| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Department of Endocrinology, Max Planck Institute of Psychiatry (M.P.P., U.R., G.K.S.), 80804 Munich, Germany; Fundacion Campomar, CONICET, FCEN, Universidad de Buenos Aires (M.F.L., O.P.), Patricias Argentinas 435, 1405 Buenos Aires, Argentina; Instituto de Investigaciones Médicas, Facultad de Medicina, Universidad de Buenos Aires (V.G.), 1427 Buenos Aires, Argentina; Hospital Santa Lucía (A.C., G.C.), 1232 Buenos Aires, Argentina; Hospital Ramos Mejía (H.M.), 1221 Buenos Aires, Argentina; the Department of Neurosurgery, University of Munich (A.M.), 81377 Munich, Germany; and Laboratory Fisiología y Biología Molecular, Universidad de Buenos Aires. Ciudad Universitaria (E.A.), 1428 Buenos Aires, Argentina
Address all correspondence and requests for reprints to: Prof. Dr. G. K. Stalla, Department of Endocrinology, Max Planck Institute of Psychiatry, Kraepelinstrasse 10, 80804 Munich, Germany. E-mail: stalla{at}mpipsykl.mpg.de or Prof. Dr. E. Arzt, Department de
 |
Abstract |
|---|
 Top Abstract IntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Matrix metalloproteinases (MMPs) are required for the degradation of
the extracellular matrix barrier. They are necessary for the expression
of an invasive and metastatic phenotype in late stage tumor progression
(5, 6, 7). Most breast, colon, and gastric adenocarcinomas are
immunoreactive for MMP-2, whereas benign proliferative disorders of
these tissues are negative (8, 9, 10). MMP-2 is expressed at an early
stage of tumor development and plays a key role in the acquisition of
the metastatic phenotype (8, 11, 12). Pro-MMP-2, secreted as an
inactive enzyme, can be activated by a membrane-type matrix
metalloproteinase (MT1-MMP) (13). The activity of MMPs is also
regulated by tissue inhibitors of MMPs (TIMPs). The maintenance of the
MMP-TIMP balance is important to limit degradation of the matrix
proteins (14). Beside this role for MMPs in metastasis, other functions
have emerged for these enzymes (15, 16). These studies have
demonstrated their participation in vascularization and apoptosis (17, 18). MMPs have also been found to be necessary for the early growth of
metastatic foci (16, 19). However, the biochemical mechanisms involved
in this proliferative action of MMPs are not known. Proposed mechanisms
for this action involve the processing of interleukin-1 and tumor
necrosis factor-
as well as the release of fibroblast growth factor
and epidermal growth factor anchored to extracellular matrix components
(15, 20, 21).
In the present work we describe in pituitary adenomas the expression of representative members of the metalloproteinase family, MMP-2, MMP-9, and MT1-MMP, and their endogenous inhibitors TIMP-1 and TIMP-2. To examine the role of MMPs in pituitary cell function we used the pharmacological inhibitor of MMPs, batimastat, a synthetic, low molecular weight hydroxamate.
|
Subjects and Methods |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Eighty-five patients, 47 women and 38 men, aged 20–82 yr, with
pituitary adenomas and clinical symptoms of Cushing’s disease,
acromegaly, prolactinoma, or nonfunctioning tumors were diagnosed by
plasma pituitary hormone testing and magnetic resonance imaging as
previously described (2, 22, 23). The clinical data of the patients and
histological characterization of the tumors after transsphenoidal
surgery are provided in Table 1
. The
purity of the samples was assessed by immunohistochemistry for the
different pituitary hormones.
|
Unless stated, all reagents were obtained from Sigma (St. Louis, MO), Roche Molecular Biochemicals (Mannheim, Germany), or Pharmacia Biotech (Uppsala, Sweden).
Pituitary adenoma explant incubation
Human or rat normal pituitary and pituitary adenoma explants were prepared as previously described (2). Briefly, normal pituitary or adenoma tissue was rinsed in phosphate-buffered saline (PBS) three times and cut into pieces of approximately 1-mm diameter using scalpels. Pieces weighing between 4–5 mg (fresh weight) were distributed in six-well culture plates (four pieces in each well) and washed again with PBS for 15 min with gentle rocking. PBS was replaced by stimulation medium (DMEM, pH 7.3, containing 2.2 g/L NaHCO3, 10 mmol/L HEPES, 2 mmol/L glutamine, and 1 g/L BSA) and incubated for 30 min with gentle rocking under a 5% CO2 atmosphere at 37 C. After incubation, the medium was replaced by fresh stimulation medium containing the indicated additives. At the end of the treatment, medium was collected for zymographic analysis and total protein measurements. Total protein content was measured with the Bradford assay according to standard procedures. Then the conditioned media were diluted with fresh medium to obtain 0.5 mg total proteins/mL. The protein content of the conditioned media showed a good correlation with the fresh weight of the explants as previously described (2). Cell viability was checked by trypan blue exclusion and acridine orange/ethidium bromide staining and was routinely more than 98% at the end of the treatment. The explants were free of mononuclear cell contamination, as there were less than 1% cells reactive by immunofluorescence with CD4, CD8, and CD14 monoclonal antibodies (Dianova, Hamburg, Germany), which label T helper, T cytotoxic, and monocyte cells, respectively.
Cell lines
HFL-1 human fetal fibroblast cells (24), MDA-MB-231 human breast cancer cells (24), IIB-MEL-LES human melanoma cells (24), AtT-20 mouse corticotropic tumor cells (22, 25), and GH3 rat lactosomatotropic tumor cells (25), were used in this study. The cell lines were cultured in DMEM, pH 7.3, containing 10% FCS, 2.2 g/L NaHCO3, 2 mmol/L glutamine, 2.5 mg/L amphotericin B, and 1 x 105 U/L penicillin/streptomycin under a 5% CO2 atmosphere at 37 C.
Cell culture on Matrigel: use of Batimastat
As a reconstituted extracellular matrix we used Matrigel (Becton Dickinson and Co., Bedford, MA), a basement membrane extract of the Engelbreth-Holm-Swarm tumor. This extracellular matrix is mainly composed of laminin, collagen IV, entactin, and proteoglycans. Growth factor-reduced Matrigel (Becton Dickinson and Co.) is prepared by ammonium sulfate treatment as described by the manufacturer. Solutions of 5 mg/mL Matrigel with or without 10% heat-inactivated FCS (Life Technologies, Inc., Eggenstein, Germany), were prepared according to the manufacturer’s instructions. Culture plates were covered with the Matrigel solutions and incubated for 1 h at 37 C to harden the gel. The cells were then plated on top of the gel in medium with or without heat-inactivated FCS as indicated. Under these conditions the cells attach to the surface of the Matrigel as a monolayer and are covered by culture medium. The cultures were incubated for 4 days in the presence of batimastat (British Biotech, Oxford, UK), which is used as a pharmacological inhibitor of MMPs (26). Batimastat (10 ng/mL to 1 mg/mL) was used in growth factor-containing or-depleted Matrigel. Twenty-four hours after the beginning of the treatment samples were collected for hormone measurements. At the end of the experiments after 4 days of treatment, proliferation was measured with the wst-1 reagent (Roche Molecular Biochemicals) according to the manufacturer’s instructions. This compound is cleaved by the mitochondrial respiratory chain, and the product dye directly correlates to the number of viable cells in culture. The reaction product was measured in an enzyme-linked immunosorbent assay (ELISA) plate reader at 450 nm.
Substrate-independent colony formation in soft agar
Growth in soft agar was performed as we previously described (27). Cells were suspended in 0.3% melted agar with culture medium and plated in dishes previously coated with 0.5% agar. The cultures were then incubated at 37 C for 20 days. At the end of the incubation, the colonies were counted under the microscope.
Pituitary hormone determinations
All anterior pituitary hormones were detected by RIA, as previously described (2, 22, 23, 24).
Zymographic analysis of matrix metalloproteinases
Gelatinolytic activity was analyzed as previously described (24, 28), using 10% polyacrylamide gels containing 0.2% gels. Conditioned media, obtained as described above, was mixed with loading buffer containing 2.5% SDS and incubated for 30 min at room temperature before loading into the gel. After electrophoresis under nondenaturing conditions at 4 C, gels were washed as described. After staining with Coomassie blue R250, gelatinase activity was observed as clear zones of proteolysis against a blue background. Incubation of the gels in the presence of 20 mmol/L ethylenediamine tetraacetate (EDTA) was performed to demonstrate the Ca2+ and Zn2+ dependence of the proteinase activity observed.
TIMP-1 measurement
TIMP-1 was measured by ELISA as previously described (28). Biotrack ELISA systems specific for human TIMP-1 were purchased from Amersham Pharmacia Biotech (Aylesbury, UK). Total human TIMP-1 was measured in conditioned media from normal pituitary or pituitary adenoma explants, standardized according to the total protein content. The ELISA procedure was performed according to the manufacturer’s instructions.
Northern blot
Ribonucleic acid (RNA) was extracted from tumor samples for
Northern blot analysis as previously described (2, 22). Briefly, total
RNA, isolated by the guanidine isothiocyanate phenol-chloroform
extraction method, was denatured with glyoxal, electrophoresed on a
1.2% agarose gel, and transferred to a nylon membrane. Filters were
baked for 2 h at 80 C and stained with methylene blue to check for
RNA integrity. A 1.1-kb SalI fragment from MT1-MMP
complementary DNA (cDNA) (13), an 0.8-kb EcoRI fragment from
human TIMP 1 cDNA (29), and a 1-kb PstI fragment from actin
cDNA (2) were labeled with a random priming kit using
[-32P]deoxy-CTP (SA, 2–4 x
108 cpm/µg). Filters were prehybridized for
1 h and hybridized overnight at 42 C with MT1-MMP probe and at 55
C with actin probe in hybridization buffer (50% formamide, 5 x
SSPE, 5 x Denhardt’s solution, 0.1% SDS, and 100 µg/mL
denatured salmon sperm DNA). The filters were reprobed after eluting
the first probe with 5 mmol/L Tris-HCl (pH 8.0), 2 mmol/L EDTA, and
0.1 x Denhardt’s solution at 65 C for 2 h. After the
previous signal was removed, as confirmed by reexposure of the filter,
the blots were prehybridized and hybridized following methods described
above. The control with the fragment of actin cDNA as probe was
performed in each blot.
Statistics
Statistics were performed using one-way ANOVA in combination with Scheffe’s test. Results are expressed as the mean ± SE.
|
Results |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
We examined MMP activity in conditioned media from 3 normal human
and 3 normal rat pituitaries and from 60 human pituitary adenomas of
different types and clinical features from the 85 adenomas used in this
study (Table 1
). The metalloproteinase nature of the gelatinase
activity was confirmed after inhibition with EDTA (not shown). The
66-kDa band comigrated with the pro-MMP-2 secreted by HFL-1 cells, and
the 92-kDa band comigrated with MMP-9 secreted by MDA cells used as
positive controls (Fig. 1). Pro-MMP-9 was
detected in normal pituitary in low levels and in 38 of 60 conditioned
media (Fig. 1). In addition, the normal pituitary and all 60 tumors
showed a 66-kDa band corresponding to pro-MMP-2, and 27% of them
showed a second band corresponding to MMP-2 (Fig. 1). Active MMP-2 was
only detected in tumors, not in the normal pituitary. This second band
was confirmed as the active form of MMP-2 because it comigrated with a
band obtained from HFL-1-conditioned medium incubated with MDA-MB231
cells treated with Con A, which was shown to activate MMP-2. MMP
activity in pituitary adenoma-conditioned media was as high as that in
conditioned media from melanoma or HFL-1 cell lines. Immunoblot
analysis further confirmed the identity of the gelatinolytic bands as
MMP-2 and MMP-9 (not shown). The activity of MMPs depends on the
proteolytic activation of the proenzymes. As pro-MMP-2 can be locally
activated by MT1-MMP, we analyzed MT1-MMP messenger RNA (mRNA)
expression. We found that 12 of 21 adenomas expressed low levels of
MT1-MMP mRNA (Fig. 2). MT1-MMP was
undetectable in normal pituitary tissue (not shown). As the activity of
MMPs in the extracellular milieu is regulated by TIMPs, we studied
TIMP-1 and TIMP-2 expression. Seventeen of 25 tumors expressed low
levels of TIMP-1 mRNA (Fig. 2), whereas TIMP-2 was not detectable by
Western blot in 20 pituitary adenoma-conditioned media (not shown). We
also measured the levels of TIMP-1 protein in conditioned media from
normal human pituitary and pituitary adenomas by ELISA. For comparison
we selected aggressive tumor types known to express high levels of MMPs
similar to those produced by pituitary adenomas (28). We found that
most of the pituitary adenomas express relatively low or undetectable
levels of TIMP-1 compared to aggressive cancer cells (Fig. 3). All of these results together show that
pituitary adenomas express high levels of MMP-2 and MMP-9 and
relatively low levels of TIMPs, which indicates that the balance would
favor extracellular matrix digestion in these tumors. However, we did
not find any significant correlation between the MMP activity or the
TIMP-1 levels and the invasiveness or other clinical features of
pituitary adenomas. The presence of MMP-2 and MMP-9 was also observed
in the corticotropic tumor cell line AtT-20 and the lactosomatotropic
tumor cell line GH3 (not shown).
|
|
|
It was previously demonstrated that MMPs can release growth
factors anchored to the extracellular matrix (21). We speculated,
therefore, that this could be a biochemical mechanism for the role of
MMPs in pituitary cells. To test this hypothesis we have measured
pituitary tumor cell line proliferation and hormone secretion in the
presence of the broad range metalloproteinase inhibitor batimastat in
Matrigel-containing or -depleted growth factors. For this purpose
we cultured the pituitary tumor cell lines AtT-20 and
GH3 on thick layers of Matrigel containing 10%
heat-inactivated FCS and culture medium without any growth factor
sources. Under these conditions, batimastat inhibited the secretion of
ACTH in AtT-20 cells and the secretion of GH and PRL in
GH3 cells after 24-h treatment (Fig. 4). No effect of batimastat was observed when
growth factor-reduced Matrigel was used in the same experiments (not
shown). After 4 days batimastat significantly and dose dependently
reduced the proliferation of both cell lines on Matrigel containing
10% serum (Fig. 5A). When FCS was not added
to the Matrigel, the same tendency was observed, but the differences
were not significant (Fig. 5B), and when growth factor-reduced Matrigel
was used, no effect of batimastat on proliferation was observed (Fig. 5C). When FCS was added to the culture media instead of adding it to
the Matrigel, no effect of batimastat was observed on cell
proliferation (4 days) or hormone production (24 h; not shown).
Together, these experiments demonstrate that factors included in the
extracellular matrix need to be released by metalloproteinases for the
cells to achieve a maximum growth rate and hormone secretion.
|
|
).
In line with this, batimastat does not affect substrate-independent
colony formation in soft agar in these cell lines (Table 3). These results clearly confirm that
batimastat does not affect the intrinsic or autocrine mechanisms of
cell proliferation and survival in these cell lines.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
Our results provide a new mechanism for the control of hormone secretion and proliferation in pituitary cells. This mechanism could be mediated by endogenous growth factors anchored to the extracellular matrix and intrinsically secreted MMPs, highlighting the importance of the paracrine interactions in the pituitary. Although the active forms of the MMPs have not been found in the normal pituitary, local activation in restricted areas or under specific physiological conditions might allow MMPs to control growth and hormone secretion in normal cells. The expression of pro-MMPs in normal pituitaries further confirms the putative role of MMPs in other processes different from invasion, such as regulation of growth and hormone secretion. Our interpretation is in line with the current models of pituitary tumor development in which adenoma growth and progression are highly dependent on the availability of growth factors (1, 30, 31). Our results show that MMP activity acts to release growth factors from the extracellular matrix that control tumor growth and hormone secretion. This is further supported by our previous observation that after extracellular matrix digestion with collagenase and hyaluronidase, c-Fos expression is abolished in pituitary tumor cells (2). Although other mechanisms could be involved, and further studies are necessary to confirm this hypothesis, we speculate that the release of growth factors anchored to the extracellular matrix would be necessary to maintain c-Fos expression and, hence, cell proliferation.
In pituitary adenomas, which are seldom invasive or metastatic, the main function of the high level of extracellular matrix-degrading activity could be to contribute to the regulation of proliferation and hormone secretion. This interpretation is in agreement with a previously proposed role for MMPs in early tumor development and growth (15, 18). It was shown that matrilysin (MMP-7) is necessary for the growth of early adenomatous foci in colon cancer. In matrilysin-null mutants the early stages of colon tumor growth seem to require MMP-2. Therefore, it was proposed that MMP-2 could be an alternative pathway for the action of matrilysin in tumor growth (32). Our results are in line with this view and further support a role in early tumor development for MMP-2, which, unlike matrilysin, is more often expressed in aggressive cancer. Moreover, we demonstrate that the general inhibition of MMP activity with batimastat can inhibit cell proliferation in adenomatous cells. Our results do not rule out the participation in pituitary cell proliferation of other metalloproteinases, which could be inhibited by batimastat. However, the high expression of MMP-2 and -9 in pituitary adenomas makes these two enzymes good candidates to be involved in proliferation and hormone secretion. The participation of MMPs in the proliferation of pituitary cells could provide the basis for understanding the mechanisms of early development of other tumor types and suggests that therapies based on MMP inhibition could be applied not only to metastatic but also to benign tumors.
In conclusion, these results define a role for MMPs in the digestion of the extracellular matrix that could limit the availability of growth factors necessary for pituitary cell proliferation during early stages of tumor formation. It is also the first evidence of the participation of MMPs in the modulation of pituitary hormone secretion. The involvement of MMPs in the proliferation of pituitary cells defines a new therapeutic target to inhibit pituitary adenoma growth and hormone secretion.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 These authors equally contributed.
Received June 16, 1999.
Revised September 3, 1999.
Accepted September 13, 1999.
|
References |
|---|
|
TopAbstractIntroductionSubjects and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  H. Jin, Y. Pan, L. He, H. Zhai, X. Li, L. Zhao, L. Sun, J. Liu, L. Hong, J. Song, et al. p75 Neurotrophin Receptor Inhibits Invasion and Metastasis of Gastric Cancer Mol. Cancer Res., May 1, 2007; 5(5): 423 - 433. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Giacomini, M. Paez-Pereda, M. Theodoropoulou, M. Labeur, D. Refojo, J. Gerez, A. Chervin, S. Berner, M. Losa, M. Buchfelder, et al. Bone Morphogenetic Protein-4 Inhibits Corticotroph Tumor Cells: Involvement in the Retinoic Acid Inhibitory Action Endocrinology, January 1, 2006; 147(1): 247 - 256. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Vila, M. Theodoropoulou, J. Stalla, J. C. Tonn, M. Losa, U. Renner, G. K. Stalla, and M. Paez-Pereda Expression and Function of Sonic Hedgehog Pathway Components in Pituitary Adenomas: Evidence for a Direct Role in Hormone Secretion and Cell Proliferation J. Clin. Endocrinol. Metab., December 1, 2005; 90(12): 6687 - 6694. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S.-R. Lee, K. Tsuji, S.-R. Lee, and E. H. Lo Role of Matrix Metalloproteinases in Delayed Neuronal Damage after Transient Global Cerebral Ischemia J. Neurosci., January 21, 2004; 24(3): 671 - 678. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Roelle, R. Grosse, A. Aigner, H. W. Krell, F. Czubayko, and T. Gudermann Matrix Metalloproteinases 2 and 9 Mediate Epidermal Growth Factor Receptor Transactivation by Gonadotropin-releasing Hormone J. Biol. Chem., November 21, 2003; 278(47): 47307 - 47318. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Paez-Pereda, D. Giacomini, D. Refojo, A. C. Nagashima, U. Hopfner, Y. Grubler, A. Chervin, V. Goldberg, R. Goya, S. T. Hentges, et al. Involvement of bone morphogenetic protein 4 (BMP-4) in pituitary prolactinoma pathogenesis through a Smad/estrogen receptor crosstalk PNAS, February 4, 2003; 100(3): 1034 - 1039. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Pascual-Le Tallec, E. Dulmet, X. Bertagna, and Y. de Keyzer Identification of Genes Associated with the Corticotroph Phenotype in Bronchial Carcinoid Tumors J. Clin. Endocrinol. Metab., November 1, 2002; 87(11): 5015 - 5022. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Sumii and E. H. Lo Involvement of Matrix Metalloproteinase in Thrombolysis-Associated Hemorrhagic Transformation After Embolic Focal Ischemia in Rats Stroke, March 1, 2002; 33(3): 831 - 836. [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 |
) from 45 patients were incubated for
24 h, then the supernatants were collected for measuring TIMP-1 by
ELISA, as described in Materials and Methods. The
samples are standardized to 0.5 µg protein/mL conditioned medium.
Samples from other tumor types (n) were tested for comparison. a and c,
Meningioma; b and d, prostate carcinoma; e, thyroid carcinoma.
