HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text Full Text (PDF) All Versions of this Article: fj.05-5500fjev1 20/9/1510    most recent Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Cella, N. Articles by Zhang, M. Search for Related Content PubMed PubMed Citation Articles by Cella, N. Articles by Zhang, M.

Maspin is physically associated with ßbeta;1 integrin regulating cell adhesion in mammary epithelial cells

Baylor College of Medicine, Department of Molecular and Cellular Biology, One Baylor Plaza, Houston, Texas, USA

¹Correspondence: Baylor College of Medicine, Department of Molecular and Cellular Biology, ALKEK Bldg., Rm. N630, One Baylor Plaza, Houston, TX 77030, USA. E-mail: mzhang{at}bcm.tmc.edu

SPECIFIC AIMS

The purposes of our study are to investigate the role of maspin in cell adhesion and to identify the molecular domain involved in this process and the underlying mechanism.

PRINCIPAL FINDINGS

1. Maspin rapidly modulates cell adhesion
In a different approach from most previous studies, we have used a nontransformed mammary epithelial cell line (MCF-10A cells), which expresses high concentration of maspin. Since adhesion primary involves the interaction between the cell membrane and the substrate, we first investigated whether maspin was present on the cell surface by immunofluorescence. Using nonpermeabilized cells to prevent internalization of the antibody (Ab), we observed maspin staining at the periphery of the cell and confirmed its expression on the external side of the membrane. We next analyzed a possible role of endogenous maspin in cell adhesion as well as exogenous addition of recombinant maspin by cell adhesion assays. We found that incubation of cells with antimaspin or maspin down-regulation by RNAi decreased cell adhesion. Accordingly, incubation of cells with recombinant maspin increased cell adhesion. These effects were observed within 30 min, suggesting that maspin acts in the early steps of the cell adhesion process.

2. Mutation analyses reveal a new domain involved in maspin effect on cell adhesion
Previous studies done in tumor cell models indicate that the maspin RSL domain (reactive site loop) is essential for its positive effect on adhesion. To determine the domain of maspin, we tested the effect of various maspin mutants on cell adhesion. Differently from the studies in mammary tumors, we found that a domain comprising amino acid 139 to 225, and not the RSL domain, retained a positive effect on cell adhesion.

3. Maspin is physically associated with ßbeta;1 integrin
Cell adhesion to extracellular matrix is mediated by integrins to a great deal. In our model system, we used the endogenous matrix deposited by MCF-10A cells, which contains predominantly laminin-5. Therefore, we hypothesized that maspin could be modulating cell adhesion via a ßbeta;1 integrin, a major laminin-5 receptor in this cell line. Using total protein extracts, we observed that ßbeta;1 integrin coimmunoprecipitated with maspin. Reciprocally, maspin was also coimmunoprecipitated with ßbeta;1 integrin.

MCF-10A adhesion to its endogenous matrix could be inhibited 50% with a function-blocking anti-ßbeta;1 integrin. Interestingly, this effect was abrogated when cells were incubated with maspin before addition of the Ab, indicating that maspin could override the inhibition given by the anti-ßbeta;1 integrin.

4. Maspin colocalizes with ßbeta;1 integrin
To further confirm and extend our findings, we performed indirect double-immunofluorescence to analyze the staining pattern of maspin and ßbeta;1 integrin by confocal microscopy. We observed a significant colocalization of these two molecules, predominantly at the cell membrane site, which was further confirmed by quantitative analysis. Since both integrin expression and signaling are altered in cells grown on flat surfaces, we intend to confirm this result in cells cultivated in matrigel, in which cells can develop into three-dimensional acinar structures which recapitulate typical features of the epithelial architecture. Similarly to cells grown on flat surfaces, maspin and ßbeta;1 integrin also colocalized in three-dimensional culture, which was observed for both basement membrane and cell-cell contact sites.

5. Maspin is associated with the Triton X-100 insoluble cytoskeleton fraction
Binding of integrin receptors to extracellular ligands involves receptor-ligand interactions at the cell-substrate interface and assembly of cytoskeletal and adhesion plaque proteins. The rapid cellular response to recombinant maspin indicates that maspin acts on the initial steps of cell-substrate interaction and could be connected to the underlying cytoskeleton framework. To test this possibility, we determined maspin’s solubility in Triton X-100, a nonionic detergent. Maspin was detected in the soluble as well as in the insoluble Triton X-100, indicating that a fraction of it is associated with the cytoskeleton. In addition, immunofluorescence analysis revealed that the insoluble maspin fraction is present at the periphery of the cell, indicating that maspin is connected to the cortical membrane cytoskeleton and may be part of the supramolecular structure of the adhesion plaque.

CONCLUSIONS AND SIGNIFICANCE

Our study forms the following conclusions: 1) maspin acts rapidly to increase cell adhesion, indicating that it takes part of the early steps of the cell adhesion process; 2) differently from what was observed in tumor and stromal models, we have found that maspin effect on adhesion does not involve the RSL domain; 3) maspin is physically associated with ßbeta;1 integrin, suggesting that it may modulate its ability to mediate cell adhesion. In agreement with that, these molecules colocalize on the cell membrane site; 4) a fraction of maspin is associated with the Triton X-100-insoluble fraction, suggesting that it is part of the supramolecular structure of adhesion plaques.

In our nontransformed cell model, we find that maspin acts rapidly and independently of the RSL domain. This is in apparent contrast with studies done with stromal and tumor cells, where maspin-induced increase in cell adhesion depended on RSL, and the effect was observed after 18–24 h of incubation with recombinant maspin. In addition, one previous study observed an increase in integrin mRNA expression following a long-term treatment of maspin, whereas in our model we found that maspin and ßbeta;1 integrin were physically linked. While this difference may be due to the use of different cellular models, it may also indicate that maspin influence ßbeta;1 via two different manners: directly, modulating ßbeta;1 integrin function in the cell membrane microenvironment, and indirectly, up-regulating its expression. Alternatively, maspin could act differently in normal vs. tumor or stromal cells. These are interesting hypotheses that remain to be tested. Finally, it is interesting to observe that, in addition to the new domain described here, two distinct domains are responsible for the effect of maspin on angiogenesis and interaction with collagens, respectively, suggesting that maspin exerts its effects via different molecular mechanisms.

Our cell fractionation analysis indicates that maspin is present in a soluble fraction as well as in a detergent-insoluble cytoskeleton fraction, suggesting that maspin may be part of the adhesion plaque. Whether the maspin-insoluble fraction is intracellular or the extracellular maspin pool is linked to the cytoskeleton via its interaction with ßbeta;1 integrin remains to be investigated.

ßbeta;1 integrin has been proven vital for several mammary epithelial cell functions, including growth, differentiation, apoptosis, and cancer progression. In agreement with these observations, targeted maspin expression in the mammary gland resulted in disrupted lobular-alveolar structure, and in tumor models maspin can block tumor growth and invasion.

In conclusion, our study provided evidence that maspin and ßbeta;1 integrin may be part of a common signaling machinery, which plays a crucial role in mammary gland function and in tumor initiation and progression.

View larger version (17K): [in this window] [in a new window]   Figure 1. Polyclonal antimaspin Ab, but not an anti-RSL Ab generated against the RSL domain of maspin, inhibits MCF-10A cell adhesion to the self-deposited matrix. A) MCF-10A cells were plated on coverslips and stained with the rabbit anti-RSL (AbS4A) Ab on ice, to prevent the internalization of the Ab. Preimmune serum revealed no staining (data not shown). This imagine is representative of two independent assays. Confocal microscope analysis revealed maspin on the cell surface. Bar, 20 µm. B) MCF-10A cells were harvested with enzyme-free dissociation buffer, and preincubated with the indicated antibodies. 2 x 104 cells/well were seeded and cell adhesion was measured after 30 min using colorimetric reaction. Polyclonal anti-maspin inhibited cell adhesion by 76%, whereas no effect was detected with the anti-RSL Ab. Control, preimmune serum. Each sample was measured in triplicate. Result is representative of 3 independent assays. Statistical analysis was done by a t test (P <0.05%).

View larger version (36K): [in this window] [in a new window]   Figure 2. Maspin and ßbeta;1 integrin coimmunoprecipitate and colocalize in MCF-10A cells grown in monolayers and in 3D culture. A) MCF-10A cells were chemically cross-linked, lysed, and 500 µg of protein extracts were immunoprecipitated with the indicated antibodies. An irrelevant rabbit antiserum was used as a negative control (lanes 3 and 6). Immunoprecipitates were separated in SDS-PAGE gels as detailed in Material and Methods and analyzed by Western blot as indicated. *Pre-ßbeta;1 integrin. B) MCF-10A cells were plated on coverslips, fixed, permeabilized, and stained with a mouse monoclonal antibody anti-human maspin (a) and a rabbit polyclonal anti-ßbeta;1 integrin (b). Nucleus is shown in blue (c) and merged image is shown in (d). Arrows indicate that maspin and ßbeta;1 integrin are located in the periphery of the cell (a and b, respectively) and are colocalized (d, Robs=0.704/Rrand=0.060±0.012); Bar 10 µm. C) MCF-10A cells were embedded in the Matrigel and allowed to form acini for 15 d. Cryosections (8 µm) were prepared and stained for maspin (a) and ßbeta;1 integrin (b), using the same antibodies mentioned above. Nuclei are shown in blue (c) and merged image is shown in (d). Arrow and arrowhead in (a) and (b) indicate localization on the basal membrane and in sites of cell–cell contact, respectively. Arrows in (d) indicate sites of maspin and ßbeta;1 integrin colocalization (Robs=0.580/Rrand=0.115±0.019); Bar 20 µm.

View larger version (25K): [in this window] [in a new window]   Figure 3. Schematic diagram illustrates and summarizes experiments and main conclusions. A) Endogenous maspin modulates MCF-10A cell adhesion, which involves its physical interaction with beta 1 integrin and with cytoskeleton. Addition of anti-maspin down-regulates cell adhesion. B) MCF-10A cells in suspension were incubated with recombinant GST-maspin, which modulates beta1 integrin and cytoskeleton function, leading to an increase in cell adhesion to extracellular matrix. This activity is independent of maspin RSL domain (not depicted).

FOOTNOTES

To read the full text of this article, go to http://www.fasebj.org/cgi/doi/10.1096/fj.05-5500fje

This article has been cited by other articles:

				L.-j. Shao, H. Y. Shi, G. Ayala, D. Rowley, and M. Zhang Haploinsufficiency of the Maspin Tumor Suppressor Gene Leads to Hyperplastic Lesions in Prostate Cancer Res., July 1, 2008; 68(13): 5143 - 5151. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text Full Text (PDF) All Versions of this Article: fj.05-5500fjev1 20/9/1510    most recent Alert me when this article is cited Alert me if a correction is posted Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Cella, N. Articles by Zhang, M. Search for Related Content PubMed PubMed Citation Articles by Cella, N. Articles by Zhang, M.