0013-7227/90/1272-0716$02.00/0 Endocrinology Copyright© 1990 by The Endocrine Society
Vol. 127, No. 2 Printed in U.S.A.
Effect of Food Withdrawal and Insulin on Growth Hormone Secretion in the Guinea Pig KEITH M. FAIRHALL, BRITT G. GABRIELSSON, AND IAIN C. A. F. ROBINSON Division of Neurophysiology and Neuropharmacology, National Institute for Medical Research, The Ridgeway Mill Hill, London, NW7 1AA England
ABSTRACT. The guinea pig is unusual in that its postnatal growth appears to be independent of GH even though its pituitary gland produces a GH molecule. The effects of fasting on the GH secretory pattern and the GH responses to insulin, GHreleasing factor (GRF), and somatostatin (SS) during fasting have now been studied by automatic microsampling of blood in chronically cannulated normal guinea pigs. Withdrawal of food in both male and female guinea pigs changed the GH secretory pattern dramatically. The normal episodic GH secretory pattern [large GH peaks occurring at 3.6 ± 0.4-h intervals over a low (~0.5-1.5 ng/ml) baseline secretion] was altered to a pattern of more continuous GH output, characterized by a 10-fold elevated baseline secretion (5-15 ng/ml) with no large secretory episodes or troughs. Glucose injections (three injections of 600 mg, iv, at hourly intervals) in fasted guinea pigs lowered their elevated
W
E RECENTLY reported the first characterization of the secretory pattern of GH in normal male and female guinea pigs, and showed that, like most other species, GH release was episodic and could be stimulated or inhibited by GH-releasing factor (GRF) or somatostatin (SS), respectively (1). Our interest in GH control in the guinea pig stems from the early observations that despite producing large amounts of a pituitary GH, they do not appear to need this hormone for postnatal growth, unlike other mammals (2-4). Since GH has important metabolic actions in many species (5), it occurred to us that the guinea pig may represent one extreme in which GH is primarily a metabolic hormone and not a growthpromoting agent. Although GH release is sensitive to insulin and nutrient supply, the responses differ radically between species. In man, GH release is stimulated acutely by insulin-induced hypoglycemia (6), and prolonged GH hypersecretion occurs during fasting in man and larger animals (7-10). By contrast, starvation, intracellular glucopenia, and diabetes in the rat all cause prolonged suppression of GH release (11-14), and insulin blocks episodic GH pulses (15). We were, therefore,
blood GH levels significantly (from 9.1 ± 1.1 to 6.5 ± 0.9 ng/ ml). Insulin injections (1, 2, or 6 U, iv) inhibited spontaneous GH pulses in normally fed animals, but had little effect on the high continuous GH tone during fasting. The elevated GH secretion in fasted animals could be inhibited by continuous infusion of SS or a single iv injection of a long-acting SS analog. The secretion of GH during fasting could be further increased, either by injections of GRF (two injections of 2 ng, iv, 90 min apart), producing peak levels of 102 ± 16 and 68 ± 21 ng/ml (above a baseline output of 8.8 ± 2.2 ng/ml), or by a continuous iv infusion of GRF (12 ng/h). Because the GH secretory pattern in the guinea pig is so sensitive to nutrition and insulin, this species may provide an interesting model in which to study selectively the metabolic, as opposed to growth-promoting, actions and regulation of GH. (Endocrinology 127: 716-723,1990)
curious to know how such metabolic stimuli affect GH secretion in the guinea pig and whether this hystricomorph rodent would resemble the rat or man in its GH responses to insulin or food deprivation.
Materials and Methods Animals Guinea pigs (Hartley; 300-500 g) were housed singly in metabolic cages, with food and water available ad libitum unless otherwise stated. The details of chronic jugular venous cannulation and automatic microsampling of blood have been extensively described (1, 16-18). Briefly, 20-^1 blood samples were withdrawn and diluted 1:5 with heparinized saline automatically and assayed directly for guinea pig GH. Infusions were given using double bore cannulae so that blood sampling could continue throughout the infusions. The vehicle was 0.9% saline containing 20 U/ml heparin and 100 Mg/ml BSA, and the infusion rate was 0.5 ml/h. In those experiments in which food was removed (for 8- to 60-h periods in different experiments), water remained available at all times. RIA of guinea pig GH The assay for guinea pig GH has recently been characterized (1) and is a heterologous assay based on antirabbit GH serum (obtained from Dr. A. L. Parlow, Pituitary Hormones and Antisera Center, Torrance, CA) and radiolabeled guinea pig
Received February 26, 1990. Address all correspondence and requests for reprints to: Dr. Iain Robinson, Division of Neurophysiology, National Institute for Medical Research, The Ridgeway Mill Hill, London, NW7 1AA England. 716
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 15:28 For personal use only. No other uses without permission. . All rights reserved.
GUINEA PIG GH SECRETION GH as tracer. Briefly, samples or standards (100 /A; usually 10 pg to 10 ng in triplicate; diluted in a 20% human blood-saline mixture) were mixed with 50 /ul antirabbit GH serum (1:75,000), followed by 50Ml (6,000 cpm) 125I-labeled guinea pig GH (freshly isolated by HPLC and radioiodinated by the Iodogen method). The assay buffer was PBS, pH 7.4 containing 3 mg BSA/ml and 0.6 mM thimerosal. After incubation for 24 h at room temperature, bound and free fractions were separated by the addition of 300 i*\ polyethylene glycol (18% in 0.1 M Tris-HCl, pH 8.5, containing 0.03% Triton X-100 and 2.5 mg/ml bovine 7-globulin). After 30 min the tubes were centrifuged, the supernatants were aspirated, and the pellets were counted in a ycounter. In all assays the detection limit for guinea pig GH in whole blood was below 1 ng/ml. As previously determined, the intra- and interassay coefficients of variation were 4.8% and 12%; all samples from every animal in any 24-h period were assayed in the same assay. Substances administered Insulin (porcine monocomponent soluble insulin; 1, 2, or 6 U; Novo Laboratories, Copenhagen, Denmark) was given by single iv injections in saline. Some animals were given three iv injections of dextrose (600 mg) in saline at hourly intervals. Blood glucose was monitored on freshly collected drops of whole blood using an Exactech blood glucose monitor (Baxter/Travenol) and disposable electrode strips. SS [SS-(1-14), Ferring AB, Malmo, Sweden] was infused iv at 25 Mg/h- A long-acting SS analog (SMS 201-995, Sandostatin, Sandoz AG, Basel, Switzerland) was given as a single iv injection (10 Mg)- The GRF used was a synthetic analog [(Nle27)human GRF-(l-29)NH2, Ferring AB] and was given either by injection (2 ng) at 90-min intervals or continuous iv infusion at 12 Analysis of results Data are expressed as nanograms per ml in terms of our working standard for guinea pig GH. By RIA, 1 ng of this material corresponds to 1.7 ng of the rabbit GH standard distributed by Dr. Parlow and to 1.5 ng recombinant human GH by UV absorption. Results are shown either as individual plasma GH profiles or as the mean (±SEM) GH levels in groups of animals identically treated. Differences between treatment groups were assessed by Student's t test, paired or unpaired as appropriate. For some experiments, the Pulsar algorithm (19), modified for a microcomputer by Sten Rosberg (Goteborg University, Sweden) was used to identify pulses, pulse intervals, and mean baseline values for each animal. Because in control animals, pulses tended to cluster in major secretory episodes (1), the interval between these bursts of episodic GH secretion was also estimated from the major peaks located in each episode.
Results Withdrawal of food had a dramatic effect on the secretory pattern of GH in the guinea pig. Figure 1 shows the results obtained in four chronically cannulated conscious male guinea pigs in which blood samples were withdrawn
717
40
40
c O) Q.
40
aa>
O
40
Time (h)
FIG. 1. GH secretion in normal and fasted guinea pigs. Blood samples were withdrawn automatically at 10-min intervals from four chronically cannulated conscious male guinea pigs and assayed for guinea pig GH by RIA. Left panel, Individual 24-h blood GH profiles during ad libitum feeding; right panel, a second set of 24-h blood GH profiles obtained from the same animals after food had been withdrawn for 36 h.
at 10-min intervals over a 24-h period and assayed for guinea pig GH. After this the food hoppers were emptied, and the animals fasted for 36 h. The same animals were then subjected to a second 24-h blood collection to determine their fasting GH secretory pattern. With normal feeding, the GH secretory pattern in these animals was pulsatile, with major secretory episodes occurring at 3.6 ± 0.4-h intervals. After 36 h of food withdrawal, plasma GH showed a continuous irregular profile, with a tonically elevated GH concentration without major episodes of pulses or periods of low GH secretion. In these animals, basal GH levels rose from 1.5 ± 0.05 to 6.0 ± 0.53 ng/ml during fasting (mean ± SEM; n = 4; P < 0.003, by paired t test). Similar results were obtained in female guinea pigs; in the next experiment, the time course of this fastinginduced change in GH secretion was examined by sampling continuously from eight female guinea pigs over 3 days, with food withdrawn on the middle day. Figure 2 shows complete blood GH profiles for three of the animals in this experiment. Again, the control day shows low basal GH secretion (0.44 ± 0,10 ng/ml) interrupted by episodes of GH release. Withdrawal of food caused a large rise in blood GH, beginning 5-6 h after food withdrawal. During the next 10 h, rapid oscillations in GH were observed, after which the profile remained stable at an elevated baseline (9.2 ± 1.6 ng/ml; P < 0.01 us. basal levels, by paired t test). Replacement of food lowered the baseline GH secretion over the next 10 h (to 3.5 ± 1.0 ng/ml; P < 0.05), but this was still significantly (P < 0.02) above the basal values for the control day in the same animals.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 15:28 For personal use only. No other uses without permission. . All rights reserved.
GUINEA PIG GH SECRETION
718 Fed
Fasted
Refed
Endo • 1990 Vol 127-No 2
20 -, Glucose
I II 65 i
X
g> 65 1
o O)
0 g> "5.
"5. co 3
'5
65
a
Day
Day 2
Day 3
FlG. 2. Time course of the fasting-induced rise in GH secretion. Blood samples were withdrawn automatically from eight conscious chronically cannulated female guinea pigs at 10-min intervals over a 72-h period and assayed for guinea pig GH. Representative individual blood GH profiles are shown for three animals with normal access to food on days 1 and 3. On day 2, food was withdrawn.
In the next experiment the blood GH profile was studied in fasting animals given iv glucose injections. Figure 3 (upper panel) shows the mean blood GH concentrations over a 7.5-h sampling period in nine male guinea pigs fasted for 16 h before the study and a control group of five normal males (lower panel). Note again the much higher (25-fold) basal GH values in the fasted animals (9.1 ± 1.1 ng/ml) compared with those in the controls in this experiment (mean ± SEM, 0.37 ± 0.10 ng/ml). After three iv injections of 600 mg glucose given at 60-min intervals, fasting GH levels had fallen to 6.5 ± 0.87 ng/ml (P < 0.02, by paired t test). The effect of glucose injections in the normally fed controls (Fig. 3, lower panel) is less apparent from pooled data because of the episodic nature of GH release in normal animals, but only one of the five control animals showed a GH pulse over the 4 h during and after glucose injection compared with the normal spontaneous GH pulse frequency of 3.6 h in these animals.
Time (h) FIG. 3. Effect of glucose injections on GH secretion in normal and fasted guinea pigs. Blood samples were withdrawn at 10-min intervals from two groups of conscious chronically cannulated male guinea pigs and assayed for guinea pig GH. Results shown are the mean ± SEM. The upper panel shows a group of nine guinea pigs fasted for 16 h before and 8 h during the study; the lower panel shows results from a group of five normal animals with food available ad libitum. After a control period, both groups received three injections of glucose (600 mg, iv) at hourly intervals (arrows).
In another experiment groups of conscious normally fed guinea pigs were given single iv injections of insulin (2 or 6 U), which caused a marked cessation of GH pulses over the next 3-4 h (Fig. 4). Since these doses produced hypoglycemic symptoms 1-2 h after injection in one or two of the animals, we chose a lower dose of insulin for the next experiments, in which the effect of insulin on GH secretion was compared in both fasted and fed animals. Figure 5 shows the blood GH profiles in a group of 7 male guinea pigs 16 h after food withdrawal. After 3 h a single iv injection of 1 U insulin was given, and blood sampling continued for a further 5 h. The results from a control experiment in a group of 10 normally fed animals is also shown in the lower panel of Fig. 5. Baseline GH secretion was again greatly elevated in this fasted group (14 ± 2.5 ng/ml) compared with that in the control group
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 15:28 For personal use only. No other uses without permission. . All rights reserved.
GUINEA PIG GH SECRETION
719 Insulin 1u
30
D) C
(D o '5.
E
0 30
(0
a>
x o "a
tc
"5
(3
*****
(0
c '5
Time (h) FIG. 5. Effects of insulin on GH secretion in fed and fasted guinea pigs. Blood samples were withdrawn automatically at 10-min intervals from 2 groups of male guinea pigs over 8 h and assayed for guinea pig GH. The lower panel shows blood GH (mean ± SEM) levels in a group of 10 control animals with food available ad libitum; the upper panel shows results from a group of 7 animals fasted for 16 h before and 8 h during the sampling study. After 3 h, a single injection of porcine insulin (1 U) was given iv (arrow). 60-,
SMS (10ng)
Time (h) FlG. 4. Effects of insulin on GH secretion in conscious guinea pigs. Blood samples were withdrawn at 10-min intervals from groups of conscious male guinea pigs and assayed for guinea pig GH. All animals were given a single iv injection of porcine insulin (arrows), either 2 U (top panel; n = 5) or 6 U (bottom panel; n = 6). Individual blood GH profiles are shown.
(0.80 ± 0.10 ng/ml), and insulin injection had a slight but insignificant effect (P = 0.1) on these high GH levels (10.8 ± 1.3 ng/ml), which remained significantly above those in the normally fed animals. Blood glucose levels in these animals 60 min after insulin injection had fallen to 1.5 ± 0.2 raM compared to 5.2 ± 0.7 mM measured in the same animals at this time on a previous control day. In the normally fed control group (Fig. 5, lower panel), iv insulin injection again suppressed the next episodic GH secretory burst; the interburst interval across the insulin injection (4.8 ± 0.3 h) was significantly longer than that between the 3 subsequent episodes of GH secretion in each animal (3.5 ± 0.2, 2.9 ± 0.3, and 2.5 ± 0.4 h; P < 0.002 in each case). The sensitivity of the fasting-induced high GH secretion to SS was investigated in the next two experiments. First, five fasted male guinea pigs were given a continuous iv infusion of SS at 25 yug/h while blood samples were drawn for GH assay. The elevated baseline GH secretion in this group of fasted animals was 14 ± 3 ng/
o J? a. a a> 3
30.
(5
lli.ii Time (h) FIG. 6. Effects of a long-acting SS analog on GH secretion in fasted guinea pigs. Blood samples were withdrawn automatically every 10 min for 8 h from a group of 7 chronically cannulated male guinea pigs that had been fasting for 16 h before and 8 h during the study. After 16 samples, a single injection of SMS 201-995 (Sandostatin; 10 fig, iv) was given. The samples were assayed for guinea pig GH, and the results shown are the mean ± SEM.
ml, and SS infusion lowered this to 8 ± 3 ng/ml (mean ± SEM; P < 0.05, by paired t test). In the second approach, seven fasting male guinea pigs were given a single injection of a long-acting SS analog (SMS 201-995; 10 ng, iv) while blood samples were drawn for GH assay (Fig. 6). Again, the SS analog significantly (P < 0.02) lowered the high blood levels of GH in fasting animals; the nadir occurred 35-45 min after injection, with GH levels re-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 15:28 For personal use only. No other uses without permission. . All rights reserved.
720
GUINEA PIG GH SECRETION
turning to preinjection values within 80 min of analog injection. Both of these experiments clearly show that elevated GH secretion during fasting in the guinea pig is sensitive to inhibition by SS. In the last pair of experiments the GH secretory responses to GRF were tested in fasted animals. In one experiment (Fig. 7) six male guinea pigs were fasted for 8 h before sampling commenced and showed a continuous high baseline GH secretion (8.8 ± 2.2 ng/ml). Two GRF injections (2 ng, iv) at 90-min intervals elicited large peaks of GH secretion (peak GH levels, 102 ± 16 and 68 ± 2 1 ng/ml, respectively) above this elevated baseline (Fig. 7). In the last experiment, continuous iv infusions of GRF (12 ng/h) were given to five fasted guinea pigs; their individual blood GH profiles are shown in Fig. 8. Before GRF infusion, the basal GH level was 20 ± 6 ng/ ml; it rose to a peak 60-90 min after the onset of GRF and fell to a new continuous level of 54 ± 10 ng/ml during GRF infusion, significantly (P < 0.05) higher than that before GRF infusion. Thus, despite high continuous basal GH release during fasting, the output can be further increased by injections or infusions of exogenous GRF.
E n d o • 1990 Vol 127 • No 2
GRF (12ng/h)
|
60
E
0 60
O)
c I
0 60
O O) CL CO
0 60
0)
c
"5
O
0 60
^
J
Discussion In this paper we extend our studies of the control of GH secretion in the guinea pig. Earlier studies showed clearly that, unlike in other mammals, hypophysectomy GRF(2ng)
100 -i
D) C
X
o
50
o>
"5. OS Qi
c
3
(3 10
Time (h) FIG. 7. Effects of GRF injections on GH secretion in fasted male guinea pigs. Blood samples were withdrawn from a group of six conscious male animals that had been fasted for 8 h before and 10 h during the study and assayed for blood GH. After a 7-h control period, two serial injections of (27Nle)humanGRF-(l-29-)NH2 (2 /zg) were given 90 min apart (arrows). Results shown are the mean ± SEM.
Time (h) FIG. 8. Effects of continuous infusion of GRF on GH secretion in fasted guinea pigs. Blood samples were withdrawn at 10-min intervals from five male guinea pigs that had been fasted for 36 h before and 5 h during the study and assayed for guinea pig GH. After 1 h, a continuous iv infusion of (27Nle)human GRF-(l-29)-NH2 (12 Mg/h) was begun (D). Individual blood GH profiles are shown.
has only a minor effect on growth in the guinea pig (2, 4, 20), but the guinea pig does produce a GH molecule that can support growth in the hypophysectomized rat (21). We have recently extended our automatic bloodsampling method to the conscious guinea pig and, by isolating guinea pig GH and developing a sensitive heterologous RIA for this hormone, were able to report the first studies of the secretory pattern of GH in the guinea pig and its responses to exogenous GRF and SS (1). Since the guinea pig secretes GH, but does not apparently require this hormone for stimulating growth, we suggested that its physiological role may be metabolic rather than growth promotion in this species. There is some circumstantial evidence for a link between GH and insulin in the guinea pig. Insulin circulates at relatively high levels in normal guinea pigs (22), perhaps compensating for its relatively weak hypoglycemic potency compared to that of other mammalian insulins (23), and Clayton and Worden (20) noted that hypophysectomy in the guinea pig induced a transient
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 15:28 For personal use only. No other uses without permission. . All rights reserved.
GUINEA PIG GH SECRETION period of marked insulin hypersensitivity. We reasoned that if guinea pig GH was involved in metabolic control, it might be particularly sensitive to insulin and nutrition. Furthermore, it would be interesting to discover if this hystricomorph rodent responded to food withdrawal with decreased GH output, as in the rat, or increased GH release, as seen in most other species, including man. Our first studies showed that removal of food did indeed have a dramatic effect on GH secretion. Instead of an episodic pattern of GH release, with clusters of pulses interspersed with periods of low or undetectable GH release, fasting induced a sustained rise in baseline secretion; the large episodic pulses were replaced by a continuous irregular rapid secretory pattern in both sexes. Thus, the responses resembled those during fasting in man and other large animals (7-10) and contrasted directly with those in the rat, in which GH secretion is reduced by fasting (11). The increased GH output in response to fasting in the guinea pig is quite rapid (detectable changes occur within a few hours of food withdrawal) and is relatively long-lasting (requiring at least 24 h after refeeding for the normal episodic pattern to return). It is not clear what aspect of fasting (calorie intake, available hexoses, amino acids, and fatty acids) stimulated increased GH release. In man, oral glucose inhibits GRF-induced GH release (24); in the rat, FFA inhibit both spontaneous and GRF-induced GH secretion (25), although moderate hyperglycemia per se does not block spontaneous GH release (15, 25, 26). In the present studies glucose injections reduced GH secretion in the guinea pig to a limited extent by delaying GH pulses in normally fed animals and lowering the elevated GH baseline secretion in fasted animals. It is also possible that food withdrawal represents a stress stimulus, and it is notable that the GH response to stress also differs in rats and primates. In the rat, stress blocks spontaneous GH release, probably by an increase in SS secretion (27), whereas GH release is stimulated by stress in primates (28). We have not yet specifically investigated the effects of stress on GH secretion in the guinea pig. The GH responses to insulin injections in the guinea pig more closely resembled those in the rat (15), with a prompt blockade of secretory episodes in normally fed animals, although to our surprise, insulin was much less effective in suppressing the high basal GH output in fasted guinea pigs, which should be more susceptible to insulin-induced hypoglycaemia. It is possible that a higher insulin dose would have been more effective, but an insulin-induced rise in GH secretion, as seen in man, was never observed in the guinea pig even at the largest doses (6 U), which caused marked hypoglycemia in some normally fed animals. It should be noted, however, that porcine insulin was used for these studies, since the
721
amounts of purified guinea pig insulin available were not sufficient for these in vivo experiments. Although porcine insulin is a potent hypoglycemic agent in the guinea pig, we cannot exclude the possibility that the endogenous guinea pig insulin might have different effects on GH secretion. It is also conceivable that insulin could inhibit GH release indirectly, for example by stimulating the release of gut SS, although this is unlikely in view of the relatively large amounts of exogenous SS required to suppress GH secretion in the guinea pig (1). How does fasting cause such a marked alteration of secretion from episodic pulses to a high continuous GH secretory pattern in the guinea pig? In the rat, a similar increase in baseline GH can be brought about by blocking SS release immunochemically (29, 30), and it seems reasonable to assume that fasting could have reduced the release of SS or blocked its action at the guinea pig pituitary gland. That the latter explanation is unlikely was shown by infusing SS or injecting a long-acting SS analog (31); both maneuvers promptly lowered the elevated basal GH release in fasted guinea pigs. Another possibility is that fasting increases endogenous GRF secretion in the guinea pig. Although GH secretion is high during fasting, it was nowhere near maximal, since injections or infusions of exogenous GRF evoked a much greater GH output. In both rat and man, infusions of GRF amplify endogenous episodic GH release (32-34), but there was no evidence for an amplification of any underlying episodic secretion by continuous GRF in fasted guinea pigs. Since the episodic release of GH seems to be determined by the interplay between GRF release and an intermittent SS tone (35, 36), our working hypothesis is that fasting exerts its effects on GH release in the guinea pig via both SS and GRF. SS output is reduced and its secretory rhythm disrupted (accounting for the change from episodic to continuous high GH tone), and this may be accompanied by an increased release of GRF or a change in its secretory pattern from infrequent pulses to more frequent or even continuous release. One other influence that we have not addressed is the effect of insulin-like growth factor-I (IGF-I) in fasted guinea pigs. In other species, fasting lowers IGF-I levels markedly (37, 38), and since IGFTI can inhibit both the release (39) and synthesis (40) of GH, it is possible that fasting increases total GH output partly by disinhibition of the pituitary gland as IGF-I levels fall. It is also possible that porcine insulin could suppress GH secretion via IGF-I receptors, rather than insulin receptors, in the guinea pig. In conclusion, these studies confirm that the guinea pig presents an interesting model of GH secretion and control. The relevance to other species of nutritional regulation of GH in the rat is questionable, since the
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 15:28 For personal use only. No other uses without permission. . All rights reserved.
GUINEA PIG GH SECRETION
722
responses are opposite those seen in most other species; the guinea pig would seem to offer a useful alternative small laboratory animal model in this regard. The fact that this species does not use GH as a growth-promoting substance suggests that it might be a good species in which to study selectively the metabolic regulation of GH. The importance of the episodic secretory pattern of GH has been stressed in both experimental and clinical studies, with the emphasis largely on the effectiveness of a pulsatile GH pattern to stimulate growth (41). The normal guinea pig also shows a highly episodic GH secretory pattern, but the significance of this must presumably be directed to the metabolic, rather than growth-promoting, effects of this hormone. Perhaps, the guinea pig will help us to focus on the importance of the
14. 15. 16. 17. 18. 19. 20.
secretory pattern and actions of GH in adult life, where
21.
growth promotion is no longer the primary function of GH in any species.
22.
Acknowledgments
23.
We are grateful to Dr. A. L. Parlow for the continued supply of assay reagents which form the basis of the guinea pig GH assay. We thank Ms. Norrie Shannon for excellent technical assistance and Sten Rosberg, Goteborg University, for his version of the Pulsar program. We also thank Ferring AB (Malmo, Sweden) and Sandoz AG (Basel) for supplying the peptides used in this study.
24.
References
26.
1. Gabrielsson B, Fairhall KM, Robinson ICAF 1990 Growth hormone secretion in the guinea pig. J Endocrinol 124:371 2. Mitchell ML, Guillemin R, Selye H 1954 The effect of somatotropic hormone in the growth of normal and hypophysectomized guinea pigs. Endocrinology 54:111 3. Knobil E, Hotchkiss J 1964 Growth hormone. Annu Rev Physiol 26:47 4. Ponse K, Vecsey A 1970 Survie et croissance insolites de Cobayes femelles hypophysectomisees. Rev Suisse Zool 77:713 5. Davidson MB 1987 Effect of growth hormone on carbohydrate and lipid metabolism. Endocr Rev 8:115 6. Roth J, Glick SM, Yalow RS, Berson SA 1963 Hypoglycaemia: a potent stimulus to secretion of growth hormone. Science 140:987 7. Bassett JM 1974 Diurnal patterns of insulin, growth hormone, corticosteroid and metabolite concentrations in fed and fasted sheep. Aust J Biol Sci 27:167 8. Breier BH, Bass JJ, Butler JH, Gluckman PD 1986 The somatotrophic axis in young steers: influence of nutritional status on pulsatile release of growth hormone and circulating concentrations of insulin-like growth factor 1. J Endocrinol 111:209 9. Merimee TJ, Fineberg SE 1974 Growth hormone secretion in starvation-a reassessment. J Clin Endocrinol Metab 39:385 10. Ho KY, Veldhuis JD, Johnson ML, Furlanetto R, Evans WS, Alberti KGHM, Thorner MO 1988 Fasting enhances growth hormone secretion and amplifies the complex rhythms of growth hormone secretion in man. J Clin Invest 81:968 11. Tannenbaum GS, Rorstad O, Brazeau P 1979 Effects of prolonged food deprivation on the ultradian growth hormone rhythm and immunoreactive somatostatin tissue levels in the rat. Endocrinology 104:1733 12. Painson J-C, Tannenbaum GS 1985 Effects of intracellular glucopenia on pulsatile growth hormone secretion: mediation in part by somatostatin. Endocrinology 11:1132 13. Tannenbaum GS 1981 Growth hormone secretory dynamics in
25.
27. 28. 29.
30.
31. 32. 33.
34.
35. 36.
E n d o • 1990 Vol 127 • No 2
streptozotocin diabetes: evidence for a role for endogenous circulating somatostatin. Endocrinology 108:76 Carlsson LMS, Clark RG, Skottner A, Robinson ICAF 1989 Growth hormone and growth in diabetic rats: effects of insulin and insulin-like growth factor-I infusions. J Endocrinol 122:661 Tannenbaum GS, Martin JB, Colle E 1976 Ultradian growth hormone rhythm in the rat: effects of feeding, hyperglycemia and insulin-induced hypoglycaemia. Endocrinology 99:720 Clark RG, Chambers G, Lewin J, Robinson ICAF 1986 Automated repetitive microsampling of blood: growth hormone secretion in conscious male rats. Endocrinology 111:27 Clark RG, Carlsson LMS, Robinson ICAF 1987 Growth hormone secretory profiles in conscious female rats. J Endocrinol 114:399 Clark RG, Robinson ICAF 1988 Paradoxical growth promoting effects induced by patterned infusions of somatostatin in female rats. Endocrinology 122:2675 Merriam GR, Wachter KW 1982 Algorithms for the study of episodic hormone secretion. Am J Physiol 243:E310 Clayton BE, Worden JM 1960 Growth in young hypophysectomized guinea pigs. J Endocrinol 20:30 Knobil E, Greep RO 1959 The physiology of growth hormone with particular reference to its action in the rhesus monkey and the "species-specificity" problem. Recent Prog Horm Res 15:1 Gorray KC, Fujimoto WY 1980 "Micro"-insulin radioimmunoassay: measurement of the insulin response during glucose tolerance tests in guinea pigs. Proc Soc Exp Biol Med 163:388 Zimmerman AE, Moule ML, Yip CC 1974 Guinea pig insulin. II. Biological activity. J Biol Chem 249:4026 Davies RR, Turner S, Johnston DG 1984 Oral glucose inhibits growth hormone secretion induced by human pancreatic growth hormone releasing factor 1-44 in normal man. Clin Endocrinol (Oxf) 21:477 Imaki T, Shibasaki T, Masuda A, Hotta M, Yamauchi N, Demura H, Shizume K, Wakabayashi I, Ling N 1986 The effect of glucose and free fatty acids on growth hormone (GH)-releasing factormediated GH secretion in rats. Endocrinology 118:2390 King RA, Smith RM, Willoughby JO 1986 Effect of sustained hyperglycemia on growth hormone secretion in free-moving rats. Horm Metab Res 18:510 Arimura A, Smith WD, Schally AV 1976 Blockade of the stressinduced decrease in blood GH by anti-somatostatin serum in rats. Endocrinology 98:540 Quabbe JJ 1985 Hypothalamic control of GH secretion: pathophysiology and clinical implications. Acta Neurochir (Wien) 75:60 Tannenbaum GS, Epelbaum J, Cole E, Brazeau P, Martin JB 1978 Antiserum to somatostatin reverses starvation-induced inhibition of growth hormone but not insulin secretion. Endocrinology 102:1909 Terry LC, Martin JB 1981 The effects of lateral hypothalamicmedial forebrain stimulation and somatostatin antiserum on pulsatile growth hormone secretion in freely behaving rats: evidence for a dual regulatory mechanism. Endocrinology 109:622 Bauer W, Briner U, Doepfner W, Haller R, Hugenin R, Marbach P, Petcher TJ, Pless J 1982 A very potent and selective octapeptide analogue of somatostatin with prolonged action. Life Sci 31:1133 Wehrenberg WB 1986 Continuous infusion of growth hormone releasing factor: effects on pulsatile growth hormone secretion in normal rats. Neuroendocrinology 43:391 Vance ML, Kaiser DL, Evans WS, Furlanetto R, Vale W, Rivier J, Thorner MO 1985 Pulsatile growth hormone secretion in normal man during a continuous 24-hour infusion of human growth hormone releasing factor (1-40). J Clin Invest 75:1584 Brain C, Hindmarsh PC, Brook CGD, Matthews DR 1988 Continuous subcutaneous growth hormone releasing factor analogue augments growth hormone secretion in normal male subjects with no desensitization of the somatotroph. Clin Endocrinol (Oxf) 28:543 Tannenbaum GS, Ling N 1984 The interrelationship of growth hormone (GH)-releasing factor and somatostatin in the generation of the ultradian rhythm of GH secretion. Endocrinology 115:1952 Plotsky PM, Vale W 1985 Patterns of growth hormone releasing factor and somatostatin secretion into the hypophysial portal circulation of the rat. Science 230:461
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 15:28 For personal use only. No other uses without permission. . All rights reserved.
GUINEA PIG GH SECRETION 37. Clemmons DR, Klibanski A, Underwood LE, McArthur JW, Ridgeway EC, Beitins IZ, van Wyk JJ 1981 Reduction of plasma immunoreactive somatomedin-C during fasting in humans. J Clin Endocrinol Metab 53:1247 38. Maes M, Underwood LE, Ketelslegers JM 1984 Low serum somatomedin-C in protein deficiency: relationship with changes in liver somatogenic and lactogenic bindir o sites. Mol Cell Endocrinol 37:301 39. Abe H, Molitch ME, van Wyk JJ, Underwood LE 1983 Human growth hormone and somatomedin-C suppress the spontaneous
723
release of growth hormone in unanesthetized rats. Endocrinology 113:1319 40. Yamashita S, Melmed S 1986 Insulin-like growth factor I action on rat anterior pituitary cells: suppression of growth hormone secretion and messenger ribonucleic acid levels. Endocrinology 118:176 41. Jansson J-O, Albertsson-Wikland K, Eden S, Thorngren K-G, Isaksson O 1982 Circumstantial evidence for a role of the secretory pattern of growth hormone in control of body growth. Acta Endocrinol (Copenh) 99:24
DON'T MISS THE ENDOCRINE SOCIETY'S 42nd POSTGRADUATE ASSEMBLY A must for anyone involved in Endocrinology or Internal Medicine October 28-November 1, 1990 Sheraton Waikiki Honolulu, Hawaii For program and registration information please contact: The Endocrine Society 9650 Rockville Pike Bethesda, MD 20814 (301) 571-1802 FAX (301) 571-1869
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 15:28 For personal use only. No other uses without permission. . All rights reserved.
Vol. 127, No. 2 Printed in U.S.A.
Effect of Food Withdrawal and Insulin on Growth Hormone Secretion in the Guinea Pig KEITH M. FAIRHALL, BRITT G. GABRIELSSON, AND IAIN C. A. F. ROBINSON Division of Neurophysiology and Neuropharmacology, National Institute for Medical Research, The Ridgeway Mill Hill, London, NW7 1AA England
ABSTRACT. The guinea pig is unusual in that its postnatal growth appears to be independent of GH even though its pituitary gland produces a GH molecule. The effects of fasting on the GH secretory pattern and the GH responses to insulin, GHreleasing factor (GRF), and somatostatin (SS) during fasting have now been studied by automatic microsampling of blood in chronically cannulated normal guinea pigs. Withdrawal of food in both male and female guinea pigs changed the GH secretory pattern dramatically. The normal episodic GH secretory pattern [large GH peaks occurring at 3.6 ± 0.4-h intervals over a low (~0.5-1.5 ng/ml) baseline secretion] was altered to a pattern of more continuous GH output, characterized by a 10-fold elevated baseline secretion (5-15 ng/ml) with no large secretory episodes or troughs. Glucose injections (three injections of 600 mg, iv, at hourly intervals) in fasted guinea pigs lowered their elevated
W
E RECENTLY reported the first characterization of the secretory pattern of GH in normal male and female guinea pigs, and showed that, like most other species, GH release was episodic and could be stimulated or inhibited by GH-releasing factor (GRF) or somatostatin (SS), respectively (1). Our interest in GH control in the guinea pig stems from the early observations that despite producing large amounts of a pituitary GH, they do not appear to need this hormone for postnatal growth, unlike other mammals (2-4). Since GH has important metabolic actions in many species (5), it occurred to us that the guinea pig may represent one extreme in which GH is primarily a metabolic hormone and not a growthpromoting agent. Although GH release is sensitive to insulin and nutrient supply, the responses differ radically between species. In man, GH release is stimulated acutely by insulin-induced hypoglycemia (6), and prolonged GH hypersecretion occurs during fasting in man and larger animals (7-10). By contrast, starvation, intracellular glucopenia, and diabetes in the rat all cause prolonged suppression of GH release (11-14), and insulin blocks episodic GH pulses (15). We were, therefore,
blood GH levels significantly (from 9.1 ± 1.1 to 6.5 ± 0.9 ng/ ml). Insulin injections (1, 2, or 6 U, iv) inhibited spontaneous GH pulses in normally fed animals, but had little effect on the high continuous GH tone during fasting. The elevated GH secretion in fasted animals could be inhibited by continuous infusion of SS or a single iv injection of a long-acting SS analog. The secretion of GH during fasting could be further increased, either by injections of GRF (two injections of 2 ng, iv, 90 min apart), producing peak levels of 102 ± 16 and 68 ± 21 ng/ml (above a baseline output of 8.8 ± 2.2 ng/ml), or by a continuous iv infusion of GRF (12 ng/h). Because the GH secretory pattern in the guinea pig is so sensitive to nutrition and insulin, this species may provide an interesting model in which to study selectively the metabolic, as opposed to growth-promoting, actions and regulation of GH. (Endocrinology 127: 716-723,1990)
curious to know how such metabolic stimuli affect GH secretion in the guinea pig and whether this hystricomorph rodent would resemble the rat or man in its GH responses to insulin or food deprivation.
Materials and Methods Animals Guinea pigs (Hartley; 300-500 g) were housed singly in metabolic cages, with food and water available ad libitum unless otherwise stated. The details of chronic jugular venous cannulation and automatic microsampling of blood have been extensively described (1, 16-18). Briefly, 20-^1 blood samples were withdrawn and diluted 1:5 with heparinized saline automatically and assayed directly for guinea pig GH. Infusions were given using double bore cannulae so that blood sampling could continue throughout the infusions. The vehicle was 0.9% saline containing 20 U/ml heparin and 100 Mg/ml BSA, and the infusion rate was 0.5 ml/h. In those experiments in which food was removed (for 8- to 60-h periods in different experiments), water remained available at all times. RIA of guinea pig GH The assay for guinea pig GH has recently been characterized (1) and is a heterologous assay based on antirabbit GH serum (obtained from Dr. A. L. Parlow, Pituitary Hormones and Antisera Center, Torrance, CA) and radiolabeled guinea pig
Received February 26, 1990. Address all correspondence and requests for reprints to: Dr. Iain Robinson, Division of Neurophysiology, National Institute for Medical Research, The Ridgeway Mill Hill, London, NW7 1AA England. 716
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 15:28 For personal use only. No other uses without permission. . All rights reserved.
GUINEA PIG GH SECRETION GH as tracer. Briefly, samples or standards (100 /A; usually 10 pg to 10 ng in triplicate; diluted in a 20% human blood-saline mixture) were mixed with 50 /ul antirabbit GH serum (1:75,000), followed by 50Ml (6,000 cpm) 125I-labeled guinea pig GH (freshly isolated by HPLC and radioiodinated by the Iodogen method). The assay buffer was PBS, pH 7.4 containing 3 mg BSA/ml and 0.6 mM thimerosal. After incubation for 24 h at room temperature, bound and free fractions were separated by the addition of 300 i*\ polyethylene glycol (18% in 0.1 M Tris-HCl, pH 8.5, containing 0.03% Triton X-100 and 2.5 mg/ml bovine 7-globulin). After 30 min the tubes were centrifuged, the supernatants were aspirated, and the pellets were counted in a ycounter. In all assays the detection limit for guinea pig GH in whole blood was below 1 ng/ml. As previously determined, the intra- and interassay coefficients of variation were 4.8% and 12%; all samples from every animal in any 24-h period were assayed in the same assay. Substances administered Insulin (porcine monocomponent soluble insulin; 1, 2, or 6 U; Novo Laboratories, Copenhagen, Denmark) was given by single iv injections in saline. Some animals were given three iv injections of dextrose (600 mg) in saline at hourly intervals. Blood glucose was monitored on freshly collected drops of whole blood using an Exactech blood glucose monitor (Baxter/Travenol) and disposable electrode strips. SS [SS-(1-14), Ferring AB, Malmo, Sweden] was infused iv at 25 Mg/h- A long-acting SS analog (SMS 201-995, Sandostatin, Sandoz AG, Basel, Switzerland) was given as a single iv injection (10 Mg)- The GRF used was a synthetic analog [(Nle27)human GRF-(l-29)NH2, Ferring AB] and was given either by injection (2 ng) at 90-min intervals or continuous iv infusion at 12 Analysis of results Data are expressed as nanograms per ml in terms of our working standard for guinea pig GH. By RIA, 1 ng of this material corresponds to 1.7 ng of the rabbit GH standard distributed by Dr. Parlow and to 1.5 ng recombinant human GH by UV absorption. Results are shown either as individual plasma GH profiles or as the mean (±SEM) GH levels in groups of animals identically treated. Differences between treatment groups were assessed by Student's t test, paired or unpaired as appropriate. For some experiments, the Pulsar algorithm (19), modified for a microcomputer by Sten Rosberg (Goteborg University, Sweden) was used to identify pulses, pulse intervals, and mean baseline values for each animal. Because in control animals, pulses tended to cluster in major secretory episodes (1), the interval between these bursts of episodic GH secretion was also estimated from the major peaks located in each episode.
Results Withdrawal of food had a dramatic effect on the secretory pattern of GH in the guinea pig. Figure 1 shows the results obtained in four chronically cannulated conscious male guinea pigs in which blood samples were withdrawn
717
40
40
c O) Q.
40
aa>
O
40
Time (h)
FIG. 1. GH secretion in normal and fasted guinea pigs. Blood samples were withdrawn automatically at 10-min intervals from four chronically cannulated conscious male guinea pigs and assayed for guinea pig GH by RIA. Left panel, Individual 24-h blood GH profiles during ad libitum feeding; right panel, a second set of 24-h blood GH profiles obtained from the same animals after food had been withdrawn for 36 h.
at 10-min intervals over a 24-h period and assayed for guinea pig GH. After this the food hoppers were emptied, and the animals fasted for 36 h. The same animals were then subjected to a second 24-h blood collection to determine their fasting GH secretory pattern. With normal feeding, the GH secretory pattern in these animals was pulsatile, with major secretory episodes occurring at 3.6 ± 0.4-h intervals. After 36 h of food withdrawal, plasma GH showed a continuous irregular profile, with a tonically elevated GH concentration without major episodes of pulses or periods of low GH secretion. In these animals, basal GH levels rose from 1.5 ± 0.05 to 6.0 ± 0.53 ng/ml during fasting (mean ± SEM; n = 4; P < 0.003, by paired t test). Similar results were obtained in female guinea pigs; in the next experiment, the time course of this fastinginduced change in GH secretion was examined by sampling continuously from eight female guinea pigs over 3 days, with food withdrawn on the middle day. Figure 2 shows complete blood GH profiles for three of the animals in this experiment. Again, the control day shows low basal GH secretion (0.44 ± 0,10 ng/ml) interrupted by episodes of GH release. Withdrawal of food caused a large rise in blood GH, beginning 5-6 h after food withdrawal. During the next 10 h, rapid oscillations in GH were observed, after which the profile remained stable at an elevated baseline (9.2 ± 1.6 ng/ml; P < 0.01 us. basal levels, by paired t test). Replacement of food lowered the baseline GH secretion over the next 10 h (to 3.5 ± 1.0 ng/ml; P < 0.05), but this was still significantly (P < 0.02) above the basal values for the control day in the same animals.
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 15:28 For personal use only. No other uses without permission. . All rights reserved.
GUINEA PIG GH SECRETION
718 Fed
Fasted
Refed
Endo • 1990 Vol 127-No 2
20 -, Glucose
I II 65 i
X
g> 65 1
o O)
0 g> "5.
"5. co 3
'5
65
a
Day
Day 2
Day 3
FlG. 2. Time course of the fasting-induced rise in GH secretion. Blood samples were withdrawn automatically from eight conscious chronically cannulated female guinea pigs at 10-min intervals over a 72-h period and assayed for guinea pig GH. Representative individual blood GH profiles are shown for three animals with normal access to food on days 1 and 3. On day 2, food was withdrawn.
In the next experiment the blood GH profile was studied in fasting animals given iv glucose injections. Figure 3 (upper panel) shows the mean blood GH concentrations over a 7.5-h sampling period in nine male guinea pigs fasted for 16 h before the study and a control group of five normal males (lower panel). Note again the much higher (25-fold) basal GH values in the fasted animals (9.1 ± 1.1 ng/ml) compared with those in the controls in this experiment (mean ± SEM, 0.37 ± 0.10 ng/ml). After three iv injections of 600 mg glucose given at 60-min intervals, fasting GH levels had fallen to 6.5 ± 0.87 ng/ml (P < 0.02, by paired t test). The effect of glucose injections in the normally fed controls (Fig. 3, lower panel) is less apparent from pooled data because of the episodic nature of GH release in normal animals, but only one of the five control animals showed a GH pulse over the 4 h during and after glucose injection compared with the normal spontaneous GH pulse frequency of 3.6 h in these animals.
Time (h) FIG. 3. Effect of glucose injections on GH secretion in normal and fasted guinea pigs. Blood samples were withdrawn at 10-min intervals from two groups of conscious chronically cannulated male guinea pigs and assayed for guinea pig GH. Results shown are the mean ± SEM. The upper panel shows a group of nine guinea pigs fasted for 16 h before and 8 h during the study; the lower panel shows results from a group of five normal animals with food available ad libitum. After a control period, both groups received three injections of glucose (600 mg, iv) at hourly intervals (arrows).
In another experiment groups of conscious normally fed guinea pigs were given single iv injections of insulin (2 or 6 U), which caused a marked cessation of GH pulses over the next 3-4 h (Fig. 4). Since these doses produced hypoglycemic symptoms 1-2 h after injection in one or two of the animals, we chose a lower dose of insulin for the next experiments, in which the effect of insulin on GH secretion was compared in both fasted and fed animals. Figure 5 shows the blood GH profiles in a group of 7 male guinea pigs 16 h after food withdrawal. After 3 h a single iv injection of 1 U insulin was given, and blood sampling continued for a further 5 h. The results from a control experiment in a group of 10 normally fed animals is also shown in the lower panel of Fig. 5. Baseline GH secretion was again greatly elevated in this fasted group (14 ± 2.5 ng/ml) compared with that in the control group
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 15:28 For personal use only. No other uses without permission. . All rights reserved.
GUINEA PIG GH SECRETION
719 Insulin 1u
30
D) C
(D o '5.
E
0 30
(0
a>
x o "a
tc
"5
(3
*****
(0
c '5
Time (h) FIG. 5. Effects of insulin on GH secretion in fed and fasted guinea pigs. Blood samples were withdrawn automatically at 10-min intervals from 2 groups of male guinea pigs over 8 h and assayed for guinea pig GH. The lower panel shows blood GH (mean ± SEM) levels in a group of 10 control animals with food available ad libitum; the upper panel shows results from a group of 7 animals fasted for 16 h before and 8 h during the sampling study. After 3 h, a single injection of porcine insulin (1 U) was given iv (arrow). 60-,
SMS (10ng)
Time (h) FlG. 4. Effects of insulin on GH secretion in conscious guinea pigs. Blood samples were withdrawn at 10-min intervals from groups of conscious male guinea pigs and assayed for guinea pig GH. All animals were given a single iv injection of porcine insulin (arrows), either 2 U (top panel; n = 5) or 6 U (bottom panel; n = 6). Individual blood GH profiles are shown.
(0.80 ± 0.10 ng/ml), and insulin injection had a slight but insignificant effect (P = 0.1) on these high GH levels (10.8 ± 1.3 ng/ml), which remained significantly above those in the normally fed animals. Blood glucose levels in these animals 60 min after insulin injection had fallen to 1.5 ± 0.2 raM compared to 5.2 ± 0.7 mM measured in the same animals at this time on a previous control day. In the normally fed control group (Fig. 5, lower panel), iv insulin injection again suppressed the next episodic GH secretory burst; the interburst interval across the insulin injection (4.8 ± 0.3 h) was significantly longer than that between the 3 subsequent episodes of GH secretion in each animal (3.5 ± 0.2, 2.9 ± 0.3, and 2.5 ± 0.4 h; P < 0.002 in each case). The sensitivity of the fasting-induced high GH secretion to SS was investigated in the next two experiments. First, five fasted male guinea pigs were given a continuous iv infusion of SS at 25 yug/h while blood samples were drawn for GH assay. The elevated baseline GH secretion in this group of fasted animals was 14 ± 3 ng/
o J? a. a a> 3
30.
(5
lli.ii Time (h) FIG. 6. Effects of a long-acting SS analog on GH secretion in fasted guinea pigs. Blood samples were withdrawn automatically every 10 min for 8 h from a group of 7 chronically cannulated male guinea pigs that had been fasting for 16 h before and 8 h during the study. After 16 samples, a single injection of SMS 201-995 (Sandostatin; 10 fig, iv) was given. The samples were assayed for guinea pig GH, and the results shown are the mean ± SEM.
ml, and SS infusion lowered this to 8 ± 3 ng/ml (mean ± SEM; P < 0.05, by paired t test). In the second approach, seven fasting male guinea pigs were given a single injection of a long-acting SS analog (SMS 201-995; 10 ng, iv) while blood samples were drawn for GH assay (Fig. 6). Again, the SS analog significantly (P < 0.02) lowered the high blood levels of GH in fasting animals; the nadir occurred 35-45 min after injection, with GH levels re-
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 15:28 For personal use only. No other uses without permission. . All rights reserved.
720
GUINEA PIG GH SECRETION
turning to preinjection values within 80 min of analog injection. Both of these experiments clearly show that elevated GH secretion during fasting in the guinea pig is sensitive to inhibition by SS. In the last pair of experiments the GH secretory responses to GRF were tested in fasted animals. In one experiment (Fig. 7) six male guinea pigs were fasted for 8 h before sampling commenced and showed a continuous high baseline GH secretion (8.8 ± 2.2 ng/ml). Two GRF injections (2 ng, iv) at 90-min intervals elicited large peaks of GH secretion (peak GH levels, 102 ± 16 and 68 ± 2 1 ng/ml, respectively) above this elevated baseline (Fig. 7). In the last experiment, continuous iv infusions of GRF (12 ng/h) were given to five fasted guinea pigs; their individual blood GH profiles are shown in Fig. 8. Before GRF infusion, the basal GH level was 20 ± 6 ng/ ml; it rose to a peak 60-90 min after the onset of GRF and fell to a new continuous level of 54 ± 10 ng/ml during GRF infusion, significantly (P < 0.05) higher than that before GRF infusion. Thus, despite high continuous basal GH release during fasting, the output can be further increased by injections or infusions of exogenous GRF.
E n d o • 1990 Vol 127 • No 2
GRF (12ng/h)
|
60
E
0 60
O)
c I
0 60
O O) CL CO
0 60
0)
c
"5
O
0 60
^
J
Discussion In this paper we extend our studies of the control of GH secretion in the guinea pig. Earlier studies showed clearly that, unlike in other mammals, hypophysectomy GRF(2ng)
100 -i
D) C
X
o
50
o>
"5. OS Qi
c
3
(3 10
Time (h) FIG. 7. Effects of GRF injections on GH secretion in fasted male guinea pigs. Blood samples were withdrawn from a group of six conscious male animals that had been fasted for 8 h before and 10 h during the study and assayed for blood GH. After a 7-h control period, two serial injections of (27Nle)humanGRF-(l-29-)NH2 (2 /zg) were given 90 min apart (arrows). Results shown are the mean ± SEM.
Time (h) FIG. 8. Effects of continuous infusion of GRF on GH secretion in fasted guinea pigs. Blood samples were withdrawn at 10-min intervals from five male guinea pigs that had been fasted for 36 h before and 5 h during the study and assayed for guinea pig GH. After 1 h, a continuous iv infusion of (27Nle)human GRF-(l-29)-NH2 (12 Mg/h) was begun (D). Individual blood GH profiles are shown.
has only a minor effect on growth in the guinea pig (2, 4, 20), but the guinea pig does produce a GH molecule that can support growth in the hypophysectomized rat (21). We have recently extended our automatic bloodsampling method to the conscious guinea pig and, by isolating guinea pig GH and developing a sensitive heterologous RIA for this hormone, were able to report the first studies of the secretory pattern of GH in the guinea pig and its responses to exogenous GRF and SS (1). Since the guinea pig secretes GH, but does not apparently require this hormone for stimulating growth, we suggested that its physiological role may be metabolic rather than growth promotion in this species. There is some circumstantial evidence for a link between GH and insulin in the guinea pig. Insulin circulates at relatively high levels in normal guinea pigs (22), perhaps compensating for its relatively weak hypoglycemic potency compared to that of other mammalian insulins (23), and Clayton and Worden (20) noted that hypophysectomy in the guinea pig induced a transient
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 15:28 For personal use only. No other uses without permission. . All rights reserved.
GUINEA PIG GH SECRETION period of marked insulin hypersensitivity. We reasoned that if guinea pig GH was involved in metabolic control, it might be particularly sensitive to insulin and nutrition. Furthermore, it would be interesting to discover if this hystricomorph rodent responded to food withdrawal with decreased GH output, as in the rat, or increased GH release, as seen in most other species, including man. Our first studies showed that removal of food did indeed have a dramatic effect on GH secretion. Instead of an episodic pattern of GH release, with clusters of pulses interspersed with periods of low or undetectable GH release, fasting induced a sustained rise in baseline secretion; the large episodic pulses were replaced by a continuous irregular rapid secretory pattern in both sexes. Thus, the responses resembled those during fasting in man and other large animals (7-10) and contrasted directly with those in the rat, in which GH secretion is reduced by fasting (11). The increased GH output in response to fasting in the guinea pig is quite rapid (detectable changes occur within a few hours of food withdrawal) and is relatively long-lasting (requiring at least 24 h after refeeding for the normal episodic pattern to return). It is not clear what aspect of fasting (calorie intake, available hexoses, amino acids, and fatty acids) stimulated increased GH release. In man, oral glucose inhibits GRF-induced GH release (24); in the rat, FFA inhibit both spontaneous and GRF-induced GH secretion (25), although moderate hyperglycemia per se does not block spontaneous GH release (15, 25, 26). In the present studies glucose injections reduced GH secretion in the guinea pig to a limited extent by delaying GH pulses in normally fed animals and lowering the elevated GH baseline secretion in fasted animals. It is also possible that food withdrawal represents a stress stimulus, and it is notable that the GH response to stress also differs in rats and primates. In the rat, stress blocks spontaneous GH release, probably by an increase in SS secretion (27), whereas GH release is stimulated by stress in primates (28). We have not yet specifically investigated the effects of stress on GH secretion in the guinea pig. The GH responses to insulin injections in the guinea pig more closely resembled those in the rat (15), with a prompt blockade of secretory episodes in normally fed animals, although to our surprise, insulin was much less effective in suppressing the high basal GH output in fasted guinea pigs, which should be more susceptible to insulin-induced hypoglycaemia. It is possible that a higher insulin dose would have been more effective, but an insulin-induced rise in GH secretion, as seen in man, was never observed in the guinea pig even at the largest doses (6 U), which caused marked hypoglycemia in some normally fed animals. It should be noted, however, that porcine insulin was used for these studies, since the
721
amounts of purified guinea pig insulin available were not sufficient for these in vivo experiments. Although porcine insulin is a potent hypoglycemic agent in the guinea pig, we cannot exclude the possibility that the endogenous guinea pig insulin might have different effects on GH secretion. It is also conceivable that insulin could inhibit GH release indirectly, for example by stimulating the release of gut SS, although this is unlikely in view of the relatively large amounts of exogenous SS required to suppress GH secretion in the guinea pig (1). How does fasting cause such a marked alteration of secretion from episodic pulses to a high continuous GH secretory pattern in the guinea pig? In the rat, a similar increase in baseline GH can be brought about by blocking SS release immunochemically (29, 30), and it seems reasonable to assume that fasting could have reduced the release of SS or blocked its action at the guinea pig pituitary gland. That the latter explanation is unlikely was shown by infusing SS or injecting a long-acting SS analog (31); both maneuvers promptly lowered the elevated basal GH release in fasted guinea pigs. Another possibility is that fasting increases endogenous GRF secretion in the guinea pig. Although GH secretion is high during fasting, it was nowhere near maximal, since injections or infusions of exogenous GRF evoked a much greater GH output. In both rat and man, infusions of GRF amplify endogenous episodic GH release (32-34), but there was no evidence for an amplification of any underlying episodic secretion by continuous GRF in fasted guinea pigs. Since the episodic release of GH seems to be determined by the interplay between GRF release and an intermittent SS tone (35, 36), our working hypothesis is that fasting exerts its effects on GH release in the guinea pig via both SS and GRF. SS output is reduced and its secretory rhythm disrupted (accounting for the change from episodic to continuous high GH tone), and this may be accompanied by an increased release of GRF or a change in its secretory pattern from infrequent pulses to more frequent or even continuous release. One other influence that we have not addressed is the effect of insulin-like growth factor-I (IGF-I) in fasted guinea pigs. In other species, fasting lowers IGF-I levels markedly (37, 38), and since IGFTI can inhibit both the release (39) and synthesis (40) of GH, it is possible that fasting increases total GH output partly by disinhibition of the pituitary gland as IGF-I levels fall. It is also possible that porcine insulin could suppress GH secretion via IGF-I receptors, rather than insulin receptors, in the guinea pig. In conclusion, these studies confirm that the guinea pig presents an interesting model of GH secretion and control. The relevance to other species of nutritional regulation of GH in the rat is questionable, since the
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 15:28 For personal use only. No other uses without permission. . All rights reserved.
GUINEA PIG GH SECRETION
722
responses are opposite those seen in most other species; the guinea pig would seem to offer a useful alternative small laboratory animal model in this regard. The fact that this species does not use GH as a growth-promoting substance suggests that it might be a good species in which to study selectively the metabolic regulation of GH. The importance of the episodic secretory pattern of GH has been stressed in both experimental and clinical studies, with the emphasis largely on the effectiveness of a pulsatile GH pattern to stimulate growth (41). The normal guinea pig also shows a highly episodic GH secretory pattern, but the significance of this must presumably be directed to the metabolic, rather than growth-promoting, effects of this hormone. Perhaps, the guinea pig will help us to focus on the importance of the
14. 15. 16. 17. 18. 19. 20.
secretory pattern and actions of GH in adult life, where
21.
growth promotion is no longer the primary function of GH in any species.
22.
Acknowledgments
23.
We are grateful to Dr. A. L. Parlow for the continued supply of assay reagents which form the basis of the guinea pig GH assay. We thank Ms. Norrie Shannon for excellent technical assistance and Sten Rosberg, Goteborg University, for his version of the Pulsar program. We also thank Ferring AB (Malmo, Sweden) and Sandoz AG (Basel) for supplying the peptides used in this study.
24.
References
26.
1. Gabrielsson B, Fairhall KM, Robinson ICAF 1990 Growth hormone secretion in the guinea pig. J Endocrinol 124:371 2. Mitchell ML, Guillemin R, Selye H 1954 The effect of somatotropic hormone in the growth of normal and hypophysectomized guinea pigs. Endocrinology 54:111 3. Knobil E, Hotchkiss J 1964 Growth hormone. Annu Rev Physiol 26:47 4. Ponse K, Vecsey A 1970 Survie et croissance insolites de Cobayes femelles hypophysectomisees. Rev Suisse Zool 77:713 5. Davidson MB 1987 Effect of growth hormone on carbohydrate and lipid metabolism. Endocr Rev 8:115 6. Roth J, Glick SM, Yalow RS, Berson SA 1963 Hypoglycaemia: a potent stimulus to secretion of growth hormone. Science 140:987 7. Bassett JM 1974 Diurnal patterns of insulin, growth hormone, corticosteroid and metabolite concentrations in fed and fasted sheep. Aust J Biol Sci 27:167 8. Breier BH, Bass JJ, Butler JH, Gluckman PD 1986 The somatotrophic axis in young steers: influence of nutritional status on pulsatile release of growth hormone and circulating concentrations of insulin-like growth factor 1. J Endocrinol 111:209 9. Merimee TJ, Fineberg SE 1974 Growth hormone secretion in starvation-a reassessment. J Clin Endocrinol Metab 39:385 10. Ho KY, Veldhuis JD, Johnson ML, Furlanetto R, Evans WS, Alberti KGHM, Thorner MO 1988 Fasting enhances growth hormone secretion and amplifies the complex rhythms of growth hormone secretion in man. J Clin Invest 81:968 11. Tannenbaum GS, Rorstad O, Brazeau P 1979 Effects of prolonged food deprivation on the ultradian growth hormone rhythm and immunoreactive somatostatin tissue levels in the rat. Endocrinology 104:1733 12. Painson J-C, Tannenbaum GS 1985 Effects of intracellular glucopenia on pulsatile growth hormone secretion: mediation in part by somatostatin. Endocrinology 11:1132 13. Tannenbaum GS 1981 Growth hormone secretory dynamics in
25.
27. 28. 29.
30.
31. 32. 33.
34.
35. 36.
E n d o • 1990 Vol 127 • No 2
streptozotocin diabetes: evidence for a role for endogenous circulating somatostatin. Endocrinology 108:76 Carlsson LMS, Clark RG, Skottner A, Robinson ICAF 1989 Growth hormone and growth in diabetic rats: effects of insulin and insulin-like growth factor-I infusions. J Endocrinol 122:661 Tannenbaum GS, Martin JB, Colle E 1976 Ultradian growth hormone rhythm in the rat: effects of feeding, hyperglycemia and insulin-induced hypoglycaemia. Endocrinology 99:720 Clark RG, Chambers G, Lewin J, Robinson ICAF 1986 Automated repetitive microsampling of blood: growth hormone secretion in conscious male rats. Endocrinology 111:27 Clark RG, Carlsson LMS, Robinson ICAF 1987 Growth hormone secretory profiles in conscious female rats. J Endocrinol 114:399 Clark RG, Robinson ICAF 1988 Paradoxical growth promoting effects induced by patterned infusions of somatostatin in female rats. Endocrinology 122:2675 Merriam GR, Wachter KW 1982 Algorithms for the study of episodic hormone secretion. Am J Physiol 243:E310 Clayton BE, Worden JM 1960 Growth in young hypophysectomized guinea pigs. J Endocrinol 20:30 Knobil E, Greep RO 1959 The physiology of growth hormone with particular reference to its action in the rhesus monkey and the "species-specificity" problem. Recent Prog Horm Res 15:1 Gorray KC, Fujimoto WY 1980 "Micro"-insulin radioimmunoassay: measurement of the insulin response during glucose tolerance tests in guinea pigs. Proc Soc Exp Biol Med 163:388 Zimmerman AE, Moule ML, Yip CC 1974 Guinea pig insulin. II. Biological activity. J Biol Chem 249:4026 Davies RR, Turner S, Johnston DG 1984 Oral glucose inhibits growth hormone secretion induced by human pancreatic growth hormone releasing factor 1-44 in normal man. Clin Endocrinol (Oxf) 21:477 Imaki T, Shibasaki T, Masuda A, Hotta M, Yamauchi N, Demura H, Shizume K, Wakabayashi I, Ling N 1986 The effect of glucose and free fatty acids on growth hormone (GH)-releasing factormediated GH secretion in rats. Endocrinology 118:2390 King RA, Smith RM, Willoughby JO 1986 Effect of sustained hyperglycemia on growth hormone secretion in free-moving rats. Horm Metab Res 18:510 Arimura A, Smith WD, Schally AV 1976 Blockade of the stressinduced decrease in blood GH by anti-somatostatin serum in rats. Endocrinology 98:540 Quabbe JJ 1985 Hypothalamic control of GH secretion: pathophysiology and clinical implications. Acta Neurochir (Wien) 75:60 Tannenbaum GS, Epelbaum J, Cole E, Brazeau P, Martin JB 1978 Antiserum to somatostatin reverses starvation-induced inhibition of growth hormone but not insulin secretion. Endocrinology 102:1909 Terry LC, Martin JB 1981 The effects of lateral hypothalamicmedial forebrain stimulation and somatostatin antiserum on pulsatile growth hormone secretion in freely behaving rats: evidence for a dual regulatory mechanism. Endocrinology 109:622 Bauer W, Briner U, Doepfner W, Haller R, Hugenin R, Marbach P, Petcher TJ, Pless J 1982 A very potent and selective octapeptide analogue of somatostatin with prolonged action. Life Sci 31:1133 Wehrenberg WB 1986 Continuous infusion of growth hormone releasing factor: effects on pulsatile growth hormone secretion in normal rats. Neuroendocrinology 43:391 Vance ML, Kaiser DL, Evans WS, Furlanetto R, Vale W, Rivier J, Thorner MO 1985 Pulsatile growth hormone secretion in normal man during a continuous 24-hour infusion of human growth hormone releasing factor (1-40). J Clin Invest 75:1584 Brain C, Hindmarsh PC, Brook CGD, Matthews DR 1988 Continuous subcutaneous growth hormone releasing factor analogue augments growth hormone secretion in normal male subjects with no desensitization of the somatotroph. Clin Endocrinol (Oxf) 28:543 Tannenbaum GS, Ling N 1984 The interrelationship of growth hormone (GH)-releasing factor and somatostatin in the generation of the ultradian rhythm of GH secretion. Endocrinology 115:1952 Plotsky PM, Vale W 1985 Patterns of growth hormone releasing factor and somatostatin secretion into the hypophysial portal circulation of the rat. Science 230:461
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 15:28 For personal use only. No other uses without permission. . All rights reserved.
GUINEA PIG GH SECRETION 37. Clemmons DR, Klibanski A, Underwood LE, McArthur JW, Ridgeway EC, Beitins IZ, van Wyk JJ 1981 Reduction of plasma immunoreactive somatomedin-C during fasting in humans. J Clin Endocrinol Metab 53:1247 38. Maes M, Underwood LE, Ketelslegers JM 1984 Low serum somatomedin-C in protein deficiency: relationship with changes in liver somatogenic and lactogenic bindir o sites. Mol Cell Endocrinol 37:301 39. Abe H, Molitch ME, van Wyk JJ, Underwood LE 1983 Human growth hormone and somatomedin-C suppress the spontaneous
723
release of growth hormone in unanesthetized rats. Endocrinology 113:1319 40. Yamashita S, Melmed S 1986 Insulin-like growth factor I action on rat anterior pituitary cells: suppression of growth hormone secretion and messenger ribonucleic acid levels. Endocrinology 118:176 41. Jansson J-O, Albertsson-Wikland K, Eden S, Thorngren K-G, Isaksson O 1982 Circumstantial evidence for a role of the secretory pattern of growth hormone in control of body growth. Acta Endocrinol (Copenh) 99:24
DON'T MISS THE ENDOCRINE SOCIETY'S 42nd POSTGRADUATE ASSEMBLY A must for anyone involved in Endocrinology or Internal Medicine October 28-November 1, 1990 Sheraton Waikiki Honolulu, Hawaii For program and registration information please contact: The Endocrine Society 9650 Rockville Pike Bethesda, MD 20814 (301) 571-1802 FAX (301) 571-1869
The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 16 November 2015. at 15:28 For personal use only. No other uses without permission. . All rights reserved.
