| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Neuroendocrine Unit (D.C.D., X.Z., Y.Z., S.R.J., A.K.) and the Department of Pathology (G.R.D., M.K.S., E.T.H.-W.), Massachusetts General Hospital, and the Department of Pathology, Brigham and Women’s Hospital and Harvard Medical School (J.A.F.), Boston, Massachusetts 02115.
Address all correspondence and requests for reprints to: Anne Klibanski, M.D., Neuroendocrine Unit, Bulfinch 457, Massachusetts General Hospital, Boston, Massachusetts 02114. E-mail: aklibanski{at}partners.org
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The anterior pituitary is composed of two different cell types, secretory cells and folliculostellate cells. It has been suggested that the folliculostellate cells, derived from neuroectoderm, modulate secretory cell functions by intercellular communication in a paracrine manner (8, 9). In the normal pituitary, folliculostellate cells express and secrete follistatin, which can bind and neutralize activin and, therefore, regulate FSH biosynthesis and secretion indirectly (10, 11). Recent data suggest that most circulating follistatin is bound to activin, and only the locally produced follistatin has functional significance in the regulation of activin-induced hormone secretion and proliferative responses (12). It has been demonstrated that folliculostellate cells also produce lipocortin-1 (13), a key inhibitory mediator of glucocorticoids on corticotropin secretion, at both the hypothalamic and pituitary levels (14, 15). The regulatory interactions between folliculostellate and lactotroph cells (16, 17, 18, 19) and the role of folliculostellate cells in immune system modulation (20, 21, 22, 23, 24) have also been reported. In the pituitary, it has been shown that a subset of pituitary adenomas has significant numbers of folliculostellate cells (25, 26, 27, 28, 29, 30), and furthermore, pituitary tumors consisting only of folliculostellate cells have been described (29).
We describe here a spontaneously transformed human cell line, PDFS, of folliculostellate origin, derived from a pituitary gonadotroph adenoma. The transformation and immortalization of this cell line are possibly due to the mutation of p53. It expresses follistatin and activin A, important peptides in the autocrine/paracrine regulatory mechanisms of normal and neoplastic pituitary, and retains an intact activin-mediated signal transduction pathway, as determined by reporter assays. Therefore, this cell line may provide a useful model for the study of regulation of cell growth and hormone production by factors endogenously expressed in the normal and neoplastic human pituitary as well as human neoplastic cell transformation.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Tumor fragments from a 71-yr-old man with a clinically
nonfunctioning pituitary macroadenoma were obtained in saline solution
after transsphenoidal surgery. A portion of the resected tumor was
embedded in paraffin for histological examination, which showed cells
immunoreactive for 
- and ß-subunits of LH and FSH, consistent with
a pituitary adenoma of gonadotroph origin. Another fragment was
enzymatically dispersed as previously described (31) and cells were
maintained in DMEM supplemented with 10% FCS, 0.1%
insulin/transferrin/selenium, 1% nonessential amino acids, and
antibiotics. After several days, a distinct focus of nonfibroblastic
cells had formed in one of the culture dishes. This cell clone was
isolated using a cloning ring, trypsinized, and replated in medium as
described above. The cells have been maintained in culture for over 1
yr and in excess of 70 passages.
Chromosomal analysis
Metaphase cells were obtained by exposing cultured cells to Colcemid (Life Technologies, Inc., Gaithersburg, MD) for 1 h. Metaphase harvesting conditions, slide making, and trypsin-Giemsa chromosome banding were previously described (32). Twenty metaphase cells were analyzed, and clonal aberrations were specified according to the 1995 International System for Human Cytogenetic Nomenclature (33).
Electron microscopy
Cultured cells were fixed with Karnovsky II fixative (2.5% glutaraldehyde, 2% paraformaldehyde, and 0.025% calcium chloride in 0.1 mol/L sodium cacodylate buffer, pH 7.4), postfixed with osmium tetroxide, en bloc stained with uranyl acetate, dehydrated through a graded ethanol series, infiltrated with 100% ethanol and epoxy resin mixture, and embedded in pure epoxy. After polymerization, the cell layer was removed from the petri dish and reembedded, so that 1-µm thick cross sections of the cells could be cut and stained with toluidine blue. Representative areas were chosen for thin sectioning by light microscopy. Thin sections were cut, stained with lead citrate, and examined with a Philips 301A electron microscope (Philips Electronic Instruments, Rahway, NJ).
Cell growth and transformation assay
PDFS cells were plated into 6-well plates at a density of 5000 cells/well in medium containing 0.2%, 5%, and 10% FCS. Cells in triplicate wells were harvested every day, and counted in a Coulter counter (Coulter Electronics, Hialeah, FL), and cell doubling time was calculated as previously described (34).
The transforming ability of PDFS cells was tested in an anchorage-independent growth assay on soft agar as previously described (35, 36). Two milliliters of 0.5% soft agar were added to 6-well plates. Ten thousand cells were mixed with 1 mL 0.3% soft agar and added to each well. Cells were incubated for 2 weeks before colonies of 60 or more cells were counted and photographed. Human osteosarcoma U2OS and fibroblast HS27 cells were used as positive and negative controls, respectively.
RT-PCR
Total ribonucleic acid (RNA) was extracted from 5 x
105 cultured PDFS cells or 100 mg normal
pituitary tissue with Trizol reagent (Life Technologies, Inc.), and RT-PCR was performed as previously described (37). We
used the following sets of primers, described in 5'-3' sequence and
with the expected product size in parentheses: vimentin,
gcaggactcggtggacttctc and ggcagccacactttcatattg (603 bp); and activin
A, ctgaacgcgatcagaaagctt and tcctccacgatcatgttctg (1015 bp). All other
primers were described previously (2, 38). Reactions were preformed in
the presence of [-32P]deoxy-CTP (100
nCi/reaction), and products were resolved by electrophoresis on 6%
nondenaturing TBE/polyacrylamide gel (Protogel, National
Diagnostics, Atlanta, GA) and visualized by autoradiography on
Kodak X-Omat film (Eastman Kodak Co.,
Rochester, NY) for 4 h. The PCR products were sequenced to confirm
identities.
Western blotting
Immunoblotting for p16ink4a, p53, and Rb protein expression in PDFS cells was performed as previously described (39). The cellular proteins were extracted in RIPA buffer [150 mmol/L NaCl, 50 mmol/L Tris (pH 7.4), 1 mmol/L ethylenediamine tetraacetate, 1% Nonidet P-40, and 0.25% sodium deoxycholate] containing 1 mmol/L phenylmethylsulfonylfluoride, and 1 µg/ml aprotinin and leupeptin. The homogenates were centrifuged, and the protein concentration in each sample was determined by Bradford assay (Sigma, St. Louis, MO). Twenty micrograms of total cellular proteins were resolved by 10% SDS-PAGE, and then electroblotted onto polyvinylidene difluoride filters (Millipore Corp., Bedford, MA). Filters were probed with specific primary antibodies, anti-Rb (G3–245, PharMingen, San Diego, CA), anti-p53 (FL-393, Santa Cruz Biotechnology, Inc., Santa Cruz, CA), and anti-p16ink4a (DCS-50.1, NeoMarker, Fremont, CA). The blots were processed by a chemiluminescence-based enhanced chemiluminescence Western blotting detection system (Amersham Pharmacia Biotech, Piscataway, NJ) according to the manufacturer’s instruction.
Immunocytochemistry
Immunostaining of tumor and cultured cells was carried out as previously described (40). Antibodies against the anterior pituitary hormones, vimentin (Santa Cruz Biotechnology, Inc.), the folliculostellate specific marker S-100 (DAKO Corp., High Wycombe, UK), and follistatin (National Hormone and Pituitary Program, Torrance, CA) were used as primary antibodies. All antisera were used at the manufacturer’s suggested concentrations and were visualized using an indirect immunoperoxidase procedure with Vectastatin Elite ABC kit (Vector Laboratories, Inc., Burlingame, CA). Controls included substitution of nonimmune serum for primary antibody, elimination of secondary antibody, or streptavidin-biotin-peroxidase complexes. Immunostaining for p53 wild-type (wt) and mutant proteins was preformed as described above, except that cells were fixed in acetone-methanol (1:1 ratio), and incubated with 0.03% H2O2 in methanol. The primary anti-p53 antibodies used were PAb1801 (which recognizes both wt and mutated p53) (41) and PAb240 (specific for the mutated p53; NeoMarker) (42).
Enzyme-linked immunosorbent assay of activin A in conditioned medium
Cells (106) were plated in 100-mm cell culture dishes with 10 mL culture medium. After 48 h, the culture medium was collected as conditioned medium, and the activin A concentration was measured using an activin A assay kit (Serotec, Oxford, UK), according to the manufacturer’s protocol.
Cell transfection and luciferase assay
Mink lung L17 cells were transiently transfected using
Lipofectamine (Life Technologies, Inc.) as previously
described (39). Cells were plated in six-well plates at a density of
2 x 105 cells/well and transfected in
triplicate with 1 µg/well wt activin receptor IB (Act-RIB) inserted
under the control of the cytomegalovirus promoter in the pCI/Neo
expression vector (Promega Corp., Madison, WI),
pCI-Act-RIB (wt), and 1 µg/well p3TPlux reporter, in which a
luciferase gene is under the control of an activin-responsive
element-containing promoter (provided by Dr. Joan Massagué) (43).
As an internal control, 0.2 µg/well pRSV-lacZ was included
in each experiment. After 4-h incubation at 37 C, the culture medium
was replaced with fresh medium, and 500 µL PDFS-conditioned medium or
50 ng human recombinant activin A (National Hormone and Pituitary
Program) were added to each well. After an additional 24-h incubation,
cells were harvested, the luciferase activity was measured, and results
were normalized to ß-galactosidase activity. PDFS cells were
similarly transiently transfected with pCI-Act-RIB (wt) or pCI-Act-RIB
(
c), which expresses a truncated form of Act-RIB lacking the
intracellular domain (provided by Dr. Wylie Vale) (44), and p3TPlux. To
quantitate the luciferase activity in this model, we transfected in
parallel L17 cells with pCI-Act-RIB (wt) and p3TPlux, and then treated
them with 10 ng/mL activin A. Cells were harvested after 24-h
incubation in fresh medium, and the activities of luciferase and
ß-galactosidase were measured. Each experiment was repeated twice,
and statistical significance was tested by Student’s t test
using SigmaPlot.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The PDFS cell line was spontaneously transformed from the primary
culture of a clinically nonfunctioning pituitary macroadenoma. It
showed an epithelial-like morphology with large nuclei and abundant
cytoplasm, with long cytoplasmic processes (Fig. 1A
). Hematoxylin/eosin staining shows
that PDFS cells have notable pleomorphism, with multiple nucleoli and
numerous mitotic figures (Fig. 1B). Electron microscopy of the cultured
cells revealed them to be tightly opposed in clusters and to have
frequent intercellular junctions, including desmosomes (Fig. 1C). Free
cell surfaces were raised into multiple filopodia. Nuclei were
euchromatic and contained prominent and multiple nucleoli. The
cytoplasm contained innumerable free ribosomes, many mitochondria, and
a few cisternae of rough endoplasmic reticulum. Some cells contained
many primary and secondary lysosomes. There was no evidence of
secretory granules, including dense core type, suggesting that it did
not originate from hormone-secreting cells. Filaments were sparse and
present in only a few cells.
|
PDFS cells showed no expression of messenger RNA (mRNA) encoding Pit-1
or any of the anterior pituitary hormones (data not shown), as
determined by RT-PCR. We therefore investigated whether the PDFS cell
line was of folliculostellate origin using specific folliculostellate
markers. Immunocytochemistry showed that the PDFS cells express
vimentin, a component of the intermediary filaments, in the cytoplasm
(Fig. 1D
). The S-100 protein, a folliculo- stellate-specific marker
in the context of pituitary, is expressed in PDFS cells, as shown by
immunocytochemistry in Fig. 2. The PDFS
cells specifically stained positively for S-100 (Fig. 2A), as confirmed
by substitution of the primary antibody against S-100 with nonimmune
serum (Fig. 2B). In Fig. 2, C and D, respectively, the S-100-positive
breast cancer cell line MDA-MB 231 (45) and the S-100-negative
pituitary secretory-cell derived line CHP2 (7) were used as
controls.
|
The growth of PDFS cells was serum dependent. Decreasing concentrations of FCS in the medium to 0.2% resulted in cell death. There were no significant differences in cellular proliferation between concentrations of 5% and 10% serum. The doubling time of the cells was 23.8 h.
The transforming ability of these cells was tested in an
anchorage-independent growth assay. We found that PDFS cells formed
colonies in soft agar (Fig. 3A), similar
to the human carcinoma cell line U2OS (Fig. 3B), whereas the human
fibroblast cells, used as a negative control, failed to form colonies
(Fig. 3C). From 104 PDFS cells, 942 ± 177
colonies were formed in 2 weeks (mean ± SEM).
|
To understand the potential mechanism underlying immortalization
of this cell line, we investigated the tumor suppressor genes
p16ink4a, Rb, and p53 status in PDFS cells. The
PDFS cells express p16ink4a protein (Fig. 4A, lane 1), as determined by Western
blotting. Rb protein was detected in PDFS cells and displayed the
correct size and a normal phosphorylation pattern compared to that in
control cells (Fig. 4B).
|
, lane 1), similar to that in T47D cells, a human
breast carcinoma cell line known to express mutant p53 (lane 2) (46).
The expression of p53 in PDFS cells was stronger than the expression
detected in normal human fibroblast HS 27 cells (lane 4). We further
studied p53 expression in PDFS cells by immunostaining with PAb 240
antibody, which specifically recognizes mutated p53 (Fig. 5, right panels), and
comparing it with PAb1801 antibody, which recognize both wt and mutant
p53 (Fig. 5, left panels). The PDFS cells stained positively
with both PAb1801 and PAb 240 antibodies (Fig. 5A), whereas the wt
control MCF-7 cells (47) stained positively only with PAb1801 (Fig. 5C). The T47D cells also stained positively with both anti-p53
antibodies (Fig. 5B), whereas the osteosarcoma Saos2 cells, p53 null
(48), did not stain with any antibody (Fig. 5D).
|
We studied the expression of follistatin, activin, and its
receptor in PDFS cells. Using RT-PCR, we found that PDFS cells express
follistatin mRNA (Fig. 6A). The cells
also express mRNA for ßA-subunit of activin/inhibin (Fig. 6B),
suggesting that these cells may produce the ßA-homodimeric activin A.
This result was further confirmed by enzyme-linked immunosorbent assay,
which measured 50 ng/mL activin A protein secreted over 48 h by
106 PDFS cells in 10 mL culture medium. RT-PCR
also detected mRNA for Act-RIB (Fig. 6C), and Act-RIIB (Fig. 6D) in
PDFS cells.
|
, PDFS-conditioned medium caused a 3-fold increase in luciferase activity
from p3TPlux reporter compared to regular medium, an effect comparable
to that of 50 ng/mL recombinant activin. The increase in reporter
activity was activin signaling pathway specific, as the conditioned
medium did not stimulate the reporter in the absence of Act-RIB.
|
). This activin-induced activation in the PDFS
cells is probably due to the presence of endogenous functional activin
receptors and the autocrine/paracrine effects of secreted activin A.
Cotransfection of p3TPlux with full-length Act-RIB in PDFS cells
further enhanced the luciferase activity, whereas cotransfection with
the truncated Act-RIB (c) significantly decreased the reporter
activity (P < 0.001; Fig. 7B). |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The study of the growth rate of the PDFS cell line revealed a doubling time of 24 h, similar to that of most established human carcinoma cell lines. In an anchorage-independent transformation assay, PDFS cells formed large colonies in soft agar, indicating that the cells had undergone malignant transformation. Cell transformation is usually caused by oncogene activation (54) or inactivation of tumor suppressor genes (55, 56, 57). One of the most frequent functionally inactivated tumor suppressor genes in human cancers is Rb, which controls progression from G1 to S phases in normal cycling cells by regulating the transcription of E2F-responsive genes (58). Another frequently mutated tumor suppressor gene is p53, which promotes growth arrest and apoptosis in response to cellular stresses in a transcription-specific manner that may correlate with the functional status of Rb (59). In cells expressing functional Rb, p53 induces G1 arrest by induction of cyclin-dependent kinase (cdk) inhibitors, whereas in cells lacking functional Rb, p53 induces apoptosis. Either of the above will inhibit neoplastic cell development. Hyperphosphorylation of Rb by G1 cdk, mainly cdk4 or cdk6, neutralizes its tumor-suppressing function. The kinase activity of cdk4/6 is inhibited by an ink family of cdk inhibitors, such as p16ink4a (60), resulting in Rb activation. Therefore, the inactivation of Rb, a result of mutation or the lack of functional p16ink4a expression, leads to uncontrolled cell growth and immortalization. Although in many human tumors the inactivation of tumor-suppressor genes Rb, p53, and p16ink4a has been found from the early stages (60, 61), mutations of these genes have not been commonly found in human pituitary adenomas (62). However, it has been demonstrated that these tumor suppressor genes are frequently mutated during establishment of various cell lines. Therefore, we studied the status of these most frequently inactivated tumor suppressor genes in PDFS cells. Although p16ink4a and Rb expressions were normal, the p53 appeared mutated in PDFS cells. This is probably a point mutation, because the protein size of p53 is not altered and the nuclear localization is preserved, as illustrated by Western blotting and cellular immunostaining. However, the mutation caused protein conformation changes in p53, which are recognized by PAb240, a mutant p53-specific antibody. In addition, the interaction with mdm2 protein, which is required for ubiqitin-mediated degradation of p53, may be altered in this cell line, resulting in its accumulation within the cell (63, 64). This may explain the high level of p53 expression in PDFS cells, similar to that found in mutant p53-expressing T47D cells. Only weak p53 expression was detected in HS-27 cells due to the very short half-life of wt p53 in nonstress conditions (65). It has been reported that p53 is prone to inactivation during cell line establishment. For example, all spontaneously immortalized cell lines derived from mouse embryonic fibroblasts contain p53 mutations (66). Therefore, it is possible that a p53 mutation was generated during the culture establishment of the PDFS cell line and helped its immortalization and transformation. However, it is unlikely to be related to formation of the original pituitary adenoma.
Due to the patient’s sex, we were unable to confirm whether the cell line was monoclonal in origin by X-linked inactivation. However, the cytogenetic analysis revealed that all cells were hyperdiploid as a result of multiple clonal trisomies and tetrasomies. Other clonal aberrations included rearrangements of a chromosome X short arm, chromosome 1 long arm, and chromosome 7 short arm. There are few published pituitary adenoma karyotypes. Rey et al. reported a GH-secreting adenoma karyotype with 58 chromosomes and clonal aberrations, primarily numerical (chromosome gains) rather than structural in nature, similar to those found in PDFS cells (67). Recent data suggest that genetic abnormalities in pituitary adenomas often involve several chromosomes, which may inactivate a tumor suppressor gene or activate an oncogene that is important in the initiation or progression of sporadic pituitary adenomas (68).
Activin and other members of the transforming growth factor-ß family of cytokines function as both growth and differentiation factors in a variety of cell types (69). Follistatin, a glycosylated monomeric protein that binds to activin, was originally discovered from ovarian follicular fluid as a suppressor of FSH secretion (70) and later was identified as a widely distributed activin-binding and -neutralizing protein, suggesting its regulatory effects in different activin-mediated actions (71, 72, 73, 74). It has been shown that in the normal human pituitary, follistatin is readily expressed in gonadotropes and folliculostellate cells (75). Locally secreted activin in the normal pituitary is a potent factor in controlling hormone biosynthesis and secretion (76, 77, 78). Activin also influences the proliferation of a number of normal and neoplastic human cell types, and the relative amounts of locally secreted activin, inhibin, and follistatin may control normal cellular proliferation (2, 79). However, the functional interaction between activin and follistatin and the physiological significance of this interaction are poorly understood. Our PDFS cell line provides a novel model to study such interactions in regard to hormone secretion and cell growth. Our data indicate that PDFS cells secrete biologically active activin A, express functional activin receptor types I and II, and retain an intact activin-mediated signal transduction pathway as determined by reporter assays. Therefore, this cell line could also be very useful for studying the molecular events underlying the regulation of cell proliferation by activin.
In conclusion, we report the first transformed human folliculostellate cell line, derived from a gonadotroph adenoma of the pituitary. This line expresses activin and follistatin, important proteins in the autocrine/paracrine regulation of anterior pituitary cells, and may serve as a model system to elucidate further the mechanisms of growth factor regulation in normal and neoplastic human pituitary.
|
Acknowledgments |
|---|
c construct. We are
grateful to Dr. Jack Ham (Department of Medicine, University of Wales
College of Medicine, Cardiff, UK) who provided the CHP2 cell line. |
Footnotes |
|---|
Received August 18, 1999.
Revised November 4, 1999.
Accepted November 12, 1999.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
and
ß in the rat pituitary gland. Endocrinology. 139:3976–3983.-subunit secretion
and messenger ribonucleic acid levels in alpha-secreting pituitary
tumors. J Clin Endocrinol Metab. 66:96–102.[Abstract]
This article has been cited by other articles:
 |
|
 |
|
  L. M Bilezikjian, A. L Blount, C. J Donaldson, and W. W Vale Pituitary actions of ligands of the TGF-{beta} family: activins and inhibins. Reproduction, August 1, 2006; 132(2): 207 - 215. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. D. Cherrington, T. A. Farmerie, C. A. Lents, J. D. Cantlon, M. S. Roberson, and C. M. Clay Activin Responsiveness of the Murine Gonadotropin-Releasing Hormone Receptor Gene Is Mediated by a Composite Enhancer Containing Spatially Distinct Regulatory Elements Mol. Endocrinol., April 1, 2005; 19(4): 898 - 912. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Zhang, Y. Zhou, K. R. Mehta, D. C. Danila, S. Scolavino, S. R. Johnson, and A. Klibanski A Pituitary-Derived MEG3 Isoform Functions as a Growth Suppressor in Tumor Cells J. Clin. Endocrinol. Metab., November 1, 2003; 88(11): 5119 - 5126. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Solito, A. Mulla, J. F. Morris, H. C. Christian, R. J. Flower, and J. C. Buckingham Dexamethasone Induces Rapid Serine-Phosphorylation and Membrane Translocation of Annexin 1 in a Human Folliculostellate Cell Line via a Novel Nongenomic Mechanism Involving the Glucocorticoid Receptor, Protein Kinase C, Phosphatidylinositol 3-Kinase, and Mitogen-Activated Protein Kinase Endocrinology, April 1, 2003; 144(4): 1164 - 1174. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. M. Bilezikjian, A. M. O. Leal, A. L. Blount, A. Z. Corrigan, A. V. Turnbull, and W. W. Vale Rat Anterior Pituitary Folliculostellate Cells Are Targets of Interleukin-1{beta} and a Major Source of Intrapituitary Follistatin Endocrinology, February 1, 2003; 144(2): 732 - 740. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. C. Danila, X. Zhang, Y. Zhou, J. N. S. Haidar, and A. Klibanski Overexpression of Wild-Type Activin Receptor Alk4-1 Restores Activin Antiproliferative Effects in Human Pituitary Tumor Cells J. Clin. Endocrinol. Metab., October 1, 2002; 87(10): 4741 - 4746. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Kawakami, Y. Fujii, Y. Okada, and S. J. Winters Paracrine Regulation of FSH by Follistatin in Folliculostellate Cell-Enriched Primate Pituitary Cell Cultures Endocrinology, June 1, 2002; 143(6): 2250 - 2258. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
X. Zhang, H. Sun, D. C. Danila, S. R. Johnson, Y. Zhou, B. Swearingen, and A. Klibanski Loss of Expression of GADD45{gamma}, a Growth Inhibitory Gene, in Human Pituitary Adenomas: Implications for Tumorigenesis J. Clin. Endocrinol. Metab., March 1, 2002; 87(3): 1262 - 1267. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. C. Danila, J. N. S. Haidar, X. Zhang, L. Katznelson, M. D. Culler, and A. Klibanski Somatostatin Receptor-Specific Analogs: Effects on Cell Proliferation and Growth Hormone Secretion in Human Somatotroph Tumors J. Clin. Endocrinol. Metab., July 1, 2001; 86(7): 2976 - 2981. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||
