNA in the human fetus. Science. 236:193–197.[Abstract/Free Full Text]
  37. Scott J, Cowell J, Robertson ME, et al. 1985 Insulin-like growth factor-II gene expression in Wilms‘ tumor and embryonic tissues. Nature. 317:260–263.[CrossRef][Medline]
  38. Shabahang M, Buras RR, Davoodi F, Schumaker LM, Nauta RJ, Evans SR. 1993 1,25-Dihydroxyvitamin D3 receptor as a marker of human colon carcinoma cell line differentiation and growth inhibition. Cancer Res. 53:3721–3728.
  39. Newell KJ, Witty JP, Rodgers WH, Matrisian LM. 1994 Expression and localization of matrix-degrading metalloproteinases during colorectal tumorigenesis. Mol Cancer. 10:199–206.
  40. Guo YS, Jin GF, Townsend CM, et al. 1995 Insulin-like growth factor-II expression in carcinoma in colon cell lines: implications for autocrine actions. J Am Coll Surg. 181:145–154.[Medline]
  41. Singh P, Dai B, Yallampalli C, Xu Z. 1994 Expression of IGF-II and IGF-binding proteins by colon cancer cells in relation to growth response to IGFs. Am J Physiol. 267:608–617.
  42. Lamonerie T, Lavialle C, de-Galle B, Binoux M, Brison O. 1995 Constitutive or inducible overexpression of the IGF-II gene in cells of a human colon carcinoma cell line. Exp Cell Res. 261:342–351.
  43. Urbansky SJ, Edwards DR, Hershfield N, et al. 1993 Expression pattern of metalloproteinases and their inhibitors changes with the progression of human sporadic colorectal neoplasia. Diagn Mol Pathol. 2:81–89.[Medline]



This article has been cited by other articles:


Home page
 Clin. Cancer Res.Home page
A. Orimo, Y. Tomioka, Y. Shimizu, M. Sato, S. Oigawa, K. Kamata, Y. Nogi, S. Inoue, M. Takahashi, T. Hata, et al.
Cancer-associated Myofibroblasts Possess Various Factors to Promote Endometrial Tumor Progression
Clin. Cancer Res., October 1, 2001; 7(10): 3097 - 3105.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
D. Luo, E. Guerin, M.-G. Ludwig, I. Stoll, P. Basset, and P. Anglard
Transcriptional Induction of Stromelysin-3 in Mesodermal Cells Is Mediated by an Upstream CCAAT/Enhancer-binding Protein Element Associated with a DNase I-hypersensitive Site
J. Biol. Chem., December 24, 1999; 274(52): 37177 - 37185.
[Abstract] [Full Text] [PDF]

Home page
 Mol. Cell. Biol.Home page
F. Frasca, G. Pandini, P. Scalia, L. Sciacca, R. Mineo, A. Costantino, I. D. Goldfine, A. Belfiore, and R. Vigneri
Insulin Receptor Isoform A, a Newly Recognized, High-Affinity Insulin-Like Growth Factor II Receptor in Fetal and Cancer Cells
Mol. Cell. Biol., May 1, 1999; 19(5): 3278 - 3288.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Submit a related Letter to the Editor
Right arrow Purchase Article
Right arrow View Shopping Cart
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Copyright Permission
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Singer, C. F.
Right arrow Articles by Cullen, K. J.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Singer, C. F.
Right arrow Articles by Cullen, K. J.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Endocrinology Endocrine Reviews J. Clin. End. & Metab.
Molecular Endocrinology Recent Prog. Horm. Res. All Endocrine Journals