Fish Physiology and Biochemistry vol. 7 nos 1-4 pp 315-321 (1989) Kugler Publications, Amsterdam/Berkeley
Survival of salmonids in seawater and the time-frame of growth hormone action Nathan L. Collie, ' 2 Jonathan P. Bolton, Hiroshi Kawauchi,3 and Tetsuya Hirano' 1 Laboratory of Physiology, Ocean Research Institute, University of Tokyo, Nakano, Tokyo 164, Japan; 2 Department of Physiology, University of California at Los Angeles, School of Medicine, Los Angeles, CA 90024, U.S.A.; 3 Laboratory of Molecular Endocrinology, School of Fisheries Sciences, Kitasato University, Sanriku, Iwate 022-01, Japan Keywords: growth hormone, seawater adaptation, rainbow trout, osmoregulation, plasma ions, GH bioassay
Abstract In salmonids, growth hormone (GH) effectively promotes adaptation of freshwater (FW) fish to seawater (SW), but it has been unclear whether GH has osmoregulatory actions apart from those consequent to an increase in body size. Our objectives were first, to examine the minimum time and dose required for GH to enhance SW adaptation; and second, to optimize the conditions for the acute GH response in developing a convenient GH bioassay based on its plasma ion lowering effect. Trout showed markedly improved SW survival when transferred from fresh water 6, 24, or 48h after a single chum salmon GH injection (0.25 g/g). Preadapting trout to 1/3 SW enhanced the plasma ion lowering effect of ovine GH (oGH) injected 48h before transfer of the fish to 80% SW. Endogenous plasma GH levels were elevated in control trout switched from low salinities to 80% SW but were depressed in oGH-injected fish after transfer. Under optimal test conditions (1/3 SW preadaptation, 48h pre-transfer injection, and 100% SW final challenge), the reduction in plasma Na+, Ca + +, and Mg + + levels of oGH-injected fish was dose-dependent. The oGH doses giving minimum and maximum responses were 50 and 200 ng/g, respectively. In short, GH exerts acute osmoregulatory actions that promote SW adaptation in the absence of changes in body size. Compared with growth GH bioassays, the osmoregulatory assay is superior in economy of time, animal costs, and hormone quantity required and potentially in specificity.
Introduction In the past two decades, several investigations have implicated growth hormone (GH) in the regulation of seawater (SW) adaptation in salmonids along with its unambiguous control of growth rate (Hirano et al. 1987). It is perhaps not surprising to find that GH has a role in hydromineral balance, for two reasons. First, GH is a protein with close structural homology to prolactin, the latter hor-
mone having known actions on ion-transporting epithelia that enhance adaptation of euryhaline fish to fresh water (Loretz and Bern 1982; Hirano 1986). What is surprising is that GH promotes seawater survival, suggesting that some unique and, as yet, unknown actions exist quite apart from those of prolactin. Nevertheless, one might have expected a second, indirect role for GH in fish undergoing adaptation to either osmotic environment simply because GH stimulates increased body size. In
Correspondenceto: Nathan L. Collie, Laboratory of Physiology, Ocean Research Institute, University of Tokyo, Nakano, Tokyo 164, Japan.
316 salmonids size is strongly correlated with SW survival (Parry 1958; Conte and Wagner 1965; Wagner et al. 1969; McCormick and Naiman 1984), although the mechanisms underlying this correlation are poorly understood. Here, however, we address the issue of GH actions on SW adaptation that are independent of effects on body size. We recently reported that both chum salmon (Oncorhynchus keta) GH (sGH) and ovine GH (oGH), when given as three injections over a oneweek period, reduced plasma ion levels in freshwater-adapted (FW) rainbow trout following transfer to 80% SW for 24h (Bolton et al. 1987). The present experiments were designed to test by how much we could reduce the time interval allowed for a single GH injection to act and still retain its efficacy. Furthermore, since GH bioassays have been traditionally based on growth-stimulating activity, which are costly in time and in hormone amounts required, we sought to optimize the conditions for a bioassay based on the rapid osmoregulatory effects of GH.
Materials and methods
Hormones Chum salmon (0. keta) GH was purified as previously described (Kawauchi et al. 1986); oGH (NIADDK-oGH-S12) was generously provided by the NIADDK (NIH, Bethesda, MD, U.S.A.). The hormones were dissolved in a minimum volume of either 0.1 N NaOH (sGH) or distilled water (oGH) and diluted to the appropriate concentrations with salmon Ringer (Collie and Bern 1982) containing 0.1 %o(w/v) bovine serum albumin (BSA). At injection, the fish were anesthetized (0.05% 2-phenoxyethanol) and given the stated hormone doses or vehicle (Ringer plus 0.1% BSA) intraperitoneally in a volume of 5 l/g fish.
Experiment 1: Effects of sGH on SW survival Fish acclimated to FW (> 14 d) were divided into four groups (n = 12), injected once with sGH (0.25 /g/g fish) or vehicle, and transferred to 100%o SW 6, 24, or 48h after the injection. After 24h in SW, the fish remaining alive were anesthetized, blood was removed from the caudal vessels with 23-gauge needles, and the plasma stored frozen at -80°C.
Fish Juvenile rainbow trout (Salmo gairdneri),weighing 10-40g, were obtained by courtesy of Samegai Trout Hatchery, Shiga, Japan and maintained in a recirculating FW aquarium at 15°C. The fish were fed an artificial diet (Masu No. 4P; Nihonhaigo Shiryo Co., Tokyo, Japan) at a level of 1.5% of body weight per day. Feeding was stopped three days before transfer of fish from low (FW or 1/3 SW) to high (80 or 100% SW) environmental salinities. The pertinent electrolyte compositions of these salinities were (Na +, Ca++, and Mg ++, respectively, in mM/l): 0.4, 0.4, and 0.1, FW; 152, 5.2, and 16.9, 1/3 SW; 369, 9.2, and 37.0, 80% SW; 414, 11.5, and 45.9, 100% SW). Fish were assigned to groups purposely to ensure the most narrow range in body weights among groups possible. The range in mean body weight for all groups was 25.3-28.6g.
Experiment 2: Effects ofpreadaptationsalinity and oGH On Day 0, four groups of FW fish (n = 12-14) were preadapted to 1/3 SW and four groups were kept in FW. On Day 5, all groups received injections of vehicle or one of two oGH doses (0.25 or 2.5 zg/g fish). Two vehicle-injected control groups from the two preadaptation salinities were transferred on Day 7 to tanks containing either the salinity of preadaptation (FW or 1/3 SW, transfer controls) or 80% SW. Also on Day 7 the two oGH dose groups from each preadaptation salinity were transferred to 80% SW. All fish were sampled after 24h in their respective "challenge" salinities. Plasma was collected for analysis of ions and endogenous GH levels.
317 Experiment 3: Effects of different doses of oGH Fish were divided into six groups (n = 12), preadapted to 1/3 SW, and injected with vehicle or one of five doses of oGH. All groups were challenged in 100% SW. The protocol timing for preadaptation, injection, and final transfer was the same as in Experiment 2. Blood samples were taken for analysis of plasma Na+ , Ca + + , and Mg+ + levels.
Analytical techniques Plasma ion total concentrations were determined by atomic absorption spectrophotometry (Hitachi 180-50). Plasma GH levels were measured using the sGH RIA described and validated previously for trout plasma and pituitary GH (Bolton et al. 1986). Because the RIA (antiserum AS9-2) shows no cross-reactivity with oGH, plasma GH values of trout injected with oGH represent endogenous levels of trout GH. Statistical differences among groups were assessed using one-way analysis of variance followed by the F-test (orthogonal comparisons, Experiments 1 and 3) or by the Tukey-Kramer multiple comparison test (Experiment 2) (Sokal and Rohlf 1981).
Results Seawater survival and the time requirementfor GH action In experiment 1, the minimum time required for GH to produce a demonstrable change in seawater survival was investigated. In controls transferred from FW to SW 24h after vehicle injection, only 50% survived for 24h. A single sGH injection (0.25 /zg/g fish), however, increased the survival rate when given 6h (80%70 survival), 24h (100%), or 48h (100%) prior to transfer. Because one-half of the control fish died (presumably owing to excessive plasma ion concentration), no valid comparison of plasma ion levels was possible between control and sGH-injected trout.
Effects of preadaptation salinity on the osmoregulatory response to GH Experiment 2 was designed to lessen the severity of the SW challenge conditions to permit comparison of plasma ion levels in control and GH-injected fish while retaining the acute time course of GH action. Two strategies were employed for that purpose: (1) a short preadaptation period of FW trout to 1/3 SW before the final transfer to high salinity; and (2) a reduction in the final challenge salinity from 100 to 80% SW. Ovine GH was injected instead of the sGH used in Experiment 1, to conserve the latter and because our previous results indicated that oGH was approximately equipotent to sGH in producing rapid, size-independent effects on plasma ions (Bolton et al. 19874L Based on the survival data described above, 48h was selected as the time of injection preceding transfer of the fish to 80% SW that would lower plasma ions maximally. Under these conditions, survival proved to be 100% in all groups. Figure 1A-C illustrates the plasma ion response to oGH. Whereas oGH had no significant effect on plasma Na+ levels in trout transferred directly from FW to 80% SW, both oGH doses produced a significant fall in plasma Na+ when fish were first preadapted to 1/3 SW and subsequently transferred to 80% SW (Fig. 1A). A similar effect of GH in 1/3 SW preadapted fish was seen for plasma Ca+ + and Mg + + (Fig. 1B, C), although the dose eliciting the greatest reduction in ion levels was different for the two ions (2.5 and 0.25 /g/g, respectively). Preadaptation to 1/3 SW per se did not significantly lower posttransfer levels of plasma Na+ and Ca + + (but did lower Mg++ levels) since vehicle-injected controls showed similar high levels for these ions whether transferred to 80% SW from FW or from 1/3 SW. Endogenous GH levels for fish used in Experiment 2 are shown in Fig. 1D. Transferring vehicleinjected trout from FW or 1/3 SW to 80% SW elevated plasma GH levels to comparably high values, though the change was significant only in the latter group (1/3 SW to 80% SW). In marked contrast, oGH-treated fish exhibited depressed
318
200
FW
FW--80% W
a
133%: 33%-380°A W d,e SW
-IJ 180 N
Z
E
4-
a z Z
160 140
: : UI
3.0
.w-
E C-
r
Lij '9
'
+0
I
endogenous plasma GH compared with controls. Again, the pattern of reduction in plasma levels was different depending on the preadaptation salinity. Only the high dose of oGH (2.5 tg/g) caused significant depression of plasma GH in FW fish transferred to 80°7o SW, whereas the low (0.25 j/g/g) as well as the high oGH dose equally suppressed endogenous GH in trout preadapted to 1/3 SW before the transfer.
2.5
* a,b T T.
,2.0
E 1.0
z: 80 60 40 0
20 0 nA
T.
I
'
I
,
f
I .
I
I
! I
ii
bc
'I
I
.:.:
*'
fi
*
-a-
II I I
. a
'b,c,d d
'a
1
'
B
,.-. T
_
II T._
T
I
II
I
a
I.
Ovine GH dose-response of plasma ions under optimized SW challenge conditions
I
,PI
-J N3
E
T
I
Controls 0o.25 2.5 I
oGH (pg/g)
Controls 0.25 2.5 oGH(pg/g)
Fig. 1. The effect of preadaptation to 33% SW on the plasma ion (A-C) and endogenous GH (D) response to oGH of trout transferred to 80% SW. Vertical dotted and dashed lines divide A-D into four columns corresponding to the transfer groups labeled at the top of A. Vehicle-injected controls were transferred from FW or 33% SW to either the same, respective salinity (open bars) or to 80% SW (closed bars). Trout were injected with oGH (0.25 s/g/g, hatched bars or 2.5 itg/g, stipled) 48h before transfer to 80% SW and sacrificed 24h following transfer. Plasma GH represents endogenous levels measured using an RIA for salmonid GH that shows no cross-reaction with oGH. Note that preadapting trout to 33% SW, but not to FW, results in significant reductions in plasma ions and endogenous GH when injected with oGH. Bars having superscripts in common are significantly different at p < 0.01 except in C where d = p < 0.05; n = 10-14 for each group.
Experiment 3 was designed to expand and define the plasma ion response range to oGH treatment by increasing the final challenge salinity to 100% SW and by testing several oGH doses (0.05 to 2.5 /tg/g fish). Figure 2 shows the change in plasma ions of trout (transferred to 100% SW for 24h from 1/3 SW) in response to a single injection of oGH given 48h preceding the challenge. No mortalities were encountered during the 24h challenge in 100% SW, suggesting that preadaptation to 1/3 SW offers a protective advantage over direct transfer of fish from FW to full-strength SW. All three ions exhibited a reduction in plasma level that was linearly related to the log dose of oGH up to approximately 0.2 ttg/g fish. Higher doses yielded generally smaller decreases in ion levels (Na+ and Mg + +) from those of controls than obtained with the 0.2 tg/g dose or else yielded no change (Ca + +).
Discussion SW survival and the time-lag GH action Under conditions which minimized the effect of GH on body size, a single sGH injection (0.25 tg/g fish) improved the survival of fish in SW, when given 6-48h before their transfer from FW to SW. Komourdjian et al. (1976) also reported that porcine GH injections in FW Atlantic salmon (Salmo salar) transferred to SW resulted in no mortalities compared with 50% in controls. However, the Atlantic salmon received multiple injections (at 5 x
319 groups in Experiment 2 was anticipated. An unexpected finding was the apparent increase in sensitivity to oGH in the 1/3 SW preadapted groups (Fig. lA-C), compared with fish transferred directly from FW to 80% SW, which were essentially unresponsive to the single oGH injection. By contrast, trout given three oGH injections over a one-week period and transferred directly from FW to 80% SW did show significantly lower plasma ion concentration compared to control levels (Bolton et al. 1987). These results together suggest that multiple injections or preadaptation to 1/3 SW represent alternative ways of "priming" the acute osmoregulatory response to GH. Iwata and Komatsu (1984) have pointed out the importance of estuarine residence for the rapid adaptation of chum salmon (0. keta) fry to SW. Perhaps the adaptive effects of a short residence in moderate salinities result in part from changes in responsiveness of salmonids to GH.
_J -j
E u)
0) c C
H o
E
U)
a.L
Control
.05 .
.2
.5
2.5
oGH (pg /g) Fig. 2. Log dose-response relationship between oGH and reduction in plasma Na + (A), Ca + + and Mg + + levels (B) (x + SEM; n = 12, each group). Rainbow trout were given a single injection of oGH 48h before transfer to 100% SW for an additional 24h. a, p < 0.05; b, p < 0.01; c, p < 0.001; compared with vehicle-injected controls.
the dose/injection used in Experiment 1) for 28d, which significantly stimulated growth. Thus, it was difficult to separate the GH effects on body size from those on SW adaptability alone. Though the survival data are compelling, the osmoregulatory basis for GH's effects is obscured by the very fact of poor survival in controls compared with GHinjected fish. Preadaptationto 1/3 SW increases responsiveness to oGH Our previous work (Bolton et al. 1987) and that of Landless (1976) had suggested that reducing the salinity of SW from 100% to 80% or less would permit survival of even small rainbow trout transferred from FW. Hence, the uniform survival of all
Ovine GH injections reduce endogenous plasma GH levels Analysis of endogenous plasma GH levels of trout in Experiment 2 revealed two points. First, untreated fish transferred from either FW or 1/3 SW to 80% SW showed elevated plasma GH, a finding consistent with the GH increase seen in other salmonids transferred from FW to SW (Sweeting et al. 1985; Hasegawa et al. 1987; Hirano et al. 1987). Second, oGH treatment reduced plasma GH levels in fish transferred to 80% SW from either FW or 1/3 SW, but more potently so in the latter group (Fig. ID). This observation raises some intriguing questions about the mechanisms regulating plasma GH levels. At least two possibilities are evident: oGH may (1) inhibit pituitary GH secretion, acting through a variety of potential feedback loops; and (2) influence plasma clearance of endogenous GH. More experiments are needed to define the factors regulating plasma GH levels during SW adaptation. The acute plasma ion responseto GH as a bioassay Testing a range of oGH doses under the combined, optimal conditions from Experiments 1 and 2 (1/3
320 SW preadaptation, 48h pre-transfer injection, and 100% SW final challenge) resulted in falls in plasma ions that were dose-dependent (Fig. 2A, B). These conditions offered one important improvement over our previous study (Bolton et al. 1987): only one injection 48h prior to the SW challenge was required for the response rather than three injections over a one-week period. Direct comparison of the response sensitivity to dose in the two studies is not strictly possible because we used oGH in these optimization experiments instead of the sGH employed previously. However, the minimum doseper injection giving a significant reduction in plasma Na + , Ca++, and Mg + + was, respectively, 5-, 25-, and 2.5-fold lower in the current experiment. The advantages of a GH bioassay based on its plasma ion lowering ability are many. Compared with growth bioassays, the osmoregulatory assay for GH requires far less time, uses less hormone, and offers potentially better specificity in distinguishing GH from prolactin. Regarding specificity, prolactin as well as GH stimulate growth in sockeye salmon (Clarke et al. 1977), rainbow trout (Kawauchi et al. 1986 and unpublished observations), and tilapia (Specker et al. 1985); in contrast, GH and prolactin have opposite effects on plasma ions (Bolton et al. 1987). Further studies, using the current protocol are needed, however, to examine the assay response to GH and prolactin isolated from several species. Finally, the GH bioassay conditions described here require only the ability to measure plasma ions, and utilize rainbow trout, a species widely available.
Enhancement of SW survival in two time-frames A synthesis of GH effects on growth and SW adaptation leads to the following conclusions. GH is a potent growth-enhancer in fish (see Donaldson et al. 1979) and, hence, clearly regulates body size, an important determinant of survival in SW. During ontogeny the chronic effects of GH on growth operate over an extended time-frame to increase body size and enhance SW adaptability. A second time-frame of GH action involves acute effects independent of body size. Accordingly, GH may pro-
mote survival during the "adjustive phase" of SW adaptation in the first few hours or days after SW entry by preventing the initial rise of plasma ion levels above lethal limits and by returning those levels to steady state values. The precise mechanisms operating in either time-frame are currently unknown (for discussion, see Bolton et al. 1987; Collie and Hirano 1987) and represent a fascinating area for future studies.
Acknowledgements We express our thanks to Dr. T. Ogasawara and Ms. S. Hasegawa of the Ocean Research Institute, University of Tokyo for their help and advice throughout the study, Dr. S. Raiti of the National Pituitary Agency, Bethesda for the oGH, and Drs. S. Fushiki and Y. Fujioka, Samegai Trout Hatchery, Shiga for provision of fish. Financial support for the research was provided by NIH NRSA (1 F32 AM07221-01) and JSPS fellowships to N.L.C., by a Royal Society/JSPS Fellowship to J.P.B., and by Grants-in-Aid for Scientific Research from the Ministry of Education, Science, and Culture to H.K. and T.H.
References cited Bolton, J.P., Takahashi, A., Kawauchi, H., Kubota, J. and Hirano, T. 1986. Development and validation of a salmon growth hormone radioimmunoassay. Gen. Comp. Endocrinol. 62: 230-238. Bolton, J.P., Collie, N.L., Kawauchi, H. and Hirano, T. 1987. Osmoregulatory actions of growth hormone in rainbow trout (Salmo gairdneri). J. Endocrinol. 112: 63-68. Clarke, W.C., Farmer, S.W. and Hartwell, K.M. 1977. Effect of teleost pituitary growth hormone on growth of Tilapia mossambica and on growth and sea water adaptation of sockeye salmon (Oncorhynchus nerka). Gen. Comp. Endocrinol. 33: 174-178. Collie, N.L. and Bern, H.A. 1982. Changes in intestinal fluid transport associated with smoltification and seawater adaptation in coho salmon, Oncorhynchus kisutch (Walbaum). J. Fish Biol. 21: 337-348. Collie, N.L. and Hirano, T. 1987. Mechanisms of hormone actions on intestinal transport. In Vertebrate Endocrinology: Fundamentals and Biomedical Implications. Vol. 2. pp. 239-270. Edited by P.K.T. Pang and M.P. Schreibman. Academic Press, New York.
321 Conte, F.P. and Wagner, H.H. 1965. Development of osmotic and ionic regulation in juvenile steelhead trout (Salmo gairdneri). Comp. Physiol. Biochem. 14: 603-620. Donaldson, E.M., Fagerlund, U.H.M., Higgs, D.A. and McBride, J.R. 1979. Hormonal enhancement of growth in fish. In Fish Physiology. Vol. VIII. pp. 456-597. Edited by W.S. Hoar, D.J. Randall and J.R. Brett. Academic Press, New York. Hasegawa, S., Hirano, T., Ogasawara, T., Iwata, M., Akiyama, T. and Arai, S. 1987. Osmoregulatory ability of chum salmon, Oncorhynchu keta, reared in fresh water for prolonged periods. Fish Physiol. Biochem. 4: 101-110. Hirano, T. 1986. The spectrum of prolactin action in teleosts. In Comparative Endocrinology: Developments and Directions. pp. 53-74. Edited by C.L. Ralph. Alan R. Liss, New York. Hirano, T., Ogasawara, T., Bolton, J.P., Collie, N.L., Hasegawa, S. and Iwata, M. 1987. Osmoregulatory role of prolactin in lower vertebrates. In Comparative Physiology of Environmental Adaptations. Vol. 1. pp. 112-124. Edited by R. Kirsch and B. Lahlou. Karger, Basel. Iwata, M. and Komatsu, S. 1984. Importance of estuarine residence for adaptation of chum salmon (Oncorhynchus keta) fry to seawater. Can. J. Fish. Aqu. Sci. 41: 744-749. Kawauchi, H., Moriyama, S., Yasuda, A., Yamaguchi, K., Shirahata, K., Kato, J. and Hirano, T. 1986. Isolation and characterization of chum salmon growth hormone. Arch. Biochem. Biophys. 244: 542-552. Komourdjian, M.P., Saunders, R.L. and Fenwick, J.C. 1976. The effect of porcine somatotropin on growth and survival in
seawater of Atlantic salmon (Salmo salar). Can. J. Zool. 54: 531-535. Landless, P.J. 1976. Acclimation of rainbow trout to sea water. Aquaculture 7: 173-179. Loretz, C.A. and Bern, H.A. 1982. Prolactin and osmoregulation in vertebrates. Neuroendocrinology 35: 292-304. McCormick, S.D. and Naiman, R.J. 1984. Osmoregulation in the brook trout, Salvelinusfontinalis: II. Effects of size, age and photoperiod on seawater survival and ionic regulation. Comp. Biochem. Physiol. 79A: 17-28. Parry, G. 1958. Size and osmoregulation in salmonid fishes. Nature, Lond. 181: 1218-1219. Sokal, R.R. and Rohlf, F.J. 1981. Biometry. W.H. Freeman, San Francisco. Specker, J.L., King, D.S., Rivas, R.J. and Young, B.K. 1985. Partial characterization of two prolactins from a cichlid fish. In Prolactin. Basic and Clinical Correlates. pp. 427-435. Edited by R.M. MacLeod, M.O. Thorner and U. Scapagnini. Liviana Press, Padova. Sweeting, R.M., Wagner, G.F. and McKeown, B.A. 1985. Changes in plasma glucose, amino acid nitrogen and growth hormone during smoltification and seawater adaptation in coho salmon, Oncorhynchus kisutch. Aquaculture 45: 185-197. Wagner, H.H., Conte, F.P. and Fesler, J.L. 1969. Development of osmotic and ionic regulation in two races of chinook salmon, (Oncorhynchus tshawytscha). Comp. Biochem. Physiol. 29: 325-341.
Survival of salmonids in seawater and the time-frame of growth hormone action Nathan L. Collie, ' 2 Jonathan P. Bolton, Hiroshi Kawauchi,3 and Tetsuya Hirano' 1 Laboratory of Physiology, Ocean Research Institute, University of Tokyo, Nakano, Tokyo 164, Japan; 2 Department of Physiology, University of California at Los Angeles, School of Medicine, Los Angeles, CA 90024, U.S.A.; 3 Laboratory of Molecular Endocrinology, School of Fisheries Sciences, Kitasato University, Sanriku, Iwate 022-01, Japan Keywords: growth hormone, seawater adaptation, rainbow trout, osmoregulation, plasma ions, GH bioassay
Abstract In salmonids, growth hormone (GH) effectively promotes adaptation of freshwater (FW) fish to seawater (SW), but it has been unclear whether GH has osmoregulatory actions apart from those consequent to an increase in body size. Our objectives were first, to examine the minimum time and dose required for GH to enhance SW adaptation; and second, to optimize the conditions for the acute GH response in developing a convenient GH bioassay based on its plasma ion lowering effect. Trout showed markedly improved SW survival when transferred from fresh water 6, 24, or 48h after a single chum salmon GH injection (0.25 g/g). Preadapting trout to 1/3 SW enhanced the plasma ion lowering effect of ovine GH (oGH) injected 48h before transfer of the fish to 80% SW. Endogenous plasma GH levels were elevated in control trout switched from low salinities to 80% SW but were depressed in oGH-injected fish after transfer. Under optimal test conditions (1/3 SW preadaptation, 48h pre-transfer injection, and 100% SW final challenge), the reduction in plasma Na+, Ca + +, and Mg + + levels of oGH-injected fish was dose-dependent. The oGH doses giving minimum and maximum responses were 50 and 200 ng/g, respectively. In short, GH exerts acute osmoregulatory actions that promote SW adaptation in the absence of changes in body size. Compared with growth GH bioassays, the osmoregulatory assay is superior in economy of time, animal costs, and hormone quantity required and potentially in specificity.
Introduction In the past two decades, several investigations have implicated growth hormone (GH) in the regulation of seawater (SW) adaptation in salmonids along with its unambiguous control of growth rate (Hirano et al. 1987). It is perhaps not surprising to find that GH has a role in hydromineral balance, for two reasons. First, GH is a protein with close structural homology to prolactin, the latter hor-
mone having known actions on ion-transporting epithelia that enhance adaptation of euryhaline fish to fresh water (Loretz and Bern 1982; Hirano 1986). What is surprising is that GH promotes seawater survival, suggesting that some unique and, as yet, unknown actions exist quite apart from those of prolactin. Nevertheless, one might have expected a second, indirect role for GH in fish undergoing adaptation to either osmotic environment simply because GH stimulates increased body size. In
Correspondenceto: Nathan L. Collie, Laboratory of Physiology, Ocean Research Institute, University of Tokyo, Nakano, Tokyo 164, Japan.
316 salmonids size is strongly correlated with SW survival (Parry 1958; Conte and Wagner 1965; Wagner et al. 1969; McCormick and Naiman 1984), although the mechanisms underlying this correlation are poorly understood. Here, however, we address the issue of GH actions on SW adaptation that are independent of effects on body size. We recently reported that both chum salmon (Oncorhynchus keta) GH (sGH) and ovine GH (oGH), when given as three injections over a oneweek period, reduced plasma ion levels in freshwater-adapted (FW) rainbow trout following transfer to 80% SW for 24h (Bolton et al. 1987). The present experiments were designed to test by how much we could reduce the time interval allowed for a single GH injection to act and still retain its efficacy. Furthermore, since GH bioassays have been traditionally based on growth-stimulating activity, which are costly in time and in hormone amounts required, we sought to optimize the conditions for a bioassay based on the rapid osmoregulatory effects of GH.
Materials and methods
Hormones Chum salmon (0. keta) GH was purified as previously described (Kawauchi et al. 1986); oGH (NIADDK-oGH-S12) was generously provided by the NIADDK (NIH, Bethesda, MD, U.S.A.). The hormones were dissolved in a minimum volume of either 0.1 N NaOH (sGH) or distilled water (oGH) and diluted to the appropriate concentrations with salmon Ringer (Collie and Bern 1982) containing 0.1 %o(w/v) bovine serum albumin (BSA). At injection, the fish were anesthetized (0.05% 2-phenoxyethanol) and given the stated hormone doses or vehicle (Ringer plus 0.1% BSA) intraperitoneally in a volume of 5 l/g fish.
Experiment 1: Effects of sGH on SW survival Fish acclimated to FW (> 14 d) were divided into four groups (n = 12), injected once with sGH (0.25 /g/g fish) or vehicle, and transferred to 100%o SW 6, 24, or 48h after the injection. After 24h in SW, the fish remaining alive were anesthetized, blood was removed from the caudal vessels with 23-gauge needles, and the plasma stored frozen at -80°C.
Fish Juvenile rainbow trout (Salmo gairdneri),weighing 10-40g, were obtained by courtesy of Samegai Trout Hatchery, Shiga, Japan and maintained in a recirculating FW aquarium at 15°C. The fish were fed an artificial diet (Masu No. 4P; Nihonhaigo Shiryo Co., Tokyo, Japan) at a level of 1.5% of body weight per day. Feeding was stopped three days before transfer of fish from low (FW or 1/3 SW) to high (80 or 100% SW) environmental salinities. The pertinent electrolyte compositions of these salinities were (Na +, Ca++, and Mg ++, respectively, in mM/l): 0.4, 0.4, and 0.1, FW; 152, 5.2, and 16.9, 1/3 SW; 369, 9.2, and 37.0, 80% SW; 414, 11.5, and 45.9, 100% SW). Fish were assigned to groups purposely to ensure the most narrow range in body weights among groups possible. The range in mean body weight for all groups was 25.3-28.6g.
Experiment 2: Effects ofpreadaptationsalinity and oGH On Day 0, four groups of FW fish (n = 12-14) were preadapted to 1/3 SW and four groups were kept in FW. On Day 5, all groups received injections of vehicle or one of two oGH doses (0.25 or 2.5 zg/g fish). Two vehicle-injected control groups from the two preadaptation salinities were transferred on Day 7 to tanks containing either the salinity of preadaptation (FW or 1/3 SW, transfer controls) or 80% SW. Also on Day 7 the two oGH dose groups from each preadaptation salinity were transferred to 80% SW. All fish were sampled after 24h in their respective "challenge" salinities. Plasma was collected for analysis of ions and endogenous GH levels.
317 Experiment 3: Effects of different doses of oGH Fish were divided into six groups (n = 12), preadapted to 1/3 SW, and injected with vehicle or one of five doses of oGH. All groups were challenged in 100% SW. The protocol timing for preadaptation, injection, and final transfer was the same as in Experiment 2. Blood samples were taken for analysis of plasma Na+ , Ca + + , and Mg+ + levels.
Analytical techniques Plasma ion total concentrations were determined by atomic absorption spectrophotometry (Hitachi 180-50). Plasma GH levels were measured using the sGH RIA described and validated previously for trout plasma and pituitary GH (Bolton et al. 1986). Because the RIA (antiserum AS9-2) shows no cross-reactivity with oGH, plasma GH values of trout injected with oGH represent endogenous levels of trout GH. Statistical differences among groups were assessed using one-way analysis of variance followed by the F-test (orthogonal comparisons, Experiments 1 and 3) or by the Tukey-Kramer multiple comparison test (Experiment 2) (Sokal and Rohlf 1981).
Results Seawater survival and the time requirementfor GH action In experiment 1, the minimum time required for GH to produce a demonstrable change in seawater survival was investigated. In controls transferred from FW to SW 24h after vehicle injection, only 50% survived for 24h. A single sGH injection (0.25 /zg/g fish), however, increased the survival rate when given 6h (80%70 survival), 24h (100%), or 48h (100%) prior to transfer. Because one-half of the control fish died (presumably owing to excessive plasma ion concentration), no valid comparison of plasma ion levels was possible between control and sGH-injected trout.
Effects of preadaptation salinity on the osmoregulatory response to GH Experiment 2 was designed to lessen the severity of the SW challenge conditions to permit comparison of plasma ion levels in control and GH-injected fish while retaining the acute time course of GH action. Two strategies were employed for that purpose: (1) a short preadaptation period of FW trout to 1/3 SW before the final transfer to high salinity; and (2) a reduction in the final challenge salinity from 100 to 80% SW. Ovine GH was injected instead of the sGH used in Experiment 1, to conserve the latter and because our previous results indicated that oGH was approximately equipotent to sGH in producing rapid, size-independent effects on plasma ions (Bolton et al. 19874L Based on the survival data described above, 48h was selected as the time of injection preceding transfer of the fish to 80% SW that would lower plasma ions maximally. Under these conditions, survival proved to be 100% in all groups. Figure 1A-C illustrates the plasma ion response to oGH. Whereas oGH had no significant effect on plasma Na+ levels in trout transferred directly from FW to 80% SW, both oGH doses produced a significant fall in plasma Na+ when fish were first preadapted to 1/3 SW and subsequently transferred to 80% SW (Fig. 1A). A similar effect of GH in 1/3 SW preadapted fish was seen for plasma Ca+ + and Mg + + (Fig. 1B, C), although the dose eliciting the greatest reduction in ion levels was different for the two ions (2.5 and 0.25 /g/g, respectively). Preadaptation to 1/3 SW per se did not significantly lower posttransfer levels of plasma Na+ and Ca + + (but did lower Mg++ levels) since vehicle-injected controls showed similar high levels for these ions whether transferred to 80% SW from FW or from 1/3 SW. Endogenous GH levels for fish used in Experiment 2 are shown in Fig. 1D. Transferring vehicleinjected trout from FW or 1/3 SW to 80% SW elevated plasma GH levels to comparably high values, though the change was significant only in the latter group (1/3 SW to 80% SW). In marked contrast, oGH-treated fish exhibited depressed
318
200
FW
FW--80% W
a
133%: 33%-380°A W d,e SW
-IJ 180 N
Z
E
4-
a z Z
160 140
: : UI
3.0
.w-
E C-
r
Lij '9
'
+0
I
endogenous plasma GH compared with controls. Again, the pattern of reduction in plasma levels was different depending on the preadaptation salinity. Only the high dose of oGH (2.5 tg/g) caused significant depression of plasma GH in FW fish transferred to 80°7o SW, whereas the low (0.25 j/g/g) as well as the high oGH dose equally suppressed endogenous GH in trout preadapted to 1/3 SW before the transfer.
2.5
* a,b T T.
,2.0
E 1.0
z: 80 60 40 0
20 0 nA
T.
I
'
I
,
f
I .
I
I
! I
ii
bc
'I
I
.:.:
*'
fi
*
-a-
II I I
. a
'b,c,d d
'a
1
'
B
,.-. T
_
II T._
T
I
II
I
a
I.
Ovine GH dose-response of plasma ions under optimized SW challenge conditions
I
,PI
-J N3
E
T
I
Controls 0o.25 2.5 I
oGH (pg/g)
Controls 0.25 2.5 oGH(pg/g)
Fig. 1. The effect of preadaptation to 33% SW on the plasma ion (A-C) and endogenous GH (D) response to oGH of trout transferred to 80% SW. Vertical dotted and dashed lines divide A-D into four columns corresponding to the transfer groups labeled at the top of A. Vehicle-injected controls were transferred from FW or 33% SW to either the same, respective salinity (open bars) or to 80% SW (closed bars). Trout were injected with oGH (0.25 s/g/g, hatched bars or 2.5 itg/g, stipled) 48h before transfer to 80% SW and sacrificed 24h following transfer. Plasma GH represents endogenous levels measured using an RIA for salmonid GH that shows no cross-reaction with oGH. Note that preadapting trout to 33% SW, but not to FW, results in significant reductions in plasma ions and endogenous GH when injected with oGH. Bars having superscripts in common are significantly different at p < 0.01 except in C where d = p < 0.05; n = 10-14 for each group.
Experiment 3 was designed to expand and define the plasma ion response range to oGH treatment by increasing the final challenge salinity to 100% SW and by testing several oGH doses (0.05 to 2.5 /tg/g fish). Figure 2 shows the change in plasma ions of trout (transferred to 100% SW for 24h from 1/3 SW) in response to a single injection of oGH given 48h preceding the challenge. No mortalities were encountered during the 24h challenge in 100% SW, suggesting that preadaptation to 1/3 SW offers a protective advantage over direct transfer of fish from FW to full-strength SW. All three ions exhibited a reduction in plasma level that was linearly related to the log dose of oGH up to approximately 0.2 ttg/g fish. Higher doses yielded generally smaller decreases in ion levels (Na+ and Mg + +) from those of controls than obtained with the 0.2 tg/g dose or else yielded no change (Ca + +).
Discussion SW survival and the time-lag GH action Under conditions which minimized the effect of GH on body size, a single sGH injection (0.25 tg/g fish) improved the survival of fish in SW, when given 6-48h before their transfer from FW to SW. Komourdjian et al. (1976) also reported that porcine GH injections in FW Atlantic salmon (Salmo salar) transferred to SW resulted in no mortalities compared with 50% in controls. However, the Atlantic salmon received multiple injections (at 5 x
319 groups in Experiment 2 was anticipated. An unexpected finding was the apparent increase in sensitivity to oGH in the 1/3 SW preadapted groups (Fig. lA-C), compared with fish transferred directly from FW to 80% SW, which were essentially unresponsive to the single oGH injection. By contrast, trout given three oGH injections over a one-week period and transferred directly from FW to 80% SW did show significantly lower plasma ion concentration compared to control levels (Bolton et al. 1987). These results together suggest that multiple injections or preadaptation to 1/3 SW represent alternative ways of "priming" the acute osmoregulatory response to GH. Iwata and Komatsu (1984) have pointed out the importance of estuarine residence for the rapid adaptation of chum salmon (0. keta) fry to SW. Perhaps the adaptive effects of a short residence in moderate salinities result in part from changes in responsiveness of salmonids to GH.
_J -j
E u)
0) c C
H o
E
U)
a.L
Control
.05 .
.2
.5
2.5
oGH (pg /g) Fig. 2. Log dose-response relationship between oGH and reduction in plasma Na + (A), Ca + + and Mg + + levels (B) (x + SEM; n = 12, each group). Rainbow trout were given a single injection of oGH 48h before transfer to 100% SW for an additional 24h. a, p < 0.05; b, p < 0.01; c, p < 0.001; compared with vehicle-injected controls.
the dose/injection used in Experiment 1) for 28d, which significantly stimulated growth. Thus, it was difficult to separate the GH effects on body size from those on SW adaptability alone. Though the survival data are compelling, the osmoregulatory basis for GH's effects is obscured by the very fact of poor survival in controls compared with GHinjected fish. Preadaptationto 1/3 SW increases responsiveness to oGH Our previous work (Bolton et al. 1987) and that of Landless (1976) had suggested that reducing the salinity of SW from 100% to 80% or less would permit survival of even small rainbow trout transferred from FW. Hence, the uniform survival of all
Ovine GH injections reduce endogenous plasma GH levels Analysis of endogenous plasma GH levels of trout in Experiment 2 revealed two points. First, untreated fish transferred from either FW or 1/3 SW to 80% SW showed elevated plasma GH, a finding consistent with the GH increase seen in other salmonids transferred from FW to SW (Sweeting et al. 1985; Hasegawa et al. 1987; Hirano et al. 1987). Second, oGH treatment reduced plasma GH levels in fish transferred to 80% SW from either FW or 1/3 SW, but more potently so in the latter group (Fig. ID). This observation raises some intriguing questions about the mechanisms regulating plasma GH levels. At least two possibilities are evident: oGH may (1) inhibit pituitary GH secretion, acting through a variety of potential feedback loops; and (2) influence plasma clearance of endogenous GH. More experiments are needed to define the factors regulating plasma GH levels during SW adaptation. The acute plasma ion responseto GH as a bioassay Testing a range of oGH doses under the combined, optimal conditions from Experiments 1 and 2 (1/3
320 SW preadaptation, 48h pre-transfer injection, and 100% SW final challenge) resulted in falls in plasma ions that were dose-dependent (Fig. 2A, B). These conditions offered one important improvement over our previous study (Bolton et al. 1987): only one injection 48h prior to the SW challenge was required for the response rather than three injections over a one-week period. Direct comparison of the response sensitivity to dose in the two studies is not strictly possible because we used oGH in these optimization experiments instead of the sGH employed previously. However, the minimum doseper injection giving a significant reduction in plasma Na + , Ca++, and Mg + + was, respectively, 5-, 25-, and 2.5-fold lower in the current experiment. The advantages of a GH bioassay based on its plasma ion lowering ability are many. Compared with growth bioassays, the osmoregulatory assay for GH requires far less time, uses less hormone, and offers potentially better specificity in distinguishing GH from prolactin. Regarding specificity, prolactin as well as GH stimulate growth in sockeye salmon (Clarke et al. 1977), rainbow trout (Kawauchi et al. 1986 and unpublished observations), and tilapia (Specker et al. 1985); in contrast, GH and prolactin have opposite effects on plasma ions (Bolton et al. 1987). Further studies, using the current protocol are needed, however, to examine the assay response to GH and prolactin isolated from several species. Finally, the GH bioassay conditions described here require only the ability to measure plasma ions, and utilize rainbow trout, a species widely available.
Enhancement of SW survival in two time-frames A synthesis of GH effects on growth and SW adaptation leads to the following conclusions. GH is a potent growth-enhancer in fish (see Donaldson et al. 1979) and, hence, clearly regulates body size, an important determinant of survival in SW. During ontogeny the chronic effects of GH on growth operate over an extended time-frame to increase body size and enhance SW adaptability. A second time-frame of GH action involves acute effects independent of body size. Accordingly, GH may pro-
mote survival during the "adjustive phase" of SW adaptation in the first few hours or days after SW entry by preventing the initial rise of plasma ion levels above lethal limits and by returning those levels to steady state values. The precise mechanisms operating in either time-frame are currently unknown (for discussion, see Bolton et al. 1987; Collie and Hirano 1987) and represent a fascinating area for future studies.
Acknowledgements We express our thanks to Dr. T. Ogasawara and Ms. S. Hasegawa of the Ocean Research Institute, University of Tokyo for their help and advice throughout the study, Dr. S. Raiti of the National Pituitary Agency, Bethesda for the oGH, and Drs. S. Fushiki and Y. Fujioka, Samegai Trout Hatchery, Shiga for provision of fish. Financial support for the research was provided by NIH NRSA (1 F32 AM07221-01) and JSPS fellowships to N.L.C., by a Royal Society/JSPS Fellowship to J.P.B., and by Grants-in-Aid for Scientific Research from the Ministry of Education, Science, and Culture to H.K. and T.H.
References cited Bolton, J.P., Takahashi, A., Kawauchi, H., Kubota, J. and Hirano, T. 1986. Development and validation of a salmon growth hormone radioimmunoassay. Gen. Comp. Endocrinol. 62: 230-238. Bolton, J.P., Collie, N.L., Kawauchi, H. and Hirano, T. 1987. Osmoregulatory actions of growth hormone in rainbow trout (Salmo gairdneri). J. Endocrinol. 112: 63-68. Clarke, W.C., Farmer, S.W. and Hartwell, K.M. 1977. Effect of teleost pituitary growth hormone on growth of Tilapia mossambica and on growth and sea water adaptation of sockeye salmon (Oncorhynchus nerka). Gen. Comp. Endocrinol. 33: 174-178. Collie, N.L. and Bern, H.A. 1982. Changes in intestinal fluid transport associated with smoltification and seawater adaptation in coho salmon, Oncorhynchus kisutch (Walbaum). J. Fish Biol. 21: 337-348. Collie, N.L. and Hirano, T. 1987. Mechanisms of hormone actions on intestinal transport. In Vertebrate Endocrinology: Fundamentals and Biomedical Implications. Vol. 2. pp. 239-270. Edited by P.K.T. Pang and M.P. Schreibman. Academic Press, New York.
321 Conte, F.P. and Wagner, H.H. 1965. Development of osmotic and ionic regulation in juvenile steelhead trout (Salmo gairdneri). Comp. Physiol. Biochem. 14: 603-620. Donaldson, E.M., Fagerlund, U.H.M., Higgs, D.A. and McBride, J.R. 1979. Hormonal enhancement of growth in fish. In Fish Physiology. Vol. VIII. pp. 456-597. Edited by W.S. Hoar, D.J. Randall and J.R. Brett. Academic Press, New York. Hasegawa, S., Hirano, T., Ogasawara, T., Iwata, M., Akiyama, T. and Arai, S. 1987. Osmoregulatory ability of chum salmon, Oncorhynchu keta, reared in fresh water for prolonged periods. Fish Physiol. Biochem. 4: 101-110. Hirano, T. 1986. The spectrum of prolactin action in teleosts. In Comparative Endocrinology: Developments and Directions. pp. 53-74. Edited by C.L. Ralph. Alan R. Liss, New York. Hirano, T., Ogasawara, T., Bolton, J.P., Collie, N.L., Hasegawa, S. and Iwata, M. 1987. Osmoregulatory role of prolactin in lower vertebrates. In Comparative Physiology of Environmental Adaptations. Vol. 1. pp. 112-124. Edited by R. Kirsch and B. Lahlou. Karger, Basel. Iwata, M. and Komatsu, S. 1984. Importance of estuarine residence for adaptation of chum salmon (Oncorhynchus keta) fry to seawater. Can. J. Fish. Aqu. Sci. 41: 744-749. Kawauchi, H., Moriyama, S., Yasuda, A., Yamaguchi, K., Shirahata, K., Kato, J. and Hirano, T. 1986. Isolation and characterization of chum salmon growth hormone. Arch. Biochem. Biophys. 244: 542-552. Komourdjian, M.P., Saunders, R.L. and Fenwick, J.C. 1976. The effect of porcine somatotropin on growth and survival in
seawater of Atlantic salmon (Salmo salar). Can. J. Zool. 54: 531-535. Landless, P.J. 1976. Acclimation of rainbow trout to sea water. Aquaculture 7: 173-179. Loretz, C.A. and Bern, H.A. 1982. Prolactin and osmoregulation in vertebrates. Neuroendocrinology 35: 292-304. McCormick, S.D. and Naiman, R.J. 1984. Osmoregulation in the brook trout, Salvelinusfontinalis: II. Effects of size, age and photoperiod on seawater survival and ionic regulation. Comp. Biochem. Physiol. 79A: 17-28. Parry, G. 1958. Size and osmoregulation in salmonid fishes. Nature, Lond. 181: 1218-1219. Sokal, R.R. and Rohlf, F.J. 1981. Biometry. W.H. Freeman, San Francisco. Specker, J.L., King, D.S., Rivas, R.J. and Young, B.K. 1985. Partial characterization of two prolactins from a cichlid fish. In Prolactin. Basic and Clinical Correlates. pp. 427-435. Edited by R.M. MacLeod, M.O. Thorner and U. Scapagnini. Liviana Press, Padova. Sweeting, R.M., Wagner, G.F. and McKeown, B.A. 1985. Changes in plasma glucose, amino acid nitrogen and growth hormone during smoltification and seawater adaptation in coho salmon, Oncorhynchus kisutch. Aquaculture 45: 185-197. Wagner, H.H., Conte, F.P. and Fesler, J.L. 1969. Development of osmotic and ionic regulation in two races of chinook salmon, (Oncorhynchus tshawytscha). Comp. Biochem. Physiol. 29: 325-341.
