| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Studies |
Departments of Pharmacology (R.I., R.F.-C., U.E., P.L.-S., A.L., H.W.) and Internal Medicine (R.W.G., G.F.), University of Innsbruck, A-6020 Innsbruck, Austria; Department of Biomedical Science and Human Oncology, University of Torino (A.P.), 10126 Torino, Italy; and Department of Pharmacology, School of Pharmacy, Central University of Venezuela (L.Z.C.), Caracas 1041 A, Venezuela.
Address correspondence and requests for reprints to: R. Fischer-Colbrie, Department of Pharmacology, University of Innsbruck, Peter-Mayr-Str. 1a, A-6020 Innsbruck, Austria. E-mail: fischer-colbrie{at}uibk.ac.at
| Abstract |
|---|
|
|
|---|
In patients with endocrine tumors like gut carcinoids, endocrine pancreatic tumors, oat cell lung carcinomas, and pheochromocytomas, SN serum levels were elevated up to 45-fold. Patients suffering from neuroblastomas, insulinomas, pituitary adenomas including acromegaly, and solid nonendocrine tumors had concentrations in the normal range.
In human serum, SN-immunoreactivity was confined to the free peptide SN; neither larger intermediate-sized forms nor the precursor secretogranin II were detected. An efficient removal of the small molecule SN from the serum by the kidney explains why SN serum levels are lower when compared to chromogranin A, which is present as large molecule in serum.
| Introduction |
|---|
|
|
|---|
Secretoneurin (SN) is a functional peptide representing 33 amino acids in the secretogranin II sequence (24) that provides a novel link between the endocrine and immune system. It induces dopamine release from rat striatum in vivo (25) and in vitro (26) and displays a potent, specific chemotactic affinity toward monocytes (8). Recently, binding sites for SN have been identified on monocytes (27, 28). Plasma SN levels may reflect SN released from endocrine organs or neurons into circulation. To establish the clinical usefulness and the molecular properties of SN, an antibody was used to characterize SN in serum and urine of healthy subjects and of patients with different types of tumors.
| Patients and Methods |
|---|
|
|
|---|
Serum samples were obtained from a group of 42 controls (24 women and 18 men) and patients with the following neuroendocrine tumors and disorders: oat cell lung carcinomas (n = 9), pheochromocytomas (n = 9), gastroenteropancreatic endocrine (GEP) tumors (n = 8; five gut carcinoid tumors and three endocrine pancreatic tumors), prolactinomas (n = 24), acromegaly (n = 16), other pituitary adenomas (n = 37), pituitary insufficiency (n = 18), resection of pituitary (n = 16), and adrenalectomy (n = 6; four unilateral, two bilateral). Three patients with an unilateral adrenalectomy suffered previously from pheochromocytomas. All serum samples besides those of patients who were adrenalectomized or had to undergo resection of the pituitary were obtained before any therapeutical intervention occurred. Serum of patients who received medical treatment such as surgery, chemotherapy, or other appropriate medication, but showed progression of tumor growth, was obtained from oat cell lung carcinoma patients (n = 4) and GEP tumor patients (n = 8). For the neuroblastoma patients (n = 26), a group of 125 healthy children who were 0.516-yr-old was used as controls.
In addition, samples were obtained from patients who had to undergo endocrine stimulation tests: insulin tolerance test (n = 6), glucose tolerance test (n = 7), TRH stimulation test (n = 6), CRF stimulation test (n = 2), combined CRF/TRH stimulation test (n = 2), and combined CRF/TRH/LH-releasing hormone (LHRH) stimulation test (n = 1). Blood samples were taken after 0, 15, 30, 45, 60, 90, and 120 min, with exception of the TRH test, where samples were taken after 0, 15, and 30 min of TRH administration. Insulin was given iv at a dose of 0.10.15 U/kg. TRH (Relefact TRH), CRF (CRH; Ferring Pharmaceuticals Ltd., Hoofdorp, The Netherlands), and LHRH (Relefact LHRH) were administered iv at a dose of 200 µg, 100 µg, and 100 µg, respectively. Glucose (100 g) was ingested orally.
To assess the possible influence of renal function on serum levels of SN, we analyzed serum samples of 21 patients with impaired renal function. The serum creatinine levels ranged from 1.510.2 mg/dL.
Furthermore, 15 serum samples were obtained from patients receiving cisplatin as a chemotherapeutic agent at the recommended dose (20120 mg/m2). Samples were taken before administration of cisplatin and after 5 h.
Urine samples were collected from one patient with a pheochromocytoma, a carcinoid, and seven normal subjects.
Preparation of samples
Blood samples were allowed to coagulate for 1 h at room temperature (sera) or anticoagulated with ethylenediaminetetraacetate (plasma) and then centrifuged for 10 min at 350 x g. Both, serum and plasma were used for analysis. We have previously shown by the addition of intact secretogranin II to rat serum that there is no breakdown of secretogranin II to smaller peptides or to SN even during an incubation for 2 h at 37 C (29). For chromatography, serum samples were boiled for 12 min and then centrifuged at 110,000 x g for 45 min.
Morning urine was collected without the addition of preservatives. Samples were boiled for 12 min, cooled on ice, and centrifuged for 12 min at 14,000 x g. Supernatants were used for analysis.
RIA
The RIA was performed as described (24), with slight modifications. In brief, samples, RIA buffer, and antiserum (final dilution, 1:36,000) were incubated for 48 h at 4 C. Then, iodinated SN (specific activity 77,000 dpm/ng) was added, and samples were incubated for an additional 24 h at 4 C. Bound/free separation was performed with dextran-coated charcoal. The antiserum raised against synthetic rat SN (secretogranin II 154186) reacts with the free peptide and all larger proteins containing the SN sequence (24).
Chromatography
Lyophilized extracts of serum and urine samples were dissolved in column buffer [75 mM sodium phosphate containing 75 mM NaCl and 0.02% NaN3 (pH 7.4)] and loaded on a Superose 12 HR 10/30 fast protein liquid chromatography gel-filtration column (Amersham Pharmacia Biotech, Uppsala, Sweden). Chromatography was performed at a flow rate of 0.4 mL/min. One-min fractions were collected and analyzed by RIA as described above.
| Results |
|---|
|
|
|---|
The mean serum concentration of SN in 42 controls (24 women, 18
men), as determined by RIA, was 22.2 ± 1.1 fmol/mL (range,
9.137.9). The concentration in men was 20.8 ± 1.4 fmol/mL and
in women averaged 23.3 ± 1.6 fmol/mL (Fig. 1
and Table 1
). SN concentrations in the serum were
age dependent (Fig. 2
). SN levels were
5-fold higher in newborns and continuously declined afterward. At the
age of 12, adult levels were reached.
|
|
|
Specific endocrine stimulation tests such as TRH-, CRF-, combined
TRH/CRF, combined TRH/CRF/LHRH stimulation test, and glucose tolerance
test did not lead to any changes of SN levels, whereas during the
insulin tolerance test an elevation of SN in serum was observed in all
patients (n = 6). The individual patients showed differences in
the time point and in the magnitude of the increase (Fig. 3
). The increase reached statistical
significance (P < 0.05) when the respective highest
elevation of each patient was compared to baseline levels. For
comparison, plasma glucose levels of each patient are presented (Fig. 3
).
|
|
|
For neuroendocrine tumors (Table 1
and Fig. 1
), the highest
concentrations of SN were found for gut carcinoids and endocrine
pancreatic tumors, where practically all values were above controls (up
to 45-fold). When the tumors were treated by surgery, by interferon
, or somatostatin analogs, which inhibit secretion from the tumor
and, therefore, also probably SN, most levels returned to normal (Table 1
). High levels were also observed in pheochromocytoma patients, where
only few patients were within the normal range. After resection of the
tumors, elevated SN values returned to normal. Approximately half of
the patients with oat-cell carcinomas showed elevated SN concentrations
in serum (Fig. 1
). Nearly all patients with neuroendocrine tumors and
disorders of the pituitary, such as prolactinomas, other pituitary
adenomas, or acromegaly, had SN concentrations in the normal range
(Fig. 1
and Table 1
). Neuroblastoma patients (04 yr) showed
age-dependent high SN concentrations, but no elevated levels, when
compared with controls of the same age (Table 1
).
The effect of cisplatin on SN concentrations in serum was studied in 15
patients with nonendocrine malignancies. In 13 of 15 patients, an
increase of SN could be detected. Pre-cisplatin levels were 16.7
± 1.4 fmol/mL (range, 9.625.8 fmol/mL) and increased to 23.9 ±
2.6 fmol/mL (range, 843.5 fmol/mL) after cisplatin. An increase of
42.8% above basal levels was calculated (P < 0.01,
paired t test). Interestingly, basal pre-cisplatin levels of
patients with solid tumors of nonendocrine origin had SN levels within
the limits for healthy subjects (Table 1
).
The molecular form of SN-IR in tumor patients with elevated SN levels
(one gut carcinoid patient, an islet cell carcinoma patient, and a
pheochromocytoma patient) was identical to that of controls (Fig. 5
).
SN levels in the urine
Urine measurement of normal controls (n = 7) demonstrated SN
concentrations of 80 ± 7.3 fmol/mL. An analysis by gel-filtration
chromatography of the urine of five controls and one carcinoid patient
(165 fmol SN/mL) demonstrated one consistent peak of SN-IR at the
position where intact SN elutes and various smaller peaks corresponding
to breakdown products of SN (Fig. 6
).
|
| Discussion |
|---|
|
|
|---|
In urine, 80 fmol/mL SN were measured. Because SN, a peptide of 33 amino acids, is likely to be filtered efficiently in the glomeruli, one would expect much higher concentrations in the urine. There is evidence, however, that serum peptides are degraded by proteases present in the tubular epithelia (30). In fact, the molecular characterization of SN in urine demonstrated considerable breakdown of SN to small peptides with a length of 620 amino acids, which were not analyzed any further. There is only one previous study (21) measuring secretogranin II immunoreactivity in serum and urine. The antibody used in that study was raised against a peptide representing amino acids 312331 of the secretogranin II sequence. Only levels of immunoreactivity were determined; a molecular characterization due to a lack of sensitivity of the RIA was not possible. The immunoreactive levels in serum were around 80 fmol/mL and in urine were 500 fmol/mL. Thus, the relative ratio between serum and urine compares quite well with our data, whereas the absolute values are somewhat higher.
Serum SN levels were dependent on the age of the subjects. The SN serum levels are higher in young children and then decline as the subject ages. This is not related to a different proteolytic processing of the secretogranin II precursor, but may indicate a difference in secretory rate from the tissues (e.g. intestine) or clearance in kidney. Although the latter seems unlikely since only when creatinine levels in patients with kidney failure become high, SN levels due to reduced clearance in kidney started to increase. In this context, it is interesting to note that higher childhood levels have also been reported for neuropeptide Y (31).
Cisplatin has been shown to release large quantities of chromogranin A from the intestine, with serum concentration increasing several-fold (32, 33). Only a very limited rise was seen for SN. This may indicate that the neuroendocrine cells (enterochromaffin cells) stimulated by cisplatin contain much more chromogranin A. It may also be related (see below) to the fact that SN is efficiently removed from serum, making a significant increase difficult to achieve and to maintain.
In patients with endocrine tumors, significant elevation in SN levels were found in all patients having carcinoids and in most with oat cell carcinoma and pheochromocytomas, whereas there was no change in prolactinomas, acromegaly, other pituitary adenomas, and patients with solid, nonendocrine tumors. In neuroblastoma patients there was no difference in comparison with age-matched children, taken their higher serum level into account.
Let us now compare all these data with those previously obtained with chromogranin A. In contrast to secretogranin II, intact chromogranin A can be found in serum (21, 34, 35), and this is also true for serum of patients with endocrine tumors (21, 34). Because chromogranin A represents a large molecule, it will not be filtered efficiently in the kidney, and if some is filtered, it should become reabsorbed in the kidney tubules like albumin. In fact, we have shown for rats that the kidney tissue contains a relatively high concentration of chromogranin A, and this can be localized to lysosomal structures involved in re-uptake and processing of large molecules (35). In contrast, only traces of SN are found in kidney tissue, indicating that this small molecule is not taken up for processing but degraded during tubular passage (see above).
What consequences does this difference of molecular size for chromogranin A and SN in serum have? SN is efficiently filtered in the kidney, whereas for the larger chromogranin A molecules this is obviously not so. Therefore, if tumors secrete chromogranin A and SN into the serum, the latter one will be removed faster. One would, therefore, expect higher steady-state levels for chromogranin A in patients with endocrine tumors and much lower ones for SN. We have already pointed out in the introduction that tumors contain comparable levels of chromogranin A and SN-IR, and, in fact, neuroblastoma contains even higher levels of the latter one. Despite this, we have found no increase in SN for neuroblastoma patients, although chromogranin A levels are elevated in this disease (36). In carcinoid and pheochromocytoma patients, chromogranin A levels up to 1000 times the normal ones are found, whereas for SN the increases were less (up to 45-fold). Thus, we would like to postulate that a large peptide secreted from tumors (like chromogranin A) may be a better serum tumor marker than a smaller one (e.g. SN), and, in fact, our results establish that chromogranin A is a much more sensitive serum tumor marker than SN.
| Footnotes |
|---|
Received July 9, 1999.
Revised October 5, 1999.
Accepted October 15, 1999.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
R. Kirchmair, M. Egger, D. H. Walter, W. Eisterer, A. Niederwanger, E. Woell, M. Nagl, M. Pedrini, T. Murayama, S. Frauscher, et al. Secretoneurin, an Angiogenic Neuropeptide, Induces Postnatal Vasculogenesis Circulation, August 31, 2004; 110(9): 1121 - 1127. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. Yon, J. Guillemot, M. Montero-Hadjadje, L. Grumolato, J. Leprince, H. Lefebvre, V. Contesse, P.-F. Plouin, H. Vaudry, and Y. Anouar Identification of the Secretogranin II-Derived Peptide EM66 in Pheochromocytomas as a Potential Marker for Discriminating Benign Versus Malignant Tumors J. Clin. Endocrinol. Metab., June 1, 2003; 88(6): 2579 - 2585. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. Taupenot, K. L. Harper, and D. T. O'Connor The Chromogranin-Secretogranin Family N. Engl. J. Med., March 20, 2003; 348(12): 1134 - 1149. [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 |