Restraint, but not frustration, induces prostaglandin-mediated hyperthermia in pigs

Physiology& Behavior,Vol. 57, No. 6, pp. 1051-1055, 1995 Copyright © 1995ElsevierScienceLtd Printedin the USA. All rights reserved 0031-9384/95 $9.50 + .00

Pergamon 0031-9384(94)00343-2

Restraint, But Not Frustration, Induces Prostaglandin-Mediated Hyperthermia in Pigs R. F. P A R R O T 1 "l A N D D. M. L L O Y D

Department of Neurobiology, The Babraham Institute, Babraham Hall, Cambridge CB2 4AT, UK R e c e i v e d 26 July 1994 PARROTr, R. F. AND D. M. LLOYD. Restraint, but not frustration, induces prostaglandin-mediated hyperthermia in pigs. PHYSIOL BEHAV 57(6) 1051- 1055, 1995.--Three experiments were carried out to investigate stress hyperthermia in prepubertal pigs. Experiment 1 examined the effect of frustrative nonreward (psychological stress) on deep body temperature in animals (n = 7) trained to make operant responses for food following a 17.5-h period of deprivation. There was no change in body temperature when the feeders were switched off whereas there was a small increase (NS) during normal operant feeding that might be attributable to physical exertion. In Experiment 2, the effects of 15-min physical restraint (snaring) were examined in the same group of animals. This procedure induced a significant (p < 0.01) rise in core temperature that was completely abolished by prior administration of a cyclooxygenase inhibitor (indomethacin, 2 mg/kg given intravenously). The final experiment investigated the effects of snaring on plasma cortisol concentrations. Blood samples were taken from indwelling catheters in pigs (n = 5) subjected to 15-min restraint with, or without, indomethacin pretreatment. Snaring produced a significant (p < 0.001) increase in cortisol release that was not affected by the administration of indomethacin. These results suggest that snaring, a physical stress that may also have energy expenditure components, induces a prostaglandin-mediated hyperthermic response in the growing pig. Pigs

Frustration

Restraint

Hyperthermia

Cortisol

THE mechanisms involved in the febrile responses to immune challenges are currently the subject of intensive research. However, there are several nondisease situations (e.g., exercise, stress, ovulation) in which hyperthermia occurs in the normal animal (13). This poses the question as to whether similar physiological processes may mediate changes in thermoregulatory set-point under both pathologic and normal conditions. A topic of particular interest to physiologists is the hyperthermia that ensues when animals are exposed to stressors. In this connection, experiments using rodents indicate that psychological stress induced by fear of handling (30), exposure to an open field (24), cage switching (17), and water immersion (19) reliably increases core temperature. Furthermore, the fact that anxiety also induces hyperthermia in man (14) suggests that this response may be a general mammalian phenomenon. If this is so, then stress hyperthermia may also occur in ungulates; this possibility was investigated in the present study using prepubertal pigs subjected to emotional and/or physical distress. Psychological stress represents an emotional response to a non-nociceptive (physically neutral) stimulus whereas physical stress involves aversive or painful sensory input. In practice, however, physical stress is likely to involve elements of emotional disturbance. To induce psychological stress in the present experiment, a variant of the food frustration paradigm (5) was employed. This involved the nondelivery of reinforcements in certain animals of a group simultaneously making operant responses for food (1). In addition, in view of the effectiveness of

Prostaglandins

restraint as a means of applying physical stress (8), a procedure of this type was also used. Forms of restraint previously used for pigs include suspension in a harness (25), immobilization in a crush cage (11) or box (2), and holding in a supine position (3). However, one of the most effective procedures is a type of selfinduced restraint known as snaring (7,23). In this manipulation, a noose is slipped around the snout which causes the pig instinctively to pull backwards and tighten the knot. When the loose end of the cord is secured, the pig remains immobile with the cord taut until it is released. This technique, which is commonly used to restrain pigs in husbandry situations, stimulates the release of cortisol (7) and prolactin (10) and induces an opioiddependent hypoalgesia (23). Moreover, it has recently been demonstrated that snaring induces the formation of c-fos protein, an indicator of neuronal activation, in certain areas of the porcine brain (the supraoptic and paraventricular nuclei) known to be associated with the stress response (22). Therefore, snaring was used as the essentially physical stressor in this study. Immunological challenge with bacterial toxins induces macrophages to release cytokines of the intedeukin series. These then activate inducible cyclooxygenase pathways to produce prostaglandins that act on thermoregulatory neurons in the preoptic region of the forebrain to raise the set-point for body temperature (6,13). Evidence supporting the involvement of prostaglandins in febrile responses is provided by the well-established inhibitory action of cyclooxygenase inhibitors, such as indomethacin (4), on endotoxin-induced pyrexia. In addition, it has recently been

To whom requests for reprints should be addressed. 1051

1052

PARROTT AND LLOYD

shown that prostaglandin-induced fever in pigs is accompanied by c-fos expression in neurons of the median preoptic nucleus (27). Such findings are also pertinent to the phenomenon of stress hyperthermia because cyclooxygenase inhibitors given either peripherally (12,17,24) or centrally (12) reduce, but do not completely eliminate, the febrile response of rodents to emotional stress. The objectives of the present experiment were, first, to determine whether frustration or snaring would produce a hyperthermic response in prepubertal pigs. If such effects were demonstrable, then the second objective was to examine whether they could be antagonized by indomethacin. Finally, it was also considered of interest to determine whether any associated adrenocortical responses could be modified by administration of indomethacin.

In the third experiment, five of the animals received treatments (b) and (d) described above. Although no temperature records were taken, blood samples (10 ml) were collected from the pigs at 5-min intervals and held on ice. These were subsequently centrifuged and the resultant plasma was stored at -30°C pending radioimmunoassay for cortisol, using previously described methodology (21). The results used in the statistical analysis of the first two experiments were the changes in core temperature from initial (0 min) readings. Treatment effects in each experiment were contrasted using an analysis of variance (ANOVA) that compared the change in the area under the response curve (i.e., during and/ or posttreatment minus baseline) for each manipulation (21). All probabilities are presented as two-tailed values. RESULTS

METHOD

Seven Large White prepubertal pigs, weighing approximately 25 kg at the start of the study, were housed in separate metabolism cages where they learned to press switch panels with their snouts to obtain food or water. Each animal was surgically prepared under closed circuit halothane anaesthesia, using sterile precautions, with a catheter in the jugular vein to permit intravenous (IV) injections and the collection of blood samples. Each pig was also provided with a blind-ending catheter abutting the carotid artery to allow the introduction of a thermistor probe for the measurement of deep body temperature. Both catheters emerged in the neck region and were covered with a protective elastic bandage. The venous catheter was flushed daily with sterile heparinized saline and a small amount of liquid paraffin was placed in the carotid catheter to facilitate the insertion of the thermistor probe. Temperature was recorded at 5-min intervals using a digital meter (YSI Ltd, USA). According to normal practice in this laboratory, food (pig weaner pellets) was available on a fixed ratio of five presses to one reinforcement for an hour in the morning (1000-1100 h) and 30 min in the afternoon (1600-1630 h) whereas water was continuously available on a fixed ratio of two presses per reinforcement. The availabilityof food was signalled by a 30-s tone, which induces a rapid rate of operant responding that normally lasts for about 30-40 min [e.g., (20)]. In Experiment 1, however, this procedure was altered so that, on a given occasion, certain pigs received only a single food reinforcement while the remainder were allowed to eat a normal meal. Body temperature was recorded for 30 min before, and 45 min after, the tone in each animal under normal feeding and food frustration conditions. Although no record was made of the amount of food consumed in this study, all of the pigs were observed to eat normally when food was available and to become agitated when only a single reinforcement was delivered. Body temperatures in Experiment 2 were recorded over a 75min period while the pigs were given various treatments. These were administered at the same time of day for a given pig, although the testing sequence for individual animals differed. The treatments were (a) no stress (control), (b) 15-min snaring applied in the middle of the test period, (c) pretreatment with indomethacin, and (d) indomethacin pretreatment followed by snaring, as in (b). Indomethacin (Sigma UK Ltd) was dissolved in 4% NaHCO3 + 0.9% sterile saline and given at an IV dose of 2 mg/ kg 20 min before the start of the test. Attempts were made to control the intensity of the snaring stimulus by ensuring that the animals were always secured in the same position in the cage, and a minimum interval of 2 days separated snaring treatments in individual pigs.

Changes in body temperature during food reward or frustration (Experiment 1) are shown in Fig. 1. Contrary to expectations, there was no marked increase in temperature when food reinforcements were withheld. Furthermore, during normal operant feeding core temperature showed a nonsignificantincrease of approximately 0.3°C. The effects of snaring (SN) on core temperature are illustrated in Fig. 2. With the exception of some initial variation when the pigs were pretreated with indomethacin (IND), there was little change in temperature under any treatment condition in the first 30 min of the test. Snaring, however, induced an abrupt hyperthermic response that differed from the control (NIL) situation both during the 15 min of snaring (p < 0.02) and in the 30-min posttreatment interval (p < 0.003), as well as throughout the 45min period following the imposition of the stress (p < 0.004). In contrast, body temperature did not change under control or indomethacin conditions, and pretreatment with indomethacin (IND + SN) abolished the hyperthermia induced by snaring. In consequence, the response to snaring during, after, and throughout the posttreatment period differed from that observed after indomethacin (p < 0.02, p < 0.004, p < 0.006, respectively), as well as when indomethacin preceded snaring (p < 0.05, p < 0.01, p < 0.01, respectively). There was also a tendency for core temperature to decrease after indomethacin: this effect was almost significant (NIL + IND, p < 0.055) in the last 30 min of the test. Changes in plasma cortisol in snared pigs with (IND + SN) or without (SN) indomethacin pretreatment are displayed in Fig. 3. This dose of indomethacin, as previously established (Parrott and Vellucci, unpublished), produced an initial rise in cortisol concentrations, although, by the start of snaring (10 min), values were not significantlydifferent from those when the pigs received no pretreatment. During snaring, the animals showed an increase in cortisol above the baseline value taken at this 10-min sampling point (SNp < 0.001; IND + SN, p < 0.02). However, there was no difference between treatments at any time during or after the application of stress. DISCUSSION

In this study, emotional stress induced by frustative nonreward failed to alter core temperature in prepubertal pigs whereas restraint using a snare elicited a marked hyperthermia and an accompanying increase in plasma cortisol. The results further show that the effect of snaring on body temperature, but not on cortisol release, was totally suppressed by pretreatment with the cyclooxygenase inhibitor indomethacin. The majority of reports describing stress hyperthermia have used techniques considered by the researchers involved to induce

R E S T R A I N T A N D H Y P E R T H E R M I A IN PIGS

0.60 ¢_) 0 Ill

1053

L • FOOD t ~ NO FOOD

0.45

rr I'-,,=t: er

LI.I

0_

0.30

LLI I'->I--t

o

0.15

Z U.I

co

0.00

Z

,< -I" 0 Z

<

-0.15

LU =E

-0.30

--7-

i

I

5

~

15

I

'

25

I

I

I

'

35 45 TIME (min)

I

I

55

I

I

65

I

75

FIG. l. Changes in body temperature (mean _+ SEM) from an initial reading at 0 min in pigs before and during (indicated by the horizontal bar) a signalled period of food availablity. The animals were tested during normal operant feeding ( 1 ) and during extinction of the operant response following the delivery of a single food reinforcement (D).

0.60 0

0

,,'

er

0.45

NIL ~=- SN -:~--IND -¢-- IND+SN

I.-.<

IT

r'r

:S

0.3,0

LU / >,, E::)

o

'\\\

0.15

en

Z LU

o Z

0.00

,<

C..3 Z

<

W

-0.15

-0.30

I 5

'

I 15

~

I 25

'

I 35 TIME (min)

I 45

'

I 55

I

I 65

'

~-75

FIG. 2. Changes in body temperature (mean +_ SEM) from an initial reading at 0 rain in pigs before, during (indicated by the horizontal bar), and after restraint using a snare. The animals were tested under the following experimental conditions: no treatment (D NIL), snaring ( 1 SN), pretreatment with indomethacin ( 0 IND), indomethacin followed by snaring ( • IND + SN). Snaring produced a significant increase in temperature that was abolished by indomethacin pretreatment (see text for further details).

1054

PARROTT AND LLOYD 100

--

80

~

• SN - ~ - IND+SN

I I

6O

/

i

i

/'//,I /'/ //,f///

I

.._1

o ~-

40

--

20

--

0

I

I

I

0

5

10

I

r

15 20 TIME (min)

' 25

30

I

i

35

40

FIG. 3. Plasma cortisol concentrations(mean _+SEM) in pigs before, during (indicatedby the horizontalbar), and after snaring in pigs with and without pretreatment with indomethacin (labelling as for Fig. 2). Snaring significantlyincreased hormone concentrationsabove a baseline value taken at the lO-minsamplingpoint but indomethacinpretreatmentdid not affect this response (see text for further details).

psychological stress (see the Introduction). Unexpectedly, though, enforced extinction of operant responding for food in pigs fasted for 17.5 h failed to alter body temperature. However, whereas a similar protocol increased plasma corticosteroids in pigs sampled by venepuncture (5), an adrenocortical response was not detected in catheterized animals tested after a longer period of food deprivation (1). Moreover, the situation is further complicated by the observation that anticipation of operant feeding is, as judged by the corticosteroid response, mildly stressful in 17-h starved pigs (9). Thus, it may be tentatively concluded that although the pigs in the present study vocalized and were apparently emotionally affected by the nondelivery of food, the procedure was insufficiently stressful to induce a hyperthermic response. By contrast, body temperature seemed to increase in Experiment 1 during normal operant feeding. It is unlikely that this change was due to a calorific effect of the food because the greatest increase (+0.3°(2) was observed while some of the pigs were still eating. In this connection, it is relevant that running in an exercise wheel induces hyperthermia after about the same time interval (30 min) in rats (26). There is little doubt that snaring is a particularly effective form of stress in the pig (7,10,23,27). The present results also demonstrate that this form of restraint induces a marked hyperthermia that gradually declines after the animal is released. When a pig is snared, the animal tugs backwards and subsequently appears to enter a semihypootic state that probably coincides with the hypoalgesia that occurs about 5 min after snaring (23). Therefore, although it is obvious that snaring represents a painful physical stressor, it also results in a degree of muscular exertion. In consequence, the increase in temperature may be due to stress, exercise, or, more probably, some combination of the two. This finding, however, is interesting because restraint reportedly in-

duces hypothermia in rabbits (18). Whether these divergent results are due to the use of different species, or to the precise nature of the stressor employed, is not clear because the effects of physical, as opposed to psychological, stress on body temperature have been little studied. Nevertheless, it has been found that restraint stimulates the production of interleukin-6 in rats (31) and this cytokine is known to have febrile effects (13). The involvement of prostaglandin synthesis in the hyperthermia induced by stress (12,13,18,24) or exercise (26) suggests that the response involves a change in thermoregulatory set-point and is, therefore, similar to that observed during fever. However, because peripheral or central administration of a cyclooxygenase inhibitor only partially reversed the pyrexia in those experiments, it was necessary to conclude that other factors might be involved. In contrast, by giving indomethacin IV rather than using the other routes adopted in previous studies, it proved to be possible to inhibit completely the hyperthermic response. This indicates that snaring activates the induced cyclooxygenase (Cox-2) pathway (16), resulting in the production of endogenous prostaglandins that act on preoptic neurons (27) to raise the thermoregulatory set-point. In addition, the present investigation showed that IV indomethacin also tended to lower body temperature, as originally reported in normal and pyrogen-treated rats (4). This suggests that the noninducible (constitutive) cyclooxygenase pathway (Cox-l); (16) may play a role in the maintenance of basal body temperature (29). The increase in cortisol concentrations during snaring supports previous findings (7) and is consistent with the observation that this stressor activates parvocellular neurones in the porcine paraventricular hypothalamic nucleus (PVN); (27). However, the cortisol response to snaring was not abolished by indomethacin. In this connection, it may be relevant that the febrile effects of a

R E S T R A I N T A N D H Y P E R T H E R M I A IN PIGS

1055

high dose of endotoxin in another ungulate, the goat, were prevented by IV injection of a cyclooxygenase inhibitor whereas the increase in cortisol was only slightly reduced (15). This compares with findings in the rat where indomethacin inhibits A C T H release during stress hyperthermia (18) but does not prevent activation of hypophysiotrophic P V N neurons in animals given foot shock (28). Therefore, the extent to which Cox-2 pathways are involved in stimulating pituitary/adrenocortical responses during stress hyperthermia may depend on the type of stressor, and may also vary between species. Clearly, further studies are required to resolve these issues.

In conclusion, the present results indicate that restraint induces p r o s t a g l a n d i n - m e d i a t e d h y p e r t h e r m i a in growing pigs. M o r e o v e r , m e a s u r e m e n t o f body t e m p e r a t u r e m a y provide a n o n i n v a s i v e m e a n s o f asssessing stress r e s p o n s e s in farm animals u n d e r c o n d i t i o n s w h e r e their welfare could be c o m p r o mised. ACKNOWLEDGEMENTS The authors are grateful to J. A. Goode for measuring plasma cortisol concentrations and to D. Brown for carrying out the statistical analysis.

REFERENCES 1. Baldwin, B. A.; Forsling, M. L.; Parrott, R. F. Enhanced vasopressin release in pigs as an indicator of stress. J. Endocrinol. 121:140; 1989 (abstract). 2. Becker, B. A.; Nienaber, J. A.; Christensen, R. K.; Manak, R. C.; De Shazer, J. A.; Hahn, G. L. Peripheral concentrations of cortisol as an indicator of stress in the pig. Am. J. Vet. Res. 46:1034-1038; 1985. 3. Brown-Borg, H. M.; Klemcke, H. G.; Blecha, F. Lymphocyte proliferative responses in neonatal pigs with high or low plasma cortisol concentration after stress induced by restraint. Am. J. Vet. Res. 54:2015-2020; 1993. 4. Clark, W. G.; Cumby, H. R. The antipyretic effect of indomethacin. J. Physiol. (Land.) 248:625-638; 1975. 5. Dantzer, R.; Arnone, M.; Morm&le, P. Effects of frustration on behaviour and plasma corticosteroid levels in pigs. Physiol. Behav. 24:1-4; 1980. 6. Dinarello, C. A.; Wolff, S. M. The role of interleukin-1 in disease. N. Engl. J. Med. 328:106-113; 1993. 7. Farmer, C.; Dubreuil, P.; Couture, Y.; Brazeau, P.; Petitclerc, D. Hormonal changes fol]lowing an acute stress in control and somatostatin-immunized pigs. Dam. Anim. Endocrinol. 8:527-536; 1991. 8. Glavin, G. B.; Part, W. P.; Sandbak, T.; Bakke, H. K.; Murison, R. Restraint stress in biomedical research: An update. Neurosci. Biobehav. Rev. 18:223-249; 1994. 9. Houpt, K. A.; Baldwin, B. A.; Houpt, T. R.; Hills, F. Humoral and cardiovascular responses to feeding in pigs. Am. J. Physiol. 244:R279-R284; 198L 10. Klemcke, H. G.; Nienaber, J. A.; Hahn, G. L. Stressor-associated alterations in porcine plasma prolactin. Proc. Sac. Exp. Biol. Med. 186:333-343; 1987. 11. Klemcke, H. G. Responses of the porcine pituitary-adrenal axis to chronic intermittent stressor. Dam. Anim. Endocrinol. 11:133-149; 1994. 12. Kluger, M. J.; O'Reilly, B. O.; Shape, T. R.; Vander, A. J. Further evidence that stress hyperthermia is a fever. Physiol. Behav. 39:763766; 1987. 13. Kluger, M. J. Fever: Role of pyrogens and cryogens. Physiol. Rev. 71:93-127; 1991. 14. Marazziti, D.; di Muro, A.; Castrogiovanni, P. Psychological stress and body temperature changes in humans. Physiol. Behav. 52:393395; 1992. 15. Massart-Le~n, A. M.; Burvenich, C.; Vandeputte-Van Messom, G.; Hilderson, H. Partial prostaglandin-mediated mechanism controlling the release of oortisol in plasma after intravenous administration of endotoxins. Dam. Anim. Endocrinol. 9:273-283; 1992. 16. Mitchell, J. A.; Akarasereenant, P.; Thiermermann, C.; Flower, R. J.; Vane, J. R. Selectivity of non-steroidal antiinflammatory drugs as inhibitors of constitutive and inducible cyclooxygenase. Prec. Natl. Acad. Sci. USA 90:11693-11697; 1993. 17. Morimoto, A.; Watanabe, T.; Morimoto, K.; Nakamori, T.; Murakami, N. Possible involvement of prostaglandins in psychological

18. 19.

20.

21.

22. 23. 24. 25.

26. 27.

28.

29.

30. 31.

stress-induced responses in rats. J. Physiol. (Land.) 443:421-429; 1991. Morimoto, A.; Watanabe, T.; Myogin, T.; Murakami, N. Restraint stress elicits acute-phase response in rabbits. Pflugers Arch. 410:554-556; 1987. Nakamori, T.; Morimoto, A.; Morimoto, K.; Tan, N.; Murakami, N. Effects of a- and/~-adrenergic antagonists on rise in body temperature induced by psychological stress in rats. Am. J. Physiol. 264:RI56-R161; 1993. Parrott, R. F.; Ebenezer, I. S.; Forsling, M. L. The effects of intravenous administration of cholecystokinin on feeding behaviour and release of pituitary hormones in pigs are not mediated by serotonergic (5-HT3) receptors. Neuropharmacology 31:863-867; 1992. Parrott, R. F.; Goode, J. A. Effects of intracerebroventricular corticotropin-releasing hormone and intravenous morphine on cortisoi, prolactin and growth hormone secretion in sheep. Dam. Anim. Endocrinol. 9:141 - 149; 1992. Parrott, R. F.; Vellucci, S. V. Stress-induced changes in c-fos immunoreactivity in the porcine brain. Br. Vet. J. 150:355-363; 1994. Rushen, J.; Ladewig, J. Stress-induced hypoalgesia and opioid inhibition of pigs' responses to restraint. Physiol. Behav. 50:10931096; 1991. Singer, R.; Harker, C. T.; Vander, A. J.; Kluger, M. J. Hyperthermia induced by open-field stress is blocked by salicylate. Physiol. Behav. 6:1179-1182; 1986. Stephens, D. B.; Perry, G. C. The effects of restraint, handling, simulated and real transport in the pig (with reference to man and other species). Appl. Anim. Behav. Sci. 28:41-55; 1990. Tanaka, H.; Yanase-Fujiwara, M.; Kanaosue, K. Effects of centrally and systemically administered indomethacin on exercising rats. Am. J. Physiol. 265:R230-234; 1993. Vellucci, S. V.; Parrott, R. F. Hyperthermia-associated changes in Fos protein in the median preoptic and other hypothalamic nuclei of the pig following intravenous administration of prostaglandin F<. Brain Res. 646:165-169; 1994. Wan, W.; Wetmore, L.; Sorensen, C. M.; Greenberg, A. H.; Nance, D. M. Neural and biochemical mediators of endotoxin and stressinduced C-fos expression in the rat brain. Brain Res. Bull. 34:7-14; 1994. Weidenfeld, J.; Siegel, R. A.; Conforti, N.; Feldman, S.; Chowens, I. Site and mode of action of indomethacin on the hypothalamohypophysol-adrenal axis: A temporal study in intact, hypothalamicdeafferentated, and hypothalamic-lesioned male rats. Endocrinology 109:205-209; 1981. Zethof, T. J. J.; van der Heyden, J. A. M.; Tolboom, J. T. B. M.; Olivier, B. Sress-induced hyperthermia in mice: A methodological study. Physiol. Behav. 55:109-115; 1994. Zhou, D.; Kusnecov, A. W.; Shurin, M. R.; De Paoli, M.; Rabin, B. S. Exposure to physical and psychological stressors elevates plasma interleukin 6: Relationship to activation of hypothalamicpituitary-adrenal axis. Endocrinology 133:2523-2530; 1993.