| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Centro di Endocrinologia ed Oncologia Sperimentale, Consiglio Nazionale delle Ricerche (G.R.), Dipartimento di Biologia e Patologia Cellulare e Molecolare (N.M., G.R., M.V.), Dipartimento di Endocrinologia ed Oncologia Molecolare e Clinica (F.M., S.D.R., G.F.F.), Università Federico II, 80131 Naples, Italy; and the Molecular Pathology Section, Laboratory of Pathology, National Cancer Institute, National Institutes of Health (M.E.S.), Bethesda, Maryland 20892
Address all correspondence and requests for reprints to: Nunzia Montuori, M.D., Dipartimento di Biologia e Patologia Cellulare e Molecolare, via S. Pansini 5, 80131 Naples, Italy. E-mail: mavitale{at}cds.unina.it
| Abstract |
|---|
|
|
|---|
1ß1,
2ß1,
3ß1,
6ß1, and
6ß4 was found in all tumor specimens and
in tumor cell lines, whereas normal thyroid cells and TAD-2 cells
lacked the expression of
6ß4. Despite the
presence of several integrin laminin receptors, adhesion of TAD-2, NPA,
and ARO cells to immobilized laminin-1 was poor, whereas WRO cells and
follicular carcinoma-derived cells displayed a strong adhesion. Indeed,
WRO and follicular carcinoma-derived cells showed expression of a
nonintegrin laminin receptor, the 67-kDa high affinity laminin receptor
(67LR). TAD-2, NPA, and ARO cells as well as nodular goiter, toxic
adenoma, follicular adenoma, and papillary carcinoma-derived cells did
not express the 67LR. Adhesion of WRO and follicular carcinoma-derived
cells to laminin-1 was specifically inhibited by a recombinant
polypeptide containing laminin-binding domains of 67LR, demonstrating
that this receptor confers to follicular carcinoma cells attachment
capacity to laminin. Moreover, tissue specimens from follicular
carcinomas expressed the 67LR, whereas follicular adenomas and normal
thyroid tissues were negative. In thyroid tumors, integrin receptors,
although abundant, participate weakly in adhesion to laminin. The
expression in follicular carcinoma cells of a functional, high affinity
67LR together with nonfunctional integrin LM receptors could be
responsible for the tendency of follicular carcinoma cells to
metastasize by mediating stable contacts with basal membranes. | Introduction |
|---|
|
|
|---|
1ß1
1), LM-2/merosin
(
2ß1
1), and LM-5/kalinin
(
3ß3
2). Epithelial and
endothelial cells adhere to the basement membrane mostly through
integrins, heterodimeric transmembrane proteins present on the cell
surface. Integrins are characterized by a common ß-chain associated
with one of the variant
-chains conferring the ligand specificity
for proteins of the extracellular matrix, such as collagen, LM (LM),
fibronectin, and vitronectin (1). Each integrin is a receptor for one
or more extracellular matrix components. Both integrin expression and
binding specificity are cell type related (2) and undergo modifications
upon differentiation, transformation, and cytokine induction (3, 4).
Thyroid follicular cells are polarized epithelial cells whose
ß1 integrin expression is restricted to
3ß1 at the basal site of the cell membrane
in vivo, whereas cultured thyrocytes also express
2ß1 (4, 5, 6). Upon neoplastic
transformation, follicular organization is disrupted, integrin
expression is changed, and polarized distribution is lost (7). Thyroid
carcinomas display aberrant expression of all
1ß1-
6ß1
integrins as well as of
6ß4, the major LM
receptor in squamous and columnar epithelia that is involved also in
hemidesmosome assembly (8).
Besides integrins, other nonintegrin LM receptors are present in epithelial cells. Among the several nonintegrin LM receptors, the 67-kDa high affinity LM receptor (67LR) is responsible for high affinity interactions (Kd = 10-9 mol/L) between cells and LM-1 (9, 10). The 67LR, which has been purified from several cell lines as well as from normal and neoplastic tissues, is poorly expressed in normal tissues and benign tumors, whereas its expression is dramatically increased in metastatic cancer cells (11). The 67LR expressed on the cell surface derives from posttranslational modifications of a 37-kDa cytosolic precursor (37 LRP) (12, 13). Two LM-1-binding domains have been localized on 37LRP (14, 15, 16), peptide G (residues 161180), and peptide 11 (residues 205229).
In this study we investigated the expression and function of integrin LM receptors and 67LR in differentiated thyroid carcinomas and in some thyroid carcinoma cell lines. We found that only follicular carcinoma-derived cells express the 67LR, by which cells attach to LM-1.
| Materials and Methods |
|---|
|
|
|---|
Fresh tissues specimens from normal thyroid gland, nodular goiters, toxic adenomas, Hurthle cell adenomas, follicular carcinomas, and papillary carcinomas were obtained from patients undergoing thyroidectomy. Diagnosis was confirmed by histology. Within a few hours after surgery, tissues were minced and digested with type IV collagenase (Sigma Chemical Co., St. Louis, MO), 1.25 mg/mL in 0.5% BSA in DMEM-Hams F-12 medium (BSA-F12), for 810 h at 8 C under rotation. Cells were centrifuged at 150 x g for 5 min, washed twice in BSA-F12 medium, and immediately analyzed by flow cytometry or alternatively cultured in vitro. An immortalized human fetal thyroid cell line, obtained by simian virus 40 infection (TAD-2), was donated by Dr. T. F. Davies, Mount Sinai Hospital (New York, NY); thyroid papillary NPA, follicular WRO, and anaplastic ARO carcinoma cell lines were donated by Dr. G. J. F. Juillard, University of California (Los Angeles, CA). All cell lines and primary cultures were cultured in a 5% CO2 atmosphere at 37 C in RPMI medium supplemented with 10% FCS. Medium was changed every 34 days. Cells to be examined were detached by 0.5 mmol/L ethylenediamine tetraacetate (EDTA) in calcium- and magnesium-free phosphate-buffered saline with (trypsin/PBS) or without (EDTA/PBS) 0.05% trypsin. Trypsin does not affect integrins because these molecules do not possess trypsin-sensitive sites (3).
Reagents
Monoclonal antibodies (MoAbs) against integrin subunits were
purchased or donated: anti-ß1: A1A5, Dr. M. E.
Hemler (Boston, MA); anti-
1: TS2/7, Dr. F.
Sanchez-Madrid (Madrid, Spain); anti-
2: 10G11, Dr.
A. E. G. Kr. von dem Borne (Amsterdam, The Netherlands);
anti-
3: J143, Dr. L. J. Old (New York, NY);
anti-
6: GoH3, Dr. A. Sonnenberg (Amsterdam, The
Netherlands); and anti-ß4: 3E1 Telios (San Diego, CA).
Fluorescein-conjugated goat antimouse IgG was purchased from Ortho
(Raritan, NJ).
A recombinant polypeptide was produced for competition experiments and for the production of anti-67LR antibodies. Briefly, complementary DNA coding for the 67LR cytosolic precursor 37LRP (9, 10) was cloned into the pTrcHis B expression vector (Invitrogen, San Diego CA) and expressed in TOP-10 bacteria (Invitrogen). The recombinant 37LRP (r37LRP), containing both LM-binding domains of 67LR (15, 16), was purified on nickel affinity columns, according to the procedures specified by Invitrogen, dialyzed in 50 mmol/L Tris (pH 7.5)-0.1% Triton X-100, and adjusted at a concentration of 1 mg/mL. An antiserum was produced as previously described by rabbit immunization with SDS-PAGE-electroeluted r37LRP (9). Anti-r37LRP antibody was affinity purified by r37LRP-conjugated Affigel 10 resin according to a previously described procedure (9). The affinity-purified polyclonal antibody reacted with both 37LRP and 67LR from whole cell lysates in immunoblots and was able to immunoprecipitate 67LR from surface-labeled cell lysates.
Flow cytometric analysis
Cell suspensions obtained by collagenase digestion were centrifuged at 150 x g for 5 min and washed once in 0.5% BSA in PBS (BSA/PBS), and erythrocytes were lysed with a NH4Cl solution (Ortho-mune lysing reagent, Ortho), washed twice in BSA/PBS, and filtered through a nylon mesh to remove clumps. Analysis of integrin expression in monodispersed cell suspensions from thyroid carcinomas was performed as previously described (5). Briefly, single cell suspensions obtained from collagenase-treated carcinoma cells from subconfluent cultures were incubated with specific MoAb for 1 h at 4 C in BSA/PBS, washed in the same buffer, and incubated again with the secondary fluorescein-conjugated antibody for 30 min at 4 C. Cells were resuspended in PBS and analyzed by flow cytometry using a FACScan (Becton Dickinson and Co., Mountain View, CA). Nonspecific mouse Igs of the same isotype of MoAbs were used as controls. The expression of each integrin subunit was represented as: relative fluorescence index = experimental mean fluorescence/control mean fluorescence.
Cell attachment assay to LM
The assay was performed in 96-well flat-bottomed microtiter plates (Costar, Cambridge, MA). The wells were filled with 100 µL of the appropriate dilution in PBS of LM-1/EHS (Collaborative Research, Bedford, MA), and after overnight incubation at 4 C, the plates were washed with PBS, filled with 100 µL 1% heat-denatured BSA, and incubated for 1 h at room temperature. The plates were then washed and filled again with 100 µL/well PBS, 0.9 mmol/L CaCl2, and 0.5 mmol/L MgCl2 containing 5 x 104 cells obtained by EDTA/PBS incubation from subconfluent cultures. After 30 min at 37 C, plates were gently washed three times with PBS, and attached cells were fixed with 3% paraformaldehyde for 10 min followed by 2% methanol for 10 min and finally stained with 0.5% crystal violet in 20% methanol. After 10 min, the plates were washed with tap water, the stain was eluted with a solution of 0.1 mol/L sodium citrate, pH 4.2, in 50% ethanol, and the absorbance at 540 nm was measured by a spectrophotometer.
In the adhesion inhibition assay, 5 x 104 cells/well were coincubated with 10 µg r37LRP in plates previously coated with 0.5 µg LM-1. All experiments were performed in quadruplicate. Results are presented as the mean ± SD.
Western blot
Cells grown to near confluence in 100-mm dishes were incubated for 10 min at 4 C in 1 mL lysis buffer [50 mmol/L Tris (pH 7.4), 0.5% Nonidet P-40, and 0.01% SDS] containing protease inhibitors. Cell lysates were collected by scraping and centrifuged at 12,000 x g for 5 min at 4 C. Frozen tissue specimens derived from normal thyroid gland, follicular adenomas, and follicular carcinomas were homogenized by Polytron (Brinkmann Instruments, Inc., Westbury, NY) in lysis buffer containing protease inhibitors. Tissue homogenates were incubated for 20 min at 4 C and centrifuged at 20,000 rpm for 30 min at 4 C. The protein concentration in cell lysates and in clarified tissue extracts was determined, and 30 µg total protein were incubated for 5 min at 90 C in Laemmli sample buffer. Cell lysates and clarified tissue extracts were electrophoresed on 10% SDS-polyacrylamide gels under reducing conditions (17). Gels were electroblotted, and the membranes were blocked with 5% nonfat dry milk, 1% ovalbumin, 5% FCS, and 7.5% glycine. After three washes, the membranes were incubated overnight at 4 C with polyclonal affinity-purified anti-67LR antibody in PBS. After three washes, filters were incubated for 30 min at room temperature with horseradish peroxidase-conjugated goat antirabbit antibody (Bio-Rad Laboratories, Inc., Richmond CA) diluted 1:2000 in PBS. The membranes were then washed as described above and stained using an enhanced chemiluminescence system (Amersham, Aylesbury, UK).
Surface biotinylation of intact cells and immunoprecipitation
TAD-2 and WRO cells grown to near confluence in 100-mm dishes were washed three times with PBS containing Ca2+ and Mg2+ and incubated for 1 h at room temperature with 0.5 mg/mL NHS-LC-biotin (Pierce Chemical Co., Rockford, IL) in PBS. Labeling medium was then removed, and the reaction was stopped by the addition of 50 mg/mL glycine in PBS for 10 min at room temperature. Surface-labeled cells were washed three times with cold PBS and lysed in 1 mL lysis buffer containing protease inhibitors. A total of 100 µg protein were incubated for 2 h at 4 C with 1 µg polyclonal affinity-purified anti-67LR antibody or 1 µg control rabbit IgG and then with 100 µL 25% protein A-Sepharose (Pharmacia Biotech, Uppsala, Sweden) for 2 h at 4 C. Immune complexes were washed five times with lysis buffer, eluted in 40 µL Laemmli sample buffer for 5 min at 90 C, and loaded on a 10% SDS-polyacrylamide gel in reducing conditions. The biotin-labeled antibody-reacting proteins were detected, after electroblotting of the gel, by incubation with horseradish peroxidase-conjugated streptavidin using an enhanced chemiluminescence kit.
| Results |
|---|
|
|
|---|
The presence of 5 integrin LM receptors
(
1ß1,
2ß1,
3ß1,
6ß1, and
6ß4) was investigated by flow
cytofluorometry with specific MoAbs in 4 papillary and 3 follicular
carcinomas obtained at surgery (Fig. 1A
).
The analysis showed that all subunits were present, although each
displayed a different intensity of fluorescence. ß1,
3, and ß4 were the most fluorescent ones,
thus confirming the complex integrin expression profile of tumor cells
(7). Thyrocytes from follicular carcinoma displayed the same integrin
profile as papillary tumors, with bright fluorescence of all 5
receptors. The expression of integrin LM receptors was also
investigated in 15 normal thyroid specimens obtained from the
controlateral lobes of carcinomas (Fig. 1B
). As previously shown (5, 6), the majority of the cells showed only the expression of the
ß1- and
3-subunits, whereas only few cells
(<3%) also expressed the subunits
1 and
6 (fluorescence of these few cells is not reported in
the diagram). As expected by the absence of the
6-chain,
ß4 was undetectable in these cells.
|
3 fluorescence was higher in normal follicular cells
from subconfluent primary cultures than in the uncultured cells shown
in Fig. 1
2-subunit was also present. In both
normal thyroid cells and TAD-2 cells, ß1 and
3 were the more fluorescent subunits,
6
was only slightly detectable in TAD-2 and was absent in normal primary
cultures, whereas ß4 was totally undetectable in both
cells. In the three carcinoma cell lines, ß1 and
3 were again the more fluorescent chains, whereas
6 and ß4 ranged from weak staining in NPA
and WRO to strong staining in ARO cells.
|
As the binding of integrin receptors to their ligands is cell type
specific and is regulated by several factors at the posttranscriptional
level, effective cell adhesion must be directly verified by adhesion
assays. To investigate whether thyroid cells attached to immobilized
LM-1, cell attachment assays were performed in 96-well flat-bottom
microtiter plates coated with different concentrations of LM-1 (Fig. 3A
). TAD-2, NPA, and ARO cells harvested
by EDTA/PBS incubation from subconfluent cultures showed poor
attachment to LM-1 even though several integrin LM receptors were
expressed. On the contrary, the adhesion of WRO cells was remarkably
higher and LM-1 concentration dependent.
|
Expression of the 67LR in normal and tumor thyroid cell lines
As WRO cells showed a trypsin-sensitive adhesion to LM, and
trypsin does not affect integrins (1), the potential involvement of a
nonintegrin LM receptor was hypothesized. The expression of the
metastasis-associated 67LR, which was extremely sensitive to trypsin
treatment of cell surface (18), was investigated. ARO, NPA, TAD-2, and
WRO cells were surface labeled with biotin, and the whole cell lysates
were then immunoprecipitated with anti-67LR antibodies (Fig. 4
). A single band with a molecular mass
of 67 kDa was immunoprecipitated only from the WRO cell lysate (lane
8), whereas ARO, NPA, and TAD-2 cell lysates were negative (lanes 2, 4,
and 6). Given its cytoplasmic localization, the 37LRP was not
immunoprecipitated from surface-labeled cells (lanes 2 and 4).
|
The role of 67LR in mediating WRO cells adhesion to LM-1 was
investigated by an attachment-inhibition assay performed in the
presence of soluble r37LRP, a recombinant polypeptide containing
LM-binding domains of the mature 67LR (Fig. 5
). As expected (18), the adhesion of WRO
cells to LM-1 was dramatically reduced by trypsin treatment of the
cells, showing a 64% inhibition of cell attachment. In the presence of
soluble r37LRP, the adhesion of WRO cells to LM was strongly inhibited,
showing an inhibition of cell adhesion corresponding to 76.3%
inhibition of the trypsin-sensitive cell adhesion.
|
The expression of 67LR was investigated by Western blot in cell
lysates from normal and carcinoma thyroid cells in primary culture
(Fig. 6
). Expression of the 67LR was
found only in follicular carcinoma cells (lane 5). As expected, all
cultured cells expressed the 37-kDa 37LRP. Indeed, in primary cultures
from nodular goiter (lane 1), toxic adenoma (lane 2), benign adenoma
with Hurthle cells (lane 3 and 4), and papillary carcinoma (lane 6),
only cytoplasmic 37LRP was present. As a positive control, the lysate
of WRO cells is shown in lane 7. Only bands corresponding to 37LRP and
67LR were competed out by the addition of soluble r37LRP to the
immunoblot reaction (not shown).
|
|
To investigate whether thyroid cells in primary culture attached
to immobilized LM-1, a cell attachment assay was performed with cells
derived from nodular goiter, toxic adenoma, papillary carcinoma, and
follicular carcinoma primary cultures (Fig. 8
). Cells harvested by EDTA/PBS
incubation from subconfluent cultures were plated in 96-well
flat-bottom microtiter plates coated with 0.5 µg LM-1. Among all
thyroid cells tested, only follicular carcinoma-derived cells were able
to attach to LM, thus confirming the observations in WRO cells.
|
| Discussion |
|---|
|
|
|---|
1ß1,
2ß1,
3ß1,
6ß1, and
6ß4)
appear to bind purified EHS/LM-1 and to mediate cell adhesion to
different LM isoforms. The
6ß4 heterodimer
has been shown to mediate stable adhesion to LM-1, LM-2, and LM-5 (25).
Certain integrins display cell type binding specificity, and in some
instances, they appear to assume multiple functional states in the same
cell. For example,
2ß1 is a receptor for
both CoG and LM in melanoma and endothelial cells, but it binds only
CoG in fibroblasts and platelets (2). More recently, it has been
reported that the breast carcinoma cell line MDA-MB-435 expresses three
potential integrin LM receptors (
2ß1,
3ß1, and
6ß1), but uses only
6ß1 to mediate adhesion and migration on
LM (26). In contrast, although
3ß1 binds
purified human and mouse LM-1 (A chain) and competes with
6ß1 for available sites (27),
3ß1-transfected K562 melanoma cells fail
to bind to LM-1/EHS, but adhere to LM-2 (28). It is possible that
either the activation state of the receptor can confer different ligand
specificities or, alternatively, that different forms of the receptor
exist (29). Changes in binding affinity can be rapidly triggered by
different factors, and integrin receptors can be converted into a fully
active form. This phenomenon is independent from de novo
protein synthesis and appears to be induced by conformational changes
in preexisting receptors (30). In the light of these data, the
expression and function of integrin receptors need to be carefully
evaluated before extrapolation to in vivo situations.
Although normal thyrocytes express only
3ß1in vivo, thyroid tumor cells displayed a more complex
integrin profile. In tumor-derived cells in primary culture as well as
in carcinoma cell lines,
1ß1,
2ß1,
3ß1, and
6ß1 integrin receptors were expressed.
Also,
6ß4 was present at high levels in
the specimen of papillary cancer and in ARO cells, whereas it was
weakly expressed in NPA and WRO cells. In many cell types integrin
expression can be regulated by a number of factors. Cytokines such as
interleukin-1ß, tumor necrosis factor-
, and transforming growth
factor-ß; cadherins in dermal keratinocytes; or cell to cell contact
in normal thyrocytes in primary culture are all factors that can
regulate integrin expression (31, 32, 33).
Despite the presence of several, highly expressed potential integrin LM
receptors, the adhesion of TAD, NPA, and ARO cells to LM-1 was poor.
Also, normal thyrocytes in primary culture, although expressing
2ß1 and
3ß1,
attach very weakly to LM-1 (33). Among all thyroid cell lines tested,
only the follicular carcinoma WRO cells were capable of a strong
binding to LM-1. The trypsin sensitivity of WRO cell adhesion to LM
suggested the involvement of nonintegrin LM receptors. Given the
ability of follicular carcinoma cells to diffuse via blood-borne
metastasis formation, our study next focused on the expression and
function of the 67-kDa high affinity LM receptor. This nonintegrin LM
receptor mediates a crucial step of the metastatic cascade: the
attachment of cancer cells to LM-coated endothelial cells (15) and to
the exposed LM of the subendothelial matrix (34). In a large variety of
human cancers, particularly in breast and colon cancers, increased
expression of the 67LR is a molecular marker of metastatic potential
and aggressiveness (35). Increased expression of the 67LR is
particularly critical for the success of hematogenous metastasis, and
the 67LR is widely recognized as a metastasis-associated receptor
(36).
Within thyroid cell lines, only the follicular carcinoma WRO expressed the 67LR. Among cells derived from different thyroid diseases, only those from follicular carcinoma displayed 67LR overexpression. As shown by competition experiments with the r37LRP, containing both LM-binding domains of the 67LR, this receptor was functionally active and was responsible for the strong adhesion to LM displayed by WRO cells and follicular carcinoma-derived cells. The expression of the 37LRP cytosolic precursor in cells derived from nonneoplastic thyroid diseases and papillary carcinomas is not inconsistent with their lack of adhesion to LM. The 37LRP is not exposed on the cell surface; thus, it does not act as a LM receptor. It is localized only in the cytoplasm, where it is posttranslationally modified into the mature surface 67LR, probably through dimerization of 37LRP molecules acylated by the fatty acids palmitate, oleate, and stearate (12, 13). These fatty acids are covalently associated via an ester or thioester linkage and are likely to be responsible for targeting the protein to the cell surface, where it can participate in the adhesion of cells to the extracellular matrix. In addition, the cytoplasmic 37LRP polypeptide may exert other functions; in particular, it has been shown to interact with ribosomes and play a role in polysome formation (37, 38).
A previous study performed only by immunohistochemistry on paraffin-embedded tissue sections reported the restricted expression of 67LR in thyroid carcinomas (39). In our study, Western blot analysis performed on extracts from frozen thyroid tissue specimens confirmed 67LR expression in follicular carcinomas, whereas it was absent in normal thyroid tissues and follicular adenomas. Thus, 67LR overexpression in follicular carcinomas could be a potentially useful tool to gain diagnostic information.
The restricted overexpression in follicular carcinoma cells of a functional 67LR together with nonfunctional integrin LM receptors suggests that the activities of various extracellular matrix receptors can be differentially modulated by cancer cells according to their invasive and metastatic potentials. The high affinity to LM of the 67LR could mediate stable contacts with basal membranes and be responsible for the tendency of follicular carcinoma cells to invade blood vessels and to disseminate via blood-borne metastasis.
| Footnotes |
|---|
Received February 12, 1998.
Revised June 25, 1998.
Revised January 26, 1999.
Accepted February 18, 1999.
| References |
|---|
|
|
|---|
6ß4 is a laminin receptor. J Cell Biol. 117:671678.
6ß1 in metastatic breast carcinoma cells assessed
by expression of a dominant-negative receptor. Cancer Res. 56:959963.
6ß1 integrins from platelets and adherent cells
by affinity chromatography on mouse laminin fragment E8 and human
laminin pepsin fragment. Exp Cell Res. 197:234244.[CrossRef][Medline]
3ß1 in epithelial basement
membranes. Cell. 65:599610.[CrossRef][Medline]
7 subunit are differentially expressed during development. J Biol Chem. 268:2677326783.This article has been cited by other articles:
![]() |
C. Selleri, P. Ragno, P. Ricci, V. Visconte, N. Scarpato, M. V. Carriero, B. Rotoli, G. Rossi, and N. Montuori The metastasis-associated 67-kDa laminin receptor is involved in G-CSF-induced hematopoietic stem cell mobilization Blood, October 1, 2006; 108(7): 2476 - 2484. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |