| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
BRIEF REPORT |
Department of Pathology and Laboratory Medicine (G.R., L.B., N.C., C.A., G.G.), University of Parma, 43100 Parma, Italy; Surgical Pathology Service (L.L., M.F.), Ospedale Leno-Manerbio, 25024 Leno (BS), Italy; Department of Human Pathology and Genetics (V.N., E.S.), University of Pavia, 27100 Pavia, Italy; and NEF Laboratory (F.D., C.B., G.-L.F.), Department of Cytomorphology, University of Cagliari, 09042 Monserrato (CA), Italy
Address all correspondence and requests for reprints to: Guido Rindi, M.D., Ph.D., Dipartimento di Patologia e Medicina di Laboratorio, Sezione di Anatomia Patologica, Università di Parma, Via Gramsci, 14, I-43100 Parma, Italy. E-mail: guido.rindi{at}unipr.it.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Aim: The objective of the study was to map proVGF peptides in human endocrine cells during development, adulthood, hyperplasia, and tumors.
Methods: Antisera were generated against peptides related to internal cleavage or cleavage-amidation sites (rat proVGF422–430 and human proVGF298–306-NH2) and the proVGF C-terminal ending (human proVGF607–615). Developing and normal adult endocrine cells, hyperplastic endocrine lesions (thyroid, parathyroid, lung, and stomach), and 120 tumors (102 endocrine) were studied. Immunogold electron microscopy was performed on normal adult pancreas and gut, and Western blotting was performed on extracts of control tissues and endocrine tumors.
Results: proVGF fragments were revealed in developing pituitary, gut, pancreas, and adrenal medulla from 10 gestational weeks, in normal adult pituitary and adrenal medulla, pancreatic glucagon, and insulin cells and gut serotonin cells, in hyperplastic thyroid calcitonin cells, lung P cells, gastric enterochromaffin-like cells, and gastrin cells, and in 88 of 102 endocrine tumors. At electron microscopy proVGF immunoreactivity was restricted to electron-dense granules. Western blotting revealed large molecular weight forms and cleavage fragments in both control tissues and tumor extracts.
Conclusions: proVGF-related peptides are present in endocrine cells early during development and adulthood and increase in hyperplasia and tumors, and proVGF fragments could be novel diagnostic tools for endocrine cells and related lesions, including tumors.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
VGF exerts a critical role in regulating body weight, basal metabolism, and the hypothalamus-hypophysis-gonad axis (3). Recently, the catabolic effect of the proVGF-derived peptide TLQP-21 was demonstrated (4). Conversely, vgf gene expression proved necessary in experimental models of obesity (5). Overall VGF is deeply involved in the control of energy balance (6).
In vitro expression of vgf is induced in neurons by the neurotrophins nerve growth factor and brain-derived neurotrophic factor via both tyrosine kinase receptors and the TNF receptor p75 (7). Other vgf-inducing stimuli include cell depolarization and bioactive molecules as fibroblast growth factor, epidermal growth factor, IL-6, insulin, cAMP, and phorbol homologs (8, 9, 10). In vivo vgf mRNA varies according to neuronal activity, aging, and circadian cycle and is modified by light, neuronal lesions, and in experimental duodenal ulcers (3, 10, 11).
Information on vgf expression mainly focuses on the nervous system (3, 8). Additionally, it was demonstrated in mammalian endocrine cells of the diffuse endocrine system (DES) in pituitary, thyroid, pancreas, and adrenal medulla (3, 8, 9, 10). DES cells constitute a complex regulatory network, share with neurons several antigens, and may generate tumors with significantly different clinicopathological behavior (12). With the exception of data on rat and sheep pituitary (13, 14), detailed information on proVGF-derived peptides in specific endocrine cell type and, more importantly, in man is missing.
The present investigation aims to identify proVGF fragments in endocrine cells of man during development and adult life and related tumors.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Histology
All samples were from the Pathology Units of Brescia, Leno-Manerbio (Brescia), Pavia, and Parma. Tissues were routinely processed into paraffin. For development studies, tissues were from 51 fetuses between 8 and 40 wk of gestation, nine neonates (seven preterm and two at term), and three lactants. For normal adult studies, tissues adjacent to tumors were investigated. Endocrine cell hyperplasia was investigated in thyroid (n = 4), parathyroid (n = 4), lung (n = 5), and gastric mucosa (n = 15). For neoplasia, 120 tumors were investigated, of which 102 were endocrine and 18 were nonendocrine (Table 1
).
|
Antisera were raised in rabbits against synthetic peptides conjugated at the N terminus (13, 14), corresponding to proVGF C terminus (human proVGF607–615), and to two peptides preceding putative cleavage and cleavage-amidation sites (rat proVGF422–430 and human proVGF298–306-NH2, respectively) (Fig. 1). Controls included substitution of layers with PBS, preimmune sera, and absorption with homologous peptide (up to 
50 nmol/ml, which virtually abolished staining). In ELISA, the proVGF298–306-NH2 antiserum cross-reacted less than 0.005% with an identical peptide extended by an additional AA at its C terminus.
|
Electron microscopy
Normal, human adult pancreas and mucosa of gastric fundus, duodenum, and colon were routinely processed in Epon-Araldite, cut, and observed in a EM902 Zeiss (Oberkochen, Germany) transmission electron microscope. Immunogold labeling was performed as described previously (15).
Western blotting
Frozen samples of normal, adult human liver (n = 2), adrenal (n = 3), pituitary (n = 3), and endocrine (n = 6) tumors were homogenized, separated by tricine SDS-PAGE, and blotted onto polyvinylidene difluoride paper (Bio-Rad, Hercules, CA). Western blotting was performed using horseradish peroxidase-coupled secondary antibodies (ECL System; GE Healthcare, Freiburg, Germany).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
proVGF fragment immunoreactivity was restricted to endocrine cells and nerve elements (data not shown). Immunoreactivity was observed with all sera, although antihuman proVGF607–615 proved more efficient. Acidic fixatives such as Bouin’s fluid appeared to better preserve proVGF fragment immunoreactivity. During development, it was observed as early as 10 gestational weeks in the pancreas and at 14–16 wk in pituitary, lung, gut, and adrenal medulla (Fig. 2A) but not in parathyroid and thyroid.
|
, B–D). Immunoreactive cells were absent or exceedingly rare in the stomach but easily detected in the ileum and colon with all sera (Fig. 2E). No immunoreactivity was detected in parathyroid and lung.
For electron microscopy, immunogold labeling was restricted to endocrine granules and consistently observed in pancreatic insulin B, glucagon A cells (Fig. 2, 
and ß), and gut enterochromaffin (EC) cells (Fig. 2, EC). Labeling was more efficient with antihuman proVGF298–306-NH2 compared with antirat proVGF422–430 (data not shown); antihuman proVGF607–615 was not effective.
For Western blotting of normal adult pituitary and adrenal (Fig. 2, F and G), antihuman proVGF607–615 and antirat proVGF422–430 identified specific bands of approximately 80 kDa, corresponding to the largest VGF form (16) or, alternatively, to fragment multimers. In both tissues, both sera identified bands at approximately 50 kDa, whereas the antirat proVGF422–430 identified an additional approximately 30 kDa band, corresponding to proVGF cleavage products. The antihuman proVGF298–306-NH2 identified a faint band at approximately 30 kDa (data not shown). No specific band was detected in liver extracts with any sera (data not shown).
proVGF fragments in hyperplasia and tumors
Hyperplasia.
proVGF fragment immunoreactivity was detected in thyroid calcitonin-producing C cells in patients with concurrent medullary thyroid carcinoma (Fig. 2H), in lung P cells of patients with pulmonary cancer, fibrosis, or tumorlets/carcinoids (Fig. 2I), in corpus enterochromaffin-like and ghrelin cells, and in antrum gastrin cells of patients with chronic atrophic gastritis and hypergastrinemia (Fig. 2, J and K). No immunoreactivity was observed in hyperplastic parathyroid.
Tumors.
Immunoreactivity for proVGF fragments was restricted to endocrine tumors (Table 1). Consistently, no immunoreactivity was observed in five schwannomas and 13 exocrine carcinomas investigated (data not shown). Sera antihuman proVGF607–615 and antirat proVGF422–430 appeared more efficient compared with serum human proVGF298–306-NH2, staining in, respectively, 78, 67, and 54 cases of 102 tested (76, 67, and 53%, respectively). Immunoreactivity for either one or the other proVGF fragment was strong and diffuse in the large majority of well-differentiated tumors/carcinomas (73 of 77, 95%). The staining intensity and percentage of stained cells varied in different tumors and between tumors with the same cell type (Fig. 2, L and M). On average, 46% of cells per positive tumor was stained with antihuman proVGF607–615, 41% with antirat proVGF422–430, and 42% with human proVGF298–306-NH2. The majority of poorly differentiated endocrine carcinomas was negative (15 of 25, 60%), the largest fraction of positive cases being pulmonary large-cell neuroendocrine carcinomas.
Negative cases were three parathyroid adenomas (consistent with negative normal and hyperplastic parathyroid) and one appendix tumor, negative for all sera, suggesting poor tissue preservation.
For Western blotting (Fig. 2a–c), the antihuman proVGF607–615 identified in all tumor extracts an approximately 80 kDa band and a lower intensity of approximately 70 kDa. Minor bands of lower intensity were detected at 30–40 kDa (pheochromocytoma), 40–50 kDa (lung carcinoid and insulinoma), and 50–60 kDa (gastric, duodenal, and ileal tumors). The antirat proVGF422–430 identified in all tumors one band of approximately 30 kDa. Higher-molecular-size bands were also seen at approximately 80 kDa in insulinoma and at approximately 40 and 50 kDa in ileal tumor. Fairly abundant bands at approximately 90, 80, 60, and 40 kDa were detected for pheochromocytoma. The antihuman proVGF298–306-NH2 displayed in all tumors a band between 30 and 40 kDa, whereas a band of approximately 50 kDa was observed in the gastric tumor only.
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
proVGF fragments are present in human endocrine cells
We detected vgf gene expression at the protein level, in situ, using three different sera, avoiding possible bias attributable to tissue fixation and processing. The widespread and intense expression during early fetal life suggests that proVGF fragments may be important for embryo development, consistent with the energy balance role postulated for VGF (6). In normal adult tissues, variable proVGF fragment immunoreactivity was observed in pituitary and adrenal medulla and in some gastroenteropancreatic endocrine cells, of which pancreatic insulin and glucagon cells and gut serotonin cells were confirmed at ultrastructural investigation. For negative endocrine cells, we cannot exclude vgf expression below the detection limits of our methods or the presence of VGF peptides undetected by our sera.
Because immunoreactivity varied in different cells and for different sera, our data suggest a cell-type-specific proVGF processing. Indeed, studies in female rat and sheep pituitary indicate modification in proVGF fragment immunoreactivity according to estrous peak and seasonal changes (13, 14).
For electron microscopy, immunolabeling for proVGF fragments was seen in large dense-core vesicles, as in rodent nerve elements (16, 17), suggesting a regulated pathway of secretion. proVGF fragments were also identified for Western blotting in normal tissue extracts, indicating active cleavage processing. In man, evidence for secretion of proVGF fragments exists in the central nervous system (CNS) fluid (18). Our data indicate that, besides neurons, endocrine cells are a potential source of circulating proVGF fragments.
proVGF fragments increase in hyperplasia and neoplasia
Neurotrophins rapidly induce and regulate vgf expression in rodent CNS during development and adulthood (19). Other vgf-inducing factors comprise several bioactive substances, including growth factors (1, 10, 17). The widespread and intense expression of vgf in endocrine cells of human embryos may reflect the constant tissue modeling of development. proVGF fragments were also abundant in hyperplastic endocrine cells, at sites in which, during adult life, vgf expression is extremely low (stomach) or undetected (thyroid and lung). Hyperplastic endocrine cells appear to regain/acquire the ability to produce detectable amounts of proVGF fragments. This finding suggests that proVGF products label an active/proliferating status of the endocrine cell in response to specific stimuli. Similarly, proVGF fragments were detected in proliferating endocrine tumor cells and in tumor extracts for Western blot, indicating proVGF processing. proVGF fragments in endocrine tumors may account for the various energy balance abnormalities often, but not exclusively, observed in patients with functioning tumors.
The successful measure of VGF peptides in CNS fluid (18) suggests the possibility of their search in the bloodstream as novel and potentially useful endocrine markers. Chromogranin A, a widely used and generally effective granular marker, may have limited efficacy in nonfunctioning tumors at the tissue level (e.g. in colorectal endocrine tumors) or in patient’s plasma (20). proVGF fragments, potentially reflecting an active/proliferating status, could represent an effective alternative for detecting preneoplastic or true neoplastic endocrine lesions.
Conclusion
Our data indicate that 1) proVGF fragments are often found in human endocrine cells during development and, in some cell types, during adult life, and 2) proVGF fragments may be induced in endocrine cells under physiological and pathological stimuli, including transformation, implying an active/proliferating status. VGF, yet another antigen shared by DES cells and nerve elements, could be exploited as a novel endocrine marker for diagnostic purposes.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Part of this work was discussed as the thesis of specialization in Anatomic Pathology by L.L.
G.-L.F. raised VGF antisera, which were characterized by C.B. and F.D.; G.R., L.L., G.G., and M.F. collected and characterized tissue samples; immunohistochemistry was performed by L.L. and N.C.; electron microscopy was performed by V.N. and Western blotting by L.B. and C.A.; G.R. supervised and coordinated the project; and the paper was written by G.R. in cooperation with G.-L.F. and E.S.
Disclosure Summary: The authors have nothing to disclose.
First Published Online April 17, 2007
Abbreviations: CNS, Central nervous system; DES, diffuse endocrine system; EC, enterochromaffin.
Received January 8, 2007.
Accepted April 9, 2007.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  F. D'Amato, B. Noli, C. Brancia, C. Cocco, G. Flore, M. Collu, P. Nicolussi, and G.-L. Ferri Differential distribution of VGF-derived peptides in the adrenal medulla and evidence for their selective modulation J. Endocrinol., May 1, 2008; 197(2): 359 - 369. [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 |
