Fish Physiology and Biochemistry vol. 7 nos 1-4 pp 367-374 (1989) Kugler Publications, Amsterdam/Berkeley

Studies on the structure and physiology of salmon teleocalcin Graham F. Wagner and Henry G. Friesen Department of Physiology, University of Manitoba, Faculty of Medicine, 770 Bannatyne Avenue, Winnipeg, Manitoba, Canada R3E OW3 Keywords: teleocalcin, calcium, corpuscles of Stannius, gill function, prolactin

Abstract The structure and physiology of salmon teleocalcin, a Ca+2 regulating hormone from the corpuscles of Stannius (CS) is reviewed. Teleocalcin is produced by the PAS +, type 1 cells in the CS. The hormone is a disulfide-linked homodimer, with a unique amino acid sequence and a carbohydrate moiety on residue 29. The teleocalcin monomer has a MW of 30 KD, whereas the pro-form of the monomer is 32 KD. The hormone is positively regulated by Ca + 2 and its function is to slow the active transport of Ca + 2 across the gill epithelium. In conjunction with prolactin, which stimulates Ca+ 2 transport, teleocalcin is one of the major factors involved in Ca + 2 homeostasis in fish.

Introduction Our studies on teleocalcin (also known as hypocalcin), the active principle from the corpuscles of Stannius (CS), began 4 years ago. In spite of more than 20 years of research on the glands, little was known about either the size or composition of the hormone. What was known however, was that the active principle inhibited gill Ca + 2 transport (Fenwick and So 1974; So and Fenwick 1977, 1979), and this was the basis for the bioassay we developed to monitor various fractions during our isolation of teleocalcin. Histological studies had also revealed that the major cell type in the glands, the type 1 cells, had secretory granules that were highly reactive to periodic acid Schiff (PAS) reagent (Krishnamurthy 1976), indicating that the hormone was possibly a glycosylated protein. However, detailed information on the size of the hormone was lacking. Originally, it had been proposed that teleocalcin was a 3 KD glycopeptide (Ma and Copp 1978),

but later it was found that the acid acetone extraction procedure used for its isolation completely destroyed what is now known to be teleocalcin, and that this 3 KD peptide was only an artifact (Wagner, unpublished data). It was also suggested that the active principle was immunologically related to parathyroid hormone (Lopez et al. 1984), but on the basis of what is now known about the size and amino acid sequence of teleocalcin (Wagner et al. 1986a; Butkus et al. 1987; Lafeber et al. 1988; Wagner et al. 1988b), there appear to be no structural similarities between these two peptide hormones. Studies using SDS electrophoretic analysis of salmon CS extracts showed that the glands contained a major group of proteins in the 27-32 KD range. These proteins were also highly reactive to PAS reagent (PAS +) (Wagner, unpublished data). Comparison of acidic and neutral extractions of salmon CS showed no difference in the electrophoretic properties of the PAS + bands, or in the

368 subsequent bioactivity of the extracts in rainbow trout (Salmo gairdneri), both of which confirmed earlier observations on the acid stability of the hormone (Fenwick 1982). These PAS + proteins were then shown to bind to and elute from Concanavalin A Sepharose (Con A), and the biological activity was seen to reside in the Con A bound fraction (Wagner, unpublished data). This confirmed the glycosylated nature of the hormone and demonstrated the ease with which it could be concentrated by conventional lectin chromatography. It had also been shown that the release of secretory granules by the type 1 cells was stimulated by increasing the levels of extracellular Ca + 2 (Aida et al. 1980). A series of studies was therefore conducted whereby whole glands were cultured in the presence or absence of 3 mM Ca + 2 . When the media from these cultures were analyzed by SDS electrophoresis, these same PAS + proteins were consistently observed in the high Ca+ 2 media alone (Wagner, unpublished data). On the basis of these findings (that the hormone appeared to be a 27-32 KD, acid stable glycoprotein), a three stage isolation procedure was developed that involved affinity, gel exclusion and ion exchange chromatography with which teleocalcin was isolated in high yield (0.2-0.4% of wet weight) from both sockeye (Oncorhynchus nerka) and coho (0. kisutch) salmon.

The structure of teleocalcin Perhaps the most notable feature of salmon teleocalcin is its unique amino acid sequence (Wagner et al. 1986a, 1988b). This also holds true for the hormone in the Australian eel (Anguilla australis)(Butkus et al. 1987) and the rainbow trout (Lafeber et al. 1988), indicating that the teleocalcin family shares no homology with any other known proteins. One other salient feature of the hormone is its homodimeric structure. This was deduced from its changing electrophoretic mobility in the native and reduced state. Native teleocalcin migrated as one or more poorly resolved bands with a MW of 40-50 KD, whereas in the presence of a reductant, two distinct bands of 27 and 30 KD were apparent. In view of the fact that amino acid sequencing studies yield-

ed just one N-terminal sequence, it was concluded that the hormone had to consist of two identical, disulfide-linked polypeptide chains. This has since been confirmed by Butkus et al. (1987), who have cloned and sequenced the cDNA encoding the teleocalcin monomer from the Australian eel and found it to be rich in 1/2 cystines; all or some of which participate in inter-chain bonding. Only one other polypeptide hormone, Mullerian-inhibiting substance, is known to have a similar homodimeric structure (Donahoe et al. 1987). The difference in the size of the monomers in the purified salmon hormone is probably due to variable glycosylation (Wagner et al. 1986a). The salmon teleocalcins are similar in both amino acid and carbohydrate composition, but there are two substitutions in the N-terminal sequence that has been deduced thus far (95% homology in the first 40 residues). The same regions in eel (Butkus et al. 1987) and trout teleocalcin (Lafeber et al. 1988) are 80% and > 90% homologous to the salmon hormones by comparison. Residue 29 has defied identification in salmon and trout. In the eel, an asparagine occupies this position and forms an integral part of the glycosylation consensus sequence - Asp-X-Ser(Thr) - at positions 39 through 31. Therefore, given the high degree of homology in the teleocalcin family, and that the insertion of an asparagine at residue 29 creates this same consensus sequence in both salmon (Wagner et al. 1988b) and trout (Lafeber et al. 1988), it is safe to assume that it is glycosylated asparagine. There is also evidence of a pro-form of teleocalcin based on the predicted sequence from the Australian eel cDNA. Butkus et al. (1987) identified a 15 residue pro-sequence upstream from the N-terminus that would add approximately 2 KD to the 30 KD monomer. We also have obtained evidence for a pro-teleocalcin in salmon and trout based on Western blot analysis. Crude CS extracts have the same banding pattern as teleocalcin on Western blots (Fig. A-C), but there is a higher mol wt band in the extracts (32 KD) that has never been observed in our purified hormone preparations (Wagner et al. 1988a). This 32 KD band has also been observed after biosynthetic labelling of primary cultured trout CS cells (Gellersen et al.

369

Fig. 2.

Fig. 1.

1988). In this case, the cells were labelled with [S]methionine for 24 hours and secretion was stimulated with 1.8 mM Ca + 2 to promote the release of newly synthesized products. The conditioned media was then treated with either teleocalcin antiserum or Con A Sepharose to recover 35 [S]teleocalcin and 3 5 [S]glycoproteins respectively. The latter were subjected to SDS electrophoresis and fluorography. The results of this study (Fig. 2) revealed that the same labelled products were recovered by both techniques; two bands with MW of 50 and 52 KD in the native state (Fig. 2A), or 30 and 32 KD in the reduced state (Fig. 2C and D). Furthermore, when the intracellular forms of teleocalcin were examined by extraction and immunoprecipitation of labelled cells, they proved to be identical in size (30 and 32 KD) to the secreted forms of the hormone (Fig. 2E). On the basis of the lectin binding properties of both bands and the specificity of our antiserum, we suspect that the 30 KD band (50 KD native) is the mature, glycosylated monomer and that the 32 KD band (52 KD native) is its precursor. The absence of pro-teleocalcin in purified hormone preparations from both salmon and trout CS (Lafeber et al. 1988), suggests that it may 35

be an extremely labile form of the hormone. However, the fact that the pro-hormone is secreted in vitro suggests that some of it must enter the circulation. At present, this is an area of research that remains completely unexplored. But, if the pro-hormone could be purified in sufficient quantities for the development of specific antisera, this would facilitate studies on rates of secretion and its fate in the bloodstream.

Cellular origin of teleocalcin Krishnamurthy (1976) has extensively reviewed the literature pertaining to the histophysiology of the CS and more recent findings have been included in a new update on CS physiology (Wendelaar Bonga and Pang 1986). As both reviews explain, the most characteristic histological feature of the glands is that the major cell type present, the type 1 cell, is PAS +. In view of the fact that purified teleocalcin is also sensitive to PAS staining (Wagner et al. 1986a), we have been attempting to prove that the type 1 cells are the source of this hormone using correlative PAS and immunocytochemical staining. The species under study have included chinook (0. tschawytcha) and coho salmon, rainbow trout, arc-

370

Fig. 3.

tic char (Salvelinus alpinus) and North American eels (A. rostrata). Following PAS staining in these species, the CS cells were sometimes weakly stained, whereas at other times they were stained quite intensely. Often there was a variable pattern of staining within a single gland. However, the pattern of immunostaining that was obtained in adjacent tissue sections was generally quite different. Irrespective of the intensity of PAS staining, most of

the cells were highly immunoreactive and homogeneously stained throughout except for isolated type 2 cells, and there was no correlation between the cells stained by the two techniques. The only instance in which we have observed good correlative staining has been in sockeye salmon after their spawning migration (Wagner et al. 1988a). The CS from these fish had a unique appearance when viewed by light microscopy, as most of the secre-

371 tory granules had fused and were densely packed together at the basolateral membranes, immediately adjacent to the capillaries. These tightly packed granules were found to be highly reactive to both PAS (Fig. 3A) and teleocalcin antiserum (Fig. 3B) and provided the first conclusive evidence that the hormone was, in fact, produced by the type 1 cells. Our inability to obtain correlative staining in the other species probably reflects the relative sensitivity of the two staining procedures. For example, the teleocalcin antiserum recognizes the mature, glycosylated hormone and the in vitro translated, nonglycosylated teleocalcin monomer as well (Wagner et al. 1988a). Therefore, when employed in immunocytochemistry this antiserum should recognize all intracellular forms of the hormone, from the newly synthesized monomer to the glycosylated homodimer. This is probably why the teleocalcin cells are homogeneously stained in most instances. On the other hand, tinctorial staining procedures such as PAS are far less sensitive as they are not enzymatically based reactions. As a consequence, cells which have granules homogeneously distributed throughout the cytoplasm (as opposed to concentrated basolaterally as in sockeye salmon) or which contain few granules, will be weakly stained or not stained at all. The one drawback to the CS histology as illustrated in Fig. 3B, however, is the difficulty in identifying the unstained, type 2 cells. The latter are actually more easily distinguished in species which have homogeneously distributed granules and, consequently, poor correlative staining. After immunocytochemical staining of the CS in these fish, the type 2 cells stand out in sharp contrast to the darkly stained type 1, teleocalcin cells (Wagner, unpublished).

Teleocalcin physiology Much of the background work on the mechanism of action of teleocalcin was done prior to isolation and characterization of the hormone. First of all, numerous laboratories reported that stanniectomy (STX) of eels resulted in hypercalcemia, which could be partially corrected by the administration

of CS extracts or prevented with ectopic CS transplants (Fontaine 1964; Pang et al. 1973, 1974). However, the observation that post-STX hypercalcemia developed in direct proportion to environmental Ca + 2 levels (Fenwick 1974; Pang et al. 1973), led to the hypothesis that accelerated gill Ca + 2 transport was the cause of the rising plasma Ca + 2 levels. The gill perfusion studies in Fenwick's laboratory confirmed this hypothesis by showing that the removal of the glands led to accelerated gill Ca + 2 transport, and that the CS contained a factor which reversed this process (So and Fenwick 1977, 1979). The bioassay for teleocalcin was based on the inhibition of gill Ca + 2 transport and involved monitoring the rate of 45Ca influx (whole body) by rainbow trout fry (Wagner et al. 1985a; Wagner et al. 1986a). With this bioassay, it was demonstrated that both salmon teleocalcins had potent inhibitory effects on gill Ca+2 transport in trout and North American eels, and could reverse the accelerated gill transport in STX eels in the same manner as crude CS extracts. However, the most interesting findings have been obtained with juvenile rainbow trout in which gill Ca + 2 influx is cyclical, with a periodicity of approximately two weeks (Wagner et al. 1985a), and where teleocalcin is only bioactive during periods of peak influx (Wagner et al. 1986a, 1988b). This suggests that there may be cyclical secretion of the factors regulating transport (prolactin and teleocalcin?) and/or fluctuations in the receptiveness of the gills to these hormones. Correlative studies on plasma hormone levels and Ca + 2 influx in juvenile trout are currently in progress and should provide answers to these questions. However, given that plasma Ca + 2 is tightly regulated in fish, speculation arises as to the underlying purpose of cyclical transport. It may be linked to the cyclical growth patterns that occur in trout (Wagner and McKeown 1985, 1986), such that there may be a fluctuating demand for Ca+2 imposed by a changing growth rate. What is clearly needed are studies on both Ca + 2 influx and growth rate in these juvenile fish to establish if there is a correlation between these two parameters. If so, this would suggest a growth-related function for the hormone in rapidly growing fish,

372

E 0)

400

zz

300

_J

00

200

w

-J

I- 100

mM IONIC CALCIUM Fig. 4.

in addition to its role in Ca + 2 balance. In spite of its effects on gill Ca + 2 transport, there are few other known functions for teleocalcin and this should be a major thrust of future work. The most obvious target for the hormone is the kidney, given the close vascular and neural connections between the corpuscles and this organ, but so far, the effects of stanniectomy on kidney function have been contradictory (Butler 1969; Fenwick 1974). Now that the purified hormone is widely available, hopefully information will soon be forthcoming with respect to effects on renal physiology. Other areas that need to be explored further include effects on the gut and vitamin D production, growth of bone, cartilage and skeletal muscle, as well as a role for teleocalcin in reproduction.

The regulation of teleocalcin secretion Numerous studies over the last 20 years have demonstrated the responsiveness of the type 1 cells to Ca + 2 through histological studies. However, the actual release of hormone in response to Ca + 2 and the range of this response have never been quantified. To address this specific issue, a primary culture system for rainbow trout CS cells and a radioimmunoassay to monitor teleocalcin secretion were developed (Gellersen et al. 1988). The results of these studies have shown that teleocalcin is positively regulated by ionic Ca + 2, the response to Ca + 2 is rapid (within 0.5 h) and that secretion is not stimu-

lated by changing osmolality (240-320 mOsm with NaCI) or equivalent concentrations of magnesium. Furthermore, Ca+2 -stimulated secretion could be reversed with cobalt and mimicked with ionophores, indicating the importance of Ca + 2 entry in stimulus-secretion coupling (Wagner et al. 1989). These studies have since been repeated with rainbow trout in vivo with essentially the same result (Wagner, unpublished data). In view of the positive regulation of teleocalcin release by Ca + 2 and the inhibitory effects of teleocalcin on gill Ca + 2 transport, it appears that the function of this hormone is the prevention of hypercalcemia. Prolactin, on the other hand, appears to act in direct opposition to teleocalcin by stimulating gill Ca + 2 transport (Flik et al. 1986) and is negatively regulated by Ca + 2 in salmon (Fargher 1988). Therefore, the two hormones function and appear to be regulated in a completely opposite manner. As prolactin has no direct effects on teleocalcin secretion (Wagner, unpublished data) it may counteract the effects of teleocalcin directly at the level of the gills and other target tissues. Whether or not teleocalcin has a direct effect on prolactin secretion still needs to be investigated, as do the potential interactions of these hormones with calcitonin and vitamin D.

Conclusions There has been dramatic progress in the last several years, in terms of our knowledge of the corpuscles of Stannius. The active principle, teleocalcin or hypocalcin, has been isolated from several salmonid species and in the case of the eel, cloned and completely sequenced. Unfortunately, our knowledge of teleocalcin physiology has not progressed nearly as much, and this should now be the thrust of future work. One of the most serious problems at present, is the absence of a suitable bioassay for standardizing the different hormone preparations that are available. Most of the bioassays in use at present employ live fish and some of these species are available only seasonally. The ideal bioassay, in our view, would be an in vitro one utilizing gill or kidney membrane preparations and measuring

373 the product(s) of hormone-receptor interaction (cAMP?). But, for this development to occur, the precise mechanism by which teleocalcin acts at the tissue level must first be elucidated. Until such time, all those in the CS field will be working with hormone preparations of variable potency and purity. However, the wide availability of the hormone and the speed with which comparative endocrinology is moving forth will ensure that there will be rapid progress in this area. Given the numerous avenues of research to be explored, some of which have been outlined in this review, the future of CS physiology has never looked brighter.

Acknowledgements We would like to thank all those who actively contributed to the work discussed in this review. They include; Harold Copp, Helle Cosby, Leonard Deftos, Rosa Estagamet, James Fenwick, Agnes Freznosa, Birgit Gellersen, Margaret Hampong, San Yu Lok, Christine Milliken and Carol Park. This work was supported by grants from the Natural Sciences and Engineering Council of Canada and the Medical Research Council of Canada.

References cited Aida, K., Nishioka, R.S. and Bern, H.A. 1980. Degranulation of the corpuscles of Stannius of coho salmon (Oncorhynchus kisutch) in response to ionic changes in vitro. Gen. Comp. Endocrinol. 41: 305-3413. Butkus, A., Roche, P.J., Fernley, R.T., Haralambidis, J., Penschow, J.D., Ryan, G.B., Trahair, J.F., Tregear, G.W. and Coghlan, J.P. 1987. Purification and cloning of a corpuscles of Stannius protein from Anguilla australis.Mol. Cell. Endocrinol. 54: 123-134. Butler, D.G. 1969. Corpuscles of Stannius and renal physiology in the eel (Anguilla rostrata). J. Fish. Res. Bd. Can. 26: 639-654. Donahoe, P.K., Cate, R.L., MacLaughlin, D.T., Epstein, J., Fuller, A.F., Takahashi, M., Coughlin, J.P., Ninfa, E.G. and Taylor, L.A. 1987. Mullerian-inhibiting substance: Gene structure and mechanism of action of a fetal regressor. Rec. Progr. Horm. Res. 43: 431-467. Fargher, R.C. 1988. Biochemical and physiological studies on salmon prolactin. Ph.D. Thesis, Simon Fraser University, Canada.

Flik, G., Fenwick, J.C., Kolar, Z., Mayer-Gostan, N. and Wendelaar Bonga, S.E. 1986. Effects of ovine prolactin on calcium uptake and distribution in Oreochromis mossambicus. Am. J. Physiol. 250: R161-166. Fenwick, J.C. 1974. The corpuscles of Stannius and calcium regulation in the North American eel (Anguilla rostrata Lesueur). Gen. Comp. Endocrinol. 29: 127-135. Fenwick, J.C. and So, Y.P. 1974. A perfusion study of the effects of Stanniectomy on the net influx of calcium-45 across an isolated eel gill. J. Exp. Zool. 188: 125-131. Fontaine, M. 1964. Corpuscles de Stannius et regulation ionique (Ca, K, et Na) du milieu interieur d'un poisson l'anguille. C.R. Acad. Sci. Ser. D529: 875-878. Gellersen, B., Wagner, G.F., Copp, D.H. and Friesen, H.G. 1988. Development of a primary culture system for rainbow trout corpuscles of Stannius and characterization of secreted teleocalcin. Endocrinology 123: 913-921. Krishnamurthy, V.G. 1976. Cytophysiology of corpuscles of Stannius. Int. Rev. Cytol. 46: 177-249. Lafeber, F.P.J.G., Hanssen, R.G.J.M., Choy, Y.M., Flik, G., Herrmann-Erlee, Pang, P.K.T. and Wendelaar Bonga, S.E. 1988. Identification of hypocalcin (teleocalcin) isolated from trout Stannius corpuscles. Gen. Comp. Endocrinol. 69: 19-30. Ma, S.W.Y. and Copp, D.H. 1978. Purification, properties and action of a glycoprotein from the corpuscles of Stannius, which affects calcium metabolism in the teleost. In Comparative Endocrinology pp. 283-286. Edited by P.J. Gaillard and H.H. Boer. Elsevier-North-Holland, Amsterdam. Pang, P.K.T., Pang, R.K. and Sawyer, W.H. 1973. Effects of environmental calcium and replacement therapy on the killifish, Fundulus heteroclitus, after surgical removal of the corpuscles of Stannius. Endocrinology 93: 705-710. So, Y.P. and Fenwick, J.C. 1977. Relationship between net 45 calcium influx across a perfused isolated eel gill and the development of post-stanniectomy hypercalcemia. J. Exp. Zool. 200: 259-264. So, Y.P. and Fenwick, J.C. 1979. In vivo and in vitro effects of Stannius corpuscle extract on the branchial uptake of 45Ca in stanniectomized North American eel (Anguilla rostrata). Gen. Comp. Endocrinol. 37: 143-149. Wagner, G.F., Copp, D.H. and Friesen, H.G. 1988a. Immunological studies on teleocalcin and salmon corpuscles of Stannius. Endocrinology 122: 2064-2070. Wagner, G.F., Fenwick, J.C., Park, C.M., Milliken, C., Copp, D.H. and Friesen, H.G. 1988b. Comparative biochemistry and physiology of teleocalcin from sockeye and coho salmon. Gen. Comp. Endocrinol. 72: 237-246. Wagner, G.F., Gellersen, B. and Friesen, H.G. 1989. Primary culture of teleocalcin cells from rainbow trout corpuscles of Stannius: Regulation of teleocalcin secretion by calcium. Mol. Cell. Endocrinol. (In press). Wagner, G.F., Hampong, M. and Copp, D.H. 1985a. A cycle for 45 Ca uptake in the rainbow trout, (Salmo gairdneri), Can. J. Zool. 63: 2778-2779. Wagner, G.F., Hampong, M., Park, C.M. and Copp, D.H.

374 1986a. Purification, characterization and bioassay of teleocalcin, a glycoprotein from salmon corpuscles of Stannius. Gen. Comp. Endocrinol. 63: 481-491. Wagner, G.F. and McKeown, B.A. 1985. Cyclical growth in juvenile rainbow trout (Salmo gairdneri). Can. J. Zool. 63: 2473-2474. Wagner, G.F. and McKeown, B.A. 1986. Development of a

salmon growth hormone radioimmunoassay. Gen. Comp. Endocrinol. 62: 452-458. Wendelaar Bonga, S.E. and Pang, P.K.T. 1986. Stannius corpuscles. In Vertebrate Endocrinology, Fundamentals and Biomedical Implications. Vol. 1. pp. 439-464. Edited by P.K.T. Pang and M.P. Schreibman. Academic Press, New York.

Studies on the structure and physiology of salmon teleocalcin.

The structure and physiology of salmon teleocalcin, a Ca(+2) regulating hormone from the corpuscles of Stannius (CS) is reviewed. Teleocalcin is produ...
626KB Sizes 0 Downloads 0 Views