| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Other Original Articles |
Departments of Pathology (M.A.J., C.S., D.I.S) and Endocrinology (A.L.C.), Hospital Universitario Virgen del Rocío, 41013 Sevilla; and Department of Physiology, Faculty of Medicine, University of Santiago de Compostela (A.G.U., C.D., C.V.A.), 15782 Santiago de Compostela, Spain
Address all correspondence and requests for reprints to: Dr. Clara V. Alvarez, Department of Physiology, Faculty of Medicine, University of Santiago de Compostela, 15782 Santiago de Compostela, Spain. E-mail: . fscralvi{at}usc.es
Abstract
Glial-derived neurotropic factor (GDNF) signaling is mediated through a 2-component system consisting of the so-called GDNF receptor-
(GFR1), which binds to GDNF. This complex activates the tyrosine kinase receptor RET. In this paper we demonstrate GDNF, GFR1, and RET mRNA and protein expression in the human anterior pituitary gland. Double immunohistochemistry of anterior pituitary sections showed GDNF immunoreactivity in more than 95% of somatotrophs and to a lesser extent in corticotrophs (20%); it was almost absent in the remaining cell types. Also, although more than 95% of somatotrophs were stained for RET, no positive immunostaining could be detected in other cell types. Furthermore, we have looked for GDNF and RET in human pituitary adenomas of various hormonal phenotypes. Strong positive immunostaining was found for c-RET in all of the GH-secreting adenomas screened as well as in 50% of ACTH-producing adenomas. Positive immunostaining for GDNF was found in all of the GH-secreting adenomas and in 10% of the corticotropinomas. Lastly, we found strong positive immunostaining for GFR1 in 90% of the somatotropinomas and 50% of the corticotropinomas as well as in 1 of 8 prolactinomas and 1 of 13 nonfunctioning adenomas. All of the remaining pituitary tumors screened were negative for RET, GDNF, and GFR1.
This study indicates that GDNF may well be acting in the regulation of somatotroph cell growth and/or cell function in the normal human anterior pituitary gland. The expression of RET in all of the somatotropinomas and in 50% of the ACTH-producing tumors implies that GDNF and RET could be involved in the pathogenesis of pituitary tumors.
PITUITARY ADENOMAS HAVE a wide range of biological behavior and can result in significant morbidity. Although they are monoclonal proliferations, somatic mutations are not commonly found in these tumors (1, 2). On the other hand, it has become increasingly evident that locally produced pituitary proteins mediate the development and cellular organization of the anterior pituitary (AP). These locally produced growth factors and cytokines can be involved in the regulation of cell division and specific pituitary hormone gene expression. Although the pathogenesis of the majority of pituitary tumors remains obscure, multifactor mechanisms appear to initiate and maintain their tumorigenesis (1, 2).
Glial cell line-derived neurotropic factor (GDNF) is a distant member of the TGFß superfamily, which is expressed in many neuronal and nonneuronal tissues during development as well as in adult animals (3, 4, 5, 6, 7). GDNF signaling is mediated through a two-component system consisting of a glycosyl phosphatidyl inositol-linked protein, the so-called GDNF receptor- (GFR1), which binds to GDNF. Thereafter, this complex binds to and activates the tyrosine kinase receptor RET (3, 4). Recently, GDNF and RET gene expression have been found in AP glands from male rats (8). In this paper we have sought to characterize the expression of both GDNF and RET in the normal human AP gland as well as to identify the specific cell types within the gland that express these proteins. To this end we used a combination of different techniques, such as RT-PCR, Western blot, and immunohistochemistry. Furthermore, we used an immunohistochemical approach to determine GDNF and RET in human pituitary adenomas of various hormonal phenotypes.
Materials and Methods
This work was approved by the faculty ethical committee. The normal pituitaries used were obtained from the archives of the Institute of Legal Medicine, University of Santiago de Compostela. Neuro- and adenopituitaries had been dissected by the pathologist and snap-frozen immediately. Each piece of tissue was divided in two and processed for both mRNA and protein extraction.
RT-PCR
mRNA was extracted according to the procedures in the Roche Molecular Biochemicals Isolation Kit (Indianapolis, IN), and 1.5 µg mRNA were used in the reverse transcriptase as previously described (8). For GFR1, c-Ret, and hypoxanthine-guanine phosphoribosyl transferase (HPRT), 6 µl of the volume of the RT reaction were used in the PCR step, whereas for GDNF, 9 µl were needed. Primers had the following sequences: GDNF, 5'-GCCCTTCGCGCTGAGCAGTGAC-3' and 5'-GTCGTACGTTGTCTCAGCTGC-3', product of 315 bp (9); c-Ret, 5'-GGATTAAAGCTGGCTATGGCA-3' (exon 10) and 5'-GGAGTAGCTGACCGGGAA-3' (exon 11), product of 267 bp (10); GFR1, 5'-GCACAGCTACGGGATGCTCTTCTG-3' and 5'-GTAGTTGGGAGTCATGACTGTGCCAA TC-3' (11), product of 286 bp; and HPRT, 5'-CAGCCCTGCCGTCGTGATTA-3' and 5'-AGCAAGACGTTCAGTCCTGTC-3' (12), product of 130 bp. For c-Ret, GFR1, and HPRT, the PCR reactions were heated by 35 cycles of 95 C for 1 min, 60 C for 1 min, and 72 C for 1 min, followed by a final extension step of 72 C for 10 min, whereas for GDNF the annealing temperature was 55 C, and the reaction was maintained for 35 or 37 cycles. RT-PCR reactions in the absence of mouse mammary tumor virus reverse transcriptase or RNA were used as negative controls.
Western blot
Tissues were homogenized at 4 C in ice-cold lysis buffer [50 mM HEPES (pH 7.5), 1% Triton X-100, 10 mM EDTA, 10 mM Na4P2O7, 10 mM Na3VO4, 10 mM NaF, 2 µg/ml aprotinin, 10 µg/ml antipain, 5 µg/ml leupeptin, 0.5 µg/ml pepstatin A, and 17 µg/ml phenylmethylsulfonylfluoride]. After centrifugation, supernatants were stored at -80 C. Equal amounts of protein were separated by SDS-PAGE (15% for GDNF, 6% for RET C-19 and C-20, and 12% for GFR1), transferred onto a nitrocellulose membrane, and immunoblotted using a chemiluminescent system (TROPIX) (8, 13). -RET (C-20 and C-19), -GFR1 (C-20), or -GDNF (D-20; Santa Cruz Biotechnology, Inc., Santa Cruz, CA) were used at 1:100 dilutions. Goat antirabbit AP (Perkin-Elmer Corp., Norwalk, CT) or rabbit antigoat AP (Santa Cruz Biotechnology, Inc.) were used at 1:5000 dilutions as secondary antibodies.
Immunohistochemistry
Pituitary adenomas were selected from the archives of the Department of Pathology, Hospital Universitario Virgen del Rocío. Following standardized protocols, tissues had been fixed in either 10% buffered formalin or methacarn (60% methanol, 30% chloroform, and 10% acetic acid), dehydrated, and embedded in paraffin. Clinicopathological diagnoses were as follows: 13 nonfunctioning adenomas, 12 GH-secreting adenomas, 10 ACTH-secreting adenomas, 8 PRL-secreting adenomas, 3 gonadotropin GnD-secreting adenomas, 3 -subunit aSu-positive adenomas, and 1 TSH-secreting adenoma.
For immunohistochemistry, 5-µm-thick tissue sections from paraffin blocks were mounted on silanized slides, dewaxed, and rehydrated. Sections were immersed in 3% H2O2 aqueous solution to inactivate endogenous peroxidase, then washed in Tris buffer saline and covered with 10% normal swine serum (Vector Laboratories, Inc., Burlingame, CA) to block nonspecific immunoreactivity. Antigen retrieval methods were not applied. Primary antibody incubation was carried out overnight at 4 C in a humid chamber. Primary antihuman GDNF (D-20), RET (C-20), and GFR1 (C-20) were used at 1:100, 1:200, and 1:50 dilutions, respectively. Secondary biotinylated antibody and streptavidin-biotin-peroxidase complex (LSAB+ Kit, DAKO Corp., Glostrup, Denmark) were applied to develop immunoreactivity, using 3,3'-diaminobenzidine (DAKO Corp.) as the chromogenic substrate. Sections where primary antibody was omitted were used as negative controls.
Double immunostaining for RET (or GDNF) and pituitary hormones was performed by using sequentially an streptavidin-biotin-alkaline phosphatase method for RET (or GDNF) and peroxidase-conjugated EnVision reagents (DAKO Corp.) for pituitary hormones. Secondary antibodies and primary antipituitary hormone antibodies were obtained from DAKO Corp. Secondary antibodies were selected to avoid cross-reactivities. Alkaline phosphatase was first developed with Fast Blue BB (Sigma, St. Louis, MO), and then peroxidase was developed with 3- amino-9-ethylcarbazole (Sigma).
Results
Figure 1
illustrates the results of RT-PCR analysis showing the expression of GDNF, c-Ret, and GFR 1 in the normal human AP gland. Expected bands of 315, 267, and 286 bp were found, respectively. No signal was detected in negative controls. Immunoblots of extracts from normal human AP gland revealed the presence of the 58-kDa expected band (3) of the GFR1 receptor. We also found the expected 20-kDa size for GDNF in the posterior pituitary and, to a greater extent, in the anterior pituitary. Similarly, in both the anterior pituitary and the posterior pituitary, immunoblots for RET exhibited a specific double band of immature 140-kDa and a fully glycosylated 160-kDa band corresponding to the short isoform of the RET receptor (C-19 antibody) and the immature 150-kDa and fully glycosylated 170-kDa bands of the long isoform of the RET receptor (C-20 antibody; Fig. 1B).
|
). To localize what pituitary cell types express RET or GDNF, we employed a double immunostaining technique. We standardized the technique labeling RET-positive cells with Fast Blue and any pituitary hormone in red, whereas double-stained cells appeared in purple (Fig. 2A). We found that PRL, TSH, LH, FSH, and ACTH cells were very frequently not stained for c-Ret (<5%). Strikingly, most of the GH-producing cells (>95%) were also positive for RET (Fig. 3B). When the same study was carried out using GDNF antiserum, we found GDNF immunoreactivity in almost all of the somatotrophs (>95%) and in some corticotrophs (20%); it was practically absent in the remaining cell types (<3%; Fig. 3B). We obtained similar results when performing an independent immunohistochemistry for RET or GDNF and the different hormones in immediately consecutive pituitary sections (data not shown).
|
|
1. Positive immunostaining for GDNF was found in all of the GH-secreting adenomas and in 10% of the corticotropinomas, whereas it was negative in all of the remaining pituitary tumors screened. Lastly, we found positive immunostaining for GFR1 in 90% of the somatotropinomas and 50% of the corticotropinomas as well as in 1 of 8 prolactinomas and 1 of 13 nonfunctioning adenomas. All of the remaining tumors were negative, but one PRL had doubtful very weak staining. It is noteworthy that there was a near 100% concordance in terms of positive immunostaining for c-Ret and GFR1, e.g. the five ACTH tumors positive for c-Ret were also positive for GFR1, and the negatives were for both of them (Table 1; Fig. 4 is shown as a sample of the immunohistochemistry in the tumors).
|
|
In the present paper GDNF and RET gene expression were unambiguously identified in normal human AP glands. The identification was obtained using several different approaches. The first, using RT-PCR analysis, showed that both GDNF and RET were normally synthesized in the human AP gland. The validity of these RT-PCR results was further confirmed by Western blot as well as by immunohistochemistry, thus indicating that the AP gland is a source and a target tissue of GDNF, similar to previous data obtained in rats (8). Taking into account that the AP gland consists of a heterogeneous cell population, it is imperative to determine which specific cell types express GDNF and the signaling receptor to understand and delineate the roles of GDNF in AP cell function. Therefore, we used double immunohistochemistry to localize the specific cell type that was expressing these genes. At variance with our previous data in rats (8), where GDNF was present in a high number of gonadotrophs (60–80%) and corticotrophs (55%), but was present in a much lesser amount (<15%) in the other cell types, the present data indicate that in the human AP gland GDNF is mostly present in somatotrophs and to a lesser extent in corticotrophs, but is not present in gonadotrophs. In contrast, the expression of RET appears to be restricted to a single cell population of the AP gland, namely the somatotrophs in normal AP glands from both rats and normal humans. Taken together, these data indicate that somatotrophs and corticotrophs are the main source of GDNF in the human AP gland and that the somatotrophs appear to be their target cell type. Therefore, GDNF can be added to the list of growth factors and cytokines produced by the human AP gland.
It is also generally accepted that multifactorial mechanisms subserve the pathogenesis of pituitary adenomas. Dysregulation of early pituitary transcription factors or chromosomal mutations may cause DNA-altering events in an early stem cell. Subsequently, multiple endocrine and paracrine growth factor signals may impinge on the previously "initiated" cell and determine clonal expansion (1, 2, 14). To our knowledge, the expression of GDNF, c-Ret, and GFR1 has not been assessed in pituitary tumors.
Taking into account that in normal pituitaries c-Ret was identified almost exclusively in the somatotroph cells, our data showing that all of the GH-secreting adenomas exhibited positive immunostaining was expected. On the other hand, the finding that 50% of corticotropinomas also showed positive immunostaining for c-Ret was quite unexpected, because corticotrophs from normal human and rat AP glands (8) were found to be negative. Furthermore, we found that the corticotropinomas positive for c-Ret were also the only ones positive for GFR-1, with a 100% concordance. This could indicate the existence of two types of corticotropinomas, one group with cells that could be a biological target of GDNF and another group devoid of the receptor(s) that mediates their effects and therefore that will be unresponsive. Nevertheless, it should be noted that in the present study we used immunohistochemistry to try to identify protein levels of c-Ret, GDNF, and GFR-1 in pituitary tumors. This approach does not allow us to identify gene mutations/deletions. Although there are not many studies, mutations in the c-ret gene do not appear to be frequent in GH-producing adenomas (15, 16). It will be interesting to assess whether the differential expression of GDNF, c-Ret, and GFR1 in ACTH-producing tumors could be correlated with tumor aggressiveness.
It is not possible to ascertain at present the role of pituitary-derived GDNF on somatotroph cell function or in corticotrophs tumors expressing c-Ret. Pituitary growth factors are considered to invariably have dual functions: regulating cell development and replication and controlling differentiated gene expression (14). Because to our knowledge no detailed studies on the effects of GDNF on any of these parameters have been carried out, the physiological significance of our finding is unclear at present. It still remains to be established whether GDNF overexpression and/or alterations in other ligands and receptors, e.g. neurturin or GFR1, at the pituitary can lead to alterations in somatotroph cell proliferation and/or GH secretion.
Acknowledgments
We thank M. Angeles Poveda for her technical assistance, and the Department of Forensic Medicine for their kind help.
Footnotes
This work was supported by grants from the Fondo de Investigación Sanitaria, the Spanish Ministry of Health, Comision Interministerial de Ciencia y Tecnologia, and Consejería de Educación y Ciencia, Junta de Andalucía.
Abbreviations: AP, Anterior pituitary; aSu, -subunit; GDNF, glial-derived neurotropic factor; GFR1, glial-derived neurotropic factor receptor-; GnD, gonadotropin; HPRT, hypoxanthine-guanine phosphoribosyl transferase.
Received February 26, 2001.
Accepted December 18, 2001.
References
with specific cellular patterns of expression suggesting roles in the developing and adult nervous system and in peripheral organs. J Neurosci 17:8506–8519 indicates multiple mechanisms of trophic actions in the adult rat CNS. J Neurosci 17:3554–3567
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| 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 |
