Influence of thyroid hormones on somatostatin processing in cultured cerebro-cortical cells

Neuropeptides (1990) 15,25-30 @ Longman Group UK Ltd 1990

Influence of Thyroid Hormones on Somatostatin Processing in Cultured Cerebro-Cortical Cells M. T. DE LOS FRAILES, F. SANCHEZ FRANCO, M. J. LORENZO, R. TOLGN and L. CAClCEDO Servicio

de Endocrinologia,

Hospital

Ramdn

y Cajal. 28034-Madrid,

Spain (Reprint

requests

to LC)

Abstract-Previous

data from our laboratory showed that prolonged exposure of cultural cerebral cortical cells to high potassium concentrations and veratridine resulted in the stimulation of immunoreactive somatostatin (IR-SRIF) synthesis and caused a major increase in its high molecular weight forms. Somatostatin (SRIF) synthesis by cortical and hypothalamic cells was also affected by thyroid hormone (TH). In the present work we have examin.ed to what extent TH might also affect SRIF processing. Cerebral cortical cells maintained as monolayer culture for 7-10 days received triiodothyronine (Ta) in concentrations of lo-” M and IO-‘M for 48h. We found that the total amount of IR-SRIF was increased by high TB concentrations as reported previously. When the IR-SRIF was characterized by high pressure liquid chromatography (HPLC) or gel filtration, it was evident that thyroid hormone treatment modified the elution profile of IR-SRIF in cells and medium on Bio-Gel P-10 and HPLC, increasing somatostatin 28 (S-28) and decreasing somatostatin 14 (S-14). The results indicate that thyroid hormones affect SRIF processing, leading to a major increase in the synthesis of its high molecular weight forms.

pre-prosomatostatin (4). It has been shown that in addition to serving as an intermediate in the biosynthesis of S-14, S-28 also possesses its own biological activity (5-6). A detailed study of the regional brain distribution of the three immunoreactive forms of somatostatin (7) showed that although S-14 was the predominant molecular species in the central nervous system, highly variable quantities of S-28 and pro-S were also found (8). The relative abundance of the three main forms is different in rat brain, gut and pancreas, (9-10) suggesting differential processing of pro-S to S-14 and S-28 in a tissue-specific manner.

Introduction Tetradecapeptide somatostatin (S-14) was the original form of somatostatin to be isolated (1). Biosynthesis studies have demonstrated the existence of another form, somatostatin-28 (S-28), and the presence of larger precursor forms (pro-S) of the hormone (2-3). S-14 and S-28 arise from the post-translational cleavage of prosomatostatin, which in turn is derived from a larger precursor,

Date received 20 June 1989 Date accepted 27 June 1989

25 B

26

NEUROPEPTIDES

Our current knowledge of the processes regulating somatostatin biosynthesis is still scarce. Recent data from our laboratory (11) demonstrate that prolonged exposure to high potassium increased immunoreactive somatostatin (IRSRIF) synthesis and caused a major increase in its high molecular weight forms. Our previous studies (12) have shown that thyroid hormone regulates somatostatin synthesis. Based on these, the present experiments were performed to determine whether the alteration in somatostatins synthesis was accompanied by modifications in its processing. Our results provide evidence for a regulatory effect of thyroid hormones on somatostatin processing.

Materials and Methods Cell culture Timed pregnant Wistar rats were raised in our laboratory. On day 17 of fetal life, the embryos were removed. The cerebral cortices were taken and collected in Hank’s Balanced Salt Solution (HBSS). Cells were dispersed by mechanical techniques and cultured in Minimum Essential Medium (MEM) supplemented with 10% fetal calf serum (FCS), 10% horse serum (HS), as described previously (12, 13). Final cell suspensions were plated at 5 x lo6 cells/35mm plate and kept in a humidified atmosphere of 5% COZ, 95% air at 37°C. Medium was changed every 5-7 days. Reagents

T3 was obtained from Henning (Berlin, West Germany). Cyclic synthetic somatostatin (S-14) and Tyrl-S-14 were supplied by Bachem Inc. (La Jolla, CA). 1251was obtained from New England Nuclear (Boston, MA). Affi-Gel 10 and Bio-Gel P-10 were obtained from Bio-Rad (Richmond, CA). MEM, HBSS, FCS and HS were purchased from Gibco (Paisley, Scotland). Mimimum Essential Medium (MEM), Hank’s Balanced Salt Solution (HBSS), fetal calf serum (FCS) and horse serum (HS) were purchased from Gibco. Experiments

with T3

On day 7-10 of culture, media were discarded and replaced with fresh MEM without serum. One and

a half ml of MEM containing T3 (lo-“M or lo-‘M) were added to each plate for 48 h. Control groups received 1.5ml MEM. T3 stock solution ( lop3 M) was prepared in NaOH 1 N and diluted in MEM before use. The final pH of the MEM solutions containing T3 was 7.4. Sample preparation

At the termination of the experiments, media were collected, acidified with 100~1 HCl O.lN/dish, boiled for 5min and centrifuged at 12000rpm for 60min. Cells were extracted in HCI O.lN (lmldish) by sonication for 30sec, boiled for 5min, and centrifuged at 12OOOrpm for 60min. The supernatants were kept at -20°C. For Bio-Gel P-10 characterization, samples were pooled from media and cell extracts (3-4ml), purified by affinity chromatography, then lyophilized and resuspended in lml 2N Acetic Acid -0.1% BSAc buffer. For HPLC characterization, samples (200 ~1) were filtered through 0.2 km filters (Sartorios Inc., Mayward, CA). IR-SRIF was quantified by RIA (14) in media and cells extracts, as well as in the eluates from the gel filtration and HPLC columns, using an antiserum against S-14 raised by immunizing rabbits with S-14 bound to bovine thyroglobulin with glutaraldehyde. The final dilution of the antiserum was l/300000 and the cross-reactivity with S-28 was 20-30%. This assay is capable of measuring S-14, S-28 and other N-terminal extensions of S-14, however, it does not recognize S-18(1.,2, or S-28,,_,4,. Assgy sensitivity was 2pg/tube (20pg/ml); the intraassay and the interassay variations were 8% and 15% respectively. Immunoafjinity chromatography

Immobilized ligand was prepared by making AffiGel 10 (3g) react in the presence of ethanolamine (1 M) with immunoglobulin G obtained from 4ml of S-14 antiserum. Media and cell extracts were neutralized with equimolar amounts of NaOH before being applied to the column. The adsorbed material was eluted with 1N acetic acid. The fractions corresponding to the specifically adsorbed material were mixed and characterized by gel filtration.

INFLUENCE OF THYROID HORMONES ON SOMATOSTATIN PROCESSING

CORTICAL CELL MEDIA VO

1.2

szr

su

VI

CONTROL

27

dextran (V,,) 20ml and salt volume (V,), 65ml (free ‘*‘I) . The KD for IR-SRIF peaks was calculated as follows, Kr,: (V,-VJV,-V,), where V, is the elution volume of the IR-SRIF peak. Recovery of immunoreactive material was 75%-80%. HPLC chromatography

Molecular forms of IR-SRIF from media and cell extracts were analyzed using a Clan Bondapak C-18 column (Waters Associates, Milford, MA) at a flow rate of 1.0 ml/min with 25% acetonitrile in 0.2M NaPOh buffer, pH 3.8. Eluted fractions (2 ml) were lyophilized to dryness, reconstituted with RIA buffer and the aliquots tested for IRSRIF. S-28, 1251S-28, S-14 and “‘1 S-14 were run under similar conditions. Recovery of IR-SRIF was 95-100%.

Results 1.2

34%

T, ( IO-’

Y I

Influence of thyroid hormone on the distribution of IR-SRIF molecular forms by gel chromatography

Fig. 1 IR-SRIF characterization in media extracts of cerebrocortical cell culture by gel filtration. The Bio-Gel P-10 column was eluted with 2.0 N acetic acid 0.1% BSA. Cultures were exposed to T3 ( 10-llM and W7M) for 48h. Control received medium alone.

Gel filtration

The material eluted from the affinity column was partitioned by gel filtration through a (70 X 1 cm) column of Bio-Gel P-10 using a 2N acetic acid -0.1% BSAc buffer at 4°C with constant gravity perfusion. For calibration the following markers were used: dextran blue, human insulin (1251h Ins, MW 5800), rat GRF (1251 GRF, MW 5233), synthetic S-28 (MW 3149), 1251S-28, Synthetic S-14 (MW 1638) 1251S-14 and 1251.One milliliter fractions were collected and IR-SRIF measured. Partition coefficients (KD) were determined by OD 280 or by RIA. The characteristics of the column were as follows: exclusion volume of blue

After 7 days in culture, cerebrocortical cells were exposed to T3(10-” and 10e7M) for 24 hours. IR-SRIF from control media and cell extracts was analyzed by Bio-Gel P-10. Partition of media from control cultures yielded six different peaks of IR-SRIF. As depicted in Figure 1 (upper panel), S-14 was the predominant form. The relative abundance of S-14 and S-28, expressed as a percentage of total IR-SRIF was 40% and 25% respectively. T3 (lo-“M) treatment altered the proportion of S-14 and S-28, increasing S-28 from 25% to 35% and decreasing S-14 from 40% to 27%. Exposure to the highest T3 concentration ( 10T7 M) also induced alterations in the chromatographic pattern. There was an increase in the highest molecular weight forms (peak I and II), which represented 34% and 17% of total IR-SRIF respectively. The proportion of Si4 decreased from 40% to 21%. Fractionation of cell extracts on Bio-Gel indicated that cellular IR-SRIF was distributed into six molecular weight species. The relative abundance of S-28 and S-14 was 31% and 45% respectively. A small amount of pro-somatostatin (peak II) was also found. Exposure to Ta (lo-“M) slightly modified the proportion of S-28 by increasing it

28

NEUROPEF’TIDES

analyzed by HPLC. Partition of media from control cultures yielded five distinct peaks of IRSRIF. As shown in Table 1, S-14 accounted for 34% in media and 25% in cells. The relative proportion of S-28 was 8% in media and 23% in cell extracts. T3 ( lo-l1 M) treatment only modified of S-14 in media decreasing it from 34% to 22%. Exposure to 10p7M T3 resulted in an increase of S-28 and a decrease of S-14 in media and cell extracts.

CORTICAL CELL ~xTRACTB VO

1

“I” “i’i’

I .2

i,

z

I .6-

B

1.2-

L

0.6-

CON1 ‘ROL

45% P

o.hs 015 oh

i T,

( IO-“Y

1

\

:

I

results demonstrate that triiodothyronine affects SRIF processing in cultured neuronal cells. These cultures provide a means for studying isolated cortical cells and examining the direct action of Ta (15). It has been shown that such preparations synthesize IR-SRIF in relatively large amounts and release the peptide after membrane depolarization (16, 17). IR-SRIF synthesis in neural cultures was present in six molecular forms, confirming many previous studies,, which suggested a precursor role for such species (18, 19). Our resolution of an IR-SRIF form coeluting with S-28 is in agreement with reports of such molecules in nearly all mammalian somatostatinergic tissues (10, 20,21). The same relative proportion of S-28 in the culture medium as in cell extracts indicates that this form is secreted by cells in culture, just as it is in vivo. S-14 was the prevalent form in the cell extracts. The relatively low proportion of this peptide in the medium could be due to a more rapid destruction of smaller forms in the serum containing medium, Our

0.4-

E

o.hs

ok

o.is i

1.2

Ts ( IO-’

M 1

43% m

0.6

KD Fig. 2 Elution profiles on Bio-Gel P-10 of IR-SRIF present in cell extracts after 48 h incubation in control media and in media containing T3 (lo-“M or lo-‘M). The Bio-Gel P-10 column was eluted with 2.ON acetic acid, 0.1% BSA.

from 31% to 39%. This effect was much more striking when cells were treated with the highest Ts concentration (10m7M), which reduced the proportion of S-14 from 45% to 25% and increased S-28 from 31% to 43%. The relative proportion of other molecular weight forms was also increased. The total amount of IR-SRIF (cell extracts + media) was increased by T3 lo-l1 M and decreased by 10V7M treatment. IR-SRIF (q/plate): control 2.2 + 0.03; T3 (lo-‘iM) 3.02 f 0.2; Ts (10-7M) 1.55 f 0.1. This effect was clearly shown in media and cell extracts separately. HPLC analysis

IR-SRIF

Discussion

from

media

and

cell extracts

was

Table Media

Cells

CONTROL T3 lo-‘M

S28

s14

S28

s14

23% 35%

25% 7%

8% 19%

34% 9%

Values represent percentage of total IR-SRIF. HPLC characterization of IR-SRIF from media and cell extracts of cultured cerebra cortical cells. Cultures were exposed to Ts (lo-“M and lo-‘M) for 48h. Extracts from pooled media and cells were dissolved in 0.01 N HCl. A A Bondapak C-18 column was eluted with 23% acetonitrile in 0.2M PO4 pH 3.8 buffer.

INFLUENCE OF THYROID HORMONES ON SOMATOSTATINPROCESSING

as proposed by Pate1 (9) to explain the predominance of the larger IR-SRIF in blood. Ts modified the chromatographic pattern of IR-SRIF in Bio-Gel suggesting that triiodothyronine regulates not only the synthesis and secretion of SRIF (12, 15,22), but also its post-translational processing. Previous data from our laboratory show that the processing of this neuropeptide is affected by prolonged exposure of cerebra-cortical cells to depolarizing concentrations of potassium and veratridine, agents that stimulate the synthesis of SRIF (11). The present data support the hypothesis that an increase in the rate of synthesis caused by low T3 (lO_l’M) would lead to the appearance of immature forms. Since high thyroid hormone concentrations inhibit protein and RNA synthesis (13), it could also affect the synthesis of enzymes involved in the post-translational processing of somatostatin, leading to the accumulation of immature high-mol-wt forms. It may also reflect changes in the activity of degrading enzymes, proteases and protein kinases induced by thyroid hormones. The possibility that exposure to T3 may increase S-14 degradation is discarded, since we have shown that TJ (lo-‘M) did not modify the degradation of ““I-S14 or synthetic S-14 added to the cultures. HPLC analysis of IR-SRIF from medium and cell extracts confirms and supports the studies using Bio-Gel. S-28 and S-14 have different biological potencies, as has been shown in several assay systems (23, 24), and specific cerebrocortical receptors for S-28 have been reported (25, 28). Thus, changes in the relative proportions of S-28 and S-14 after thyroid hormone treatment could be physiologically significant, providing evidence that thyroid hormones participate in somatostatin processing. Acknowledgements This work was supported by Grants 87/0832 and 87/0841 from the Fondo de Investigaciones Sanitarias. We thank Anna Steele for the preparation of the manuscript.

References 1. Brazeau, P., Vale, W. Burgus, R., Ling, N., Buchter, M., Rivier, J. and Guillemin, R. (1973). Hypothalamicpeptide

29

that inhibits the secretion of immunoreactive pituitary growth hormone. Science 179: 77-79. 2. Noe, B. D., Fletcher, D. J. and Spiess, J. (1979). Evidence for the existence of a biosynthetic precursor for somatostatin. Diabetes 28: 724-730. 3. Noe, B. D. and Spiess, J. (1983). Evidence for biosynthesis and differential post-translational proteolytic processing of different (pre)prosomatostatins in pancreatic islets. J. Biol. Chem. 258: 1121-1129. 4. Goodman, R. H., Montminy, M. R.. Low, M. J. and Habener, J. F. (1985). Biosynthesis of rat preprosomatostatin. In: Y. C. Pate1 and G. S. Tannenbaum (eds) Somatostatin. Plenum Press, New York p. 31-47. 5. Mandarino, L., Stenner. D., Blanchard, W., Nissen, S.. Gerich. J., Ling, N.. Brazeau, P., Bohlen, P., Esch. F. and Guillemin, R. (1981). Selective effects of somatostatin-14. -25. and -28 on in vitro insulin and glucagon secretion. Nature 291: 76-77. 6. Brown, M., Rivier, J. and Vale, W. (1981). Somatostatin 28: selective action on the pancreatic B-cell and brain. Endocrinology 108: 2391-2393. 7. Gomez, S., Morel, A., Nicolas, P. and Cohen, P. ( 1983). Regional distribution of the Mu 15000 somatostatin precursor, somatostatin-28 and somatostatin 14 in the rat brain suggest a differential intracellular processing of the high molecular weight species. Biochem Biophys. Res. Commun. 112: 297-302. 8. Morrison, J. M., Benoit, R., Magistretti, P. J., Bloom, S. R. (1983). Immunohistochemical distribution of prosomatostatin-related peptides in cerebral cortex. Brain Res. 262: 344-351. 9. Patel, Y. C., Wheatly, T. and Ning, C. (1981). Multiple forms of immunoreactive somatostatin: comparison of distribution in neural and non-neural tissues and portal plasma of the rat. Endocrinology 109: 1943-1949. 10. Rostad, 0. P., Epelbaum, J., Brazeau, P. and Martin. J. M. (1979). Chromatographic and biological properties of immunoreactive somatostatin in hypothalamic and extrahypothalamic brain regions of the rat. Endocrinology 10.5: 1083- 1092 11. Cacicedo, L., de 10s Frailes, M. T., Lorenzo. M. J.. Fernandez, G. and Sanchez-France, F. (1989). Influence of chronic depolarization on synthesis and the relative amounts of different forms of somatostatin in cultured fetal cerebrocortical cells. Neuroendocrine Perspective J. A. H. Wass and M. F. Scanlon (eds). Vol6: 229-234. 12. de 10s Frailes, M. T., Cacicedo, L.. Lorenzo, M. J., Fernandez, G. and Sanchez-France. F. (1988). Thyroid Hormone Action on Biosynthesis of Somatostatin by Fetal Rat brain Cells in Culture. Endocrinology 123: 898-904. 13. de 10s Frailes, M. T., Cacicedo, L., Lorenzo, M. J. and Sanchez-France, F. (1989). Divergent effects of acute depolarization on somatostatin release and protein synthesis in cultured fetal neonatal rat brain cells. J. of Neurochemistry 52: 1333-1339. 14. Patel, Y. C. and Reichlin, S. (1978). Somatostatin in hypothalamus, extrahypothalamic brain and peripheral tissues of the rat. Endocrinology 102: 523-530.

30 15. Peterfreund, R. A., Sawchenko, P. E. and Vale, W. (1985). Thyroid hormones reversibly suppress somatostatin secretion and immunoreactivity in cultured neocortical cells. Brain Res. 328: 259-270. 16. Delfs, J., Robbins, R., Connolly, J. L., Dichter, M. and Reichlin. S. (1980). Somatostatin production by rat cerebral neurons in dissociated cells in culture’. Nature 283: 676-677. 17. Robbins, R. J., Sutton, R. E. and Reichlin, S. (1982). Sodium-and-calcium-dependent somatostatin release from dissociated cerebral cortical cells in culture. Endocrinology 110: 496-499. 18. Robbins, R. and Reichlin, S. (1983). Somatostatin biosynthesis by cerebral cortical cells in monolayer culture. Endocrinology 113: 574-581. 19. Spiess, J. and Vale, W. (1980). Multiple forms of somatostatin-like activity in the rat hypothalamus. Biochemistry 19: 2861-2864. 20. Patel, Y. C. and Srikant, C. G. (1985). Somatostatin-14 like immunoreactive forms in the rat: characterization, distribution and biosynthesis. In: Patel, Y. C. and Tannen-

NEUROPEPTIDES

21.

22.

23.

24. 25.

baum, G. S. (eds). Somatostatin. Plenum Press, New York, p, 71-88. Conlon, J., Zyznar. E., Vale, W. and Unger, R. M. (1978). Multiple forms of somatatin-like immunoreactivity in canine pancreas. FEBS Lett 96: 327-330. Berelowitz, M., Maeda, K., Harris, S. and Frohman, L. A. (1980). The effect of alterations in the pituitary-thyroid axis on hypothalamic content and in vitro release of somatostatin-like immunoreactivity. Endocrinology 107: 24-29. Brown, M. R. and Fisher, L. A. (1985). Central system actions of somatostatin-related peptides. In: Patel, Y. C. and Tannenbaum, G. S. (ed) Somatostatin. Plenum Press. New York: 217-228. Srikant, C. B. and Patel, Y. C. (1981). Receptor binding of somatostatin-28 is tissue specific. Nature 294: 259-260. Reubi, J. C., Rivier, J., Perrin, M., Brown, M. and Vale, W. (1982). Specific high affinity binding sites for somatostatinon pancreatic P-cells: differences with brain somatostatin receptors. Endocrinology 110: 1049-1051.