- Email: [email protected]
Comparative biochemical responses of rats to different stressful stimuli
Physiology & Behavior, Voi. 34, pp. 595--599.Copyright©Pergamon Press Lid., 1985. Primed in the U.S.A.
0031-9384/85 $3.00 + .00
Comparative Biochemical Responses of Rats to Different Stressful S t i m u l i 1 MAURICIO
R. O D I O 2 A N D R O G E R P. M A I C K E L
Department o f Pharmacology and Toxicology, School o f Pharmacy and Pharmacal Sciences Purdue University, West Lafayette, I N 47907 R e c e i v e d 16 J a n u a r y 1984 ODIO, M. R. AND R. P. MAICKEL. Comparative biochemical responses of rats to different stressful stimuli. PHYSIOL BEHAV 34(4) 595-599, 1985.--Adult male rats were exposed to single applications of one of three stressful stimuli (low environmental temperature, immobilization, random footshock) for periods up to 4 hours and plasma levels of corticosterone (PCS), fatty acids (PFA), and glucose (PGL) were determined at various points during the stress exposure and 1 and 2 hours post-exposure. The levels of PCS were increased by all 3 stressful stimuli in a similar temporal pattern, with the greatest magnitude of effect seen for immobilization and the least for cold exposure. The time courses of increased PFA levels were similar for immobilization and cold exposure; the response to foot shock was delayed in onset by 2 hours. The PGL response was minimal for cold exposure and foot shock, but showed a marked elevation during the first 2 hours of immobilization. The results suggest that the response pattern obtained is characteristic of the stressful stimulus employed, with PCS showing the least degree of specificity. Stressful stimuli Plasma fatty acids
Cold exposure Immobilization Plasma glucose
Random footshock
Plasma corticosterone
ferential patterns of responses by Lenox et al. [18], who showed varied responses of brain cGMP, pituitary c A M P and circulating prolactin levels in rats immobilized or exposed to cold. Keim and Sigg [12] found that the temporal pattern o f responses, i.e., changes in hypothalamic N E and PCS observed in rats exposed to a novel environment, 2 mA electric footshock, or restraint, were not similar. Even as subtle a difference as whether footshock was applied to rats according to a fixed or variable schedule resulted in differential time courses for PCS responses [1]. While single time point measurements may suggest that the hypothalamicpituitary adrenocortical axis responds in a non-specific manner to different stressful stimuli, examination o f temporal response "profiles" supports an opposite conclusion. Thus, Vernikos-Dannellis and Heyback [31] stress the importance of time-course measurements to describe response phenomena, rather than single time point values. In terms o f the overall " r e s p o n s e " to stressful stimuli, two recent reports may be o f particular relevance. Ursin [30] examined the relationship between various types o f personalities and the involvement o f response networks in the experience o f parachute jumping. The results show three response groups: one with a predominantly adrenal medullary (E plus NE) response, a second with a predominantly adrenocortical (PCS) response, and a third with a response mainly in plasma fatty acids (PFA). Henry [10] reviewed this and other data and concluded that differential responses
S E L Y E [26] first reported reactions o f higher organisms to the application o f noxious or threatening stimuli, and later characterized these responses to stress as the General Adaptation Syndrome [27]. Numerous subsequent publications have contradicted and supported various aspects o f the Selye hypothesis. In 1971, Mason [24] reported that initial exposure to virtually any unfamiliar stimulus evoked an increase in pituitary-adrenocortical activity as measured by elevated levels of plasma corticosterone (PCS), while animals acclimated to the various experimental procedures demonstrated differential PCS responses to the stressors. Forberg et al. [6] reported elevated urinary excretion of epinephrine (E) and norepinephrine (NE) in normal human volunteers viewing either pleasant, comic movies or fearful, rage-eliciting films. These questions of the nonspecificity of stress responses are interesting since the pituitaryadrenocortical, adrenal medullary and post-ganglionic sympathetic nervous systems are all presumably included in the overall " r e s p o n s e " to stress as defined by Selye's G.A.S. [27]. Mason [25] concluded that stimulus-specific responses o f the organism could be confounded by biochemical phenomena evoked by the heightened state of arousal induced in many experimental conditions; thus, the "stimulus-independent" response o f Selye could be the result of cognitive-emotional factors that would confound the experimental results. Different stressful stimuli have been shown to elicit dif-
~Research supported in part by a LASPAU-University of Costa Pica Graduate Scholarship and by a Purdue University David Ross XR Fellowship. 2Taken in part from a thesis submitted in partial fulfillment of the requirements for the Ph.D. degree, Purdue University, 1983.
595
596
ODIO A N D M A I C K E L
such as PCS or plasma E or NE could be correlated with both the type of stimulus and the background and experience of the experimental subjects. In the present report, plasma corticosterone (PCS) levels have been selected to characterize the hypothalamopituitary-adrenocortical system [29]; plasma fatty acid (PFA) levels have been selected to characterize the sympathetic nervous-adipose tissue system [2]; and plasma glucose (PGL) levels have been selected to characterize the adrenomedullary-liver system [7,32]. While no one of these parameters represents the optimal choice, characterization of the time-response curves resulting from activation by several stressful stimuli may yield a fingerprint of the stress response for each stimulus [31]. Three stressful stimuli, sufficiently different in nature to enhance the probability of producing stimulus-specific response patterns, were used. Exposure to an environmental temperature of 4°C (CLD) was selected as a primarily physiological stressor; all three response systems are clearly involved [19,20]; physical immobilization (IMM) was selected as a primarily psychological stressor [15,16]; again involving all three response systems [15,16]; and inescapable random electric footshock (ESK) was selected as a third stressful stimulus, one which combined physiological (slight discomfort) with psychological (an element of anticipation) characteristics and also differed from CLD and IMM in being discontinuous. E S K has been reported to induce responses in PCS [1], P F A [22], and P G L [5].
CLD
6O
J4k-'~L .... - - a k
8
!m - ....
,,,, "~\
!
....... o ........
ESK
---¢~----
IMM
............. ~:,,.
O
-',
if) c~ O c3
k.
X
03 0_ 10
%
I I
1 I 2 5 STRESSFUL STIMULI
Lt/ 4" /:
I I RECOVERY
I 2 (HOURS) ._
FIG. 1. Response of plasma corticosterone (PCS) to stressful stimuli. Each point in the mean of values obtained from 6 rats exposed to each stressful stimulus, except that t=0 points are the mean of values obtained from 12 rats (2 killed at each corresponding time point) as described in the Method section. Solid symbols represent values that differ significantly from t=0 values for each stressful stimulus (0<0.05).
METHOD
Subjects
Adult, male Sprague-Dawley-derived rats (total N = 144) weighing 200-225 g at delivery were purchased from Murphy Breeding Laboratories, Plainfield, IN. They were group housed (6/cage) in stainless steel cag_es (60x 57 x 50 era) on an ad lib diet of Wayne " L a b l o x " and tap water in an animal room maintained at 21-25°C with a 14/10 lighting cycle (on at 0600, off at 2000 hr). The rats were allowed a 9 day acclimation period prior to experimental use.
Chemicals
All chemicals used in the assay procedures were analytical reagent quality purchased from commercial suppliers. Heparin sodium was purchased from Abbott Laboratories, North Chicago, IL. Chemical standards used were: corticosterone and palmitic acid (Sigma Chemical Co., St. Louis, MO) and glucose (Beckman Instruments, Chicago, IL). Procedures
Prior to collection of blood samples, 50 ml beakers and 13 ml disposable borosilicate glass culture tubes were rinsed with a solution containing 125 U heparin sodium/ml 0.9% NaCI. The tubes were emptied, inverted for 10 min, then chilled in ice. Animals were killed by decapitation and trunk blood was collected into a chilled beaker, transferred to a chilled tube and centrifuged (2,000 rpm for 15 min) at 40°C. Following centrifugatmn, the tuoes were again placed in ice while four aliquots of plasma were transferred to sterile, plastic stoppered cryotubes for storage at -70°C until assay. Prior to assay, each tube was thawed at room temperature and mixed by gentle agitation to insure a homogeneous sample. Each cryotube was used for a single assay (i.e.,
PCS, PFA, or PGL) with replicate sampling; the fourth tube was maintained as a spare. Plasma corticosterone (PCS) was determined by the fluorometric method of Guillemin et al. [9] with activation and emission wavelengths of 467 and 525 nm, respectively. Plasma fatty acids (PFA) were determined by the colorimetric method of Duneombe [4] using the absorbance peak at 440 nm. Plasma glucose (PGL) was determined directly in a Beckman Model I Glucose Analyzer, using the glucose oxidase procedure o f Kadish et aL [11]. Cold exposure (CLD) was applied by placing individual rats in stainless steel cages (27x20x 17 cm) with wire mesh floors located in a walk in cold room maintained at 2.5-5.5°C with continual air circulation. Immobilization (IMM) was applied by placing individual rats in a metal cage (30x 15 x 17 cm) with screen wire sides and a moveable upper lid capable of continuous (thumbscrew) adjustment in the vertical plane. Pressure was applied to the animal until movement was completely restricted; at this point respiration was not impeded and no physical damage was produced. Electric footshock stress (ESK) was applied by placing individual animals in Lehigh Valley operant chambers; the grid floor was connected to a Grason Stadler shock generator supplying 0.5 mA shocks with a duration of 2.0 sec on a randomized schedule of 6 shocks per 60 rain. Control animals were placed in individual plastic cages (27x 18x 13 cm) for time periods comparable to the stressed animals. No food or water was available to the animals during the stress periods. Each test group consisted of 6 rats exposed to a given stress and 2 additional rats as controls. The 6 sets of controls were not significantly different from each other; consequently, they were pooled to yield a total N=12. Stressful stimuli were applied for 0.5, 1.0, 2.0 or 4.0 hr; additional groups were exposed to stressful stimuli for 4 hr, then placed
RESPONSES TO S T R E S S F U L STIMULI
597
TABLE 1 AREA UNDER THE CURVE (AUC) VALUES FOR RESPONSES TO STRESSFUL STIMULI IN RATS
Stressful Stimulus
Onset
Maintenance
Recovery
PCS 0zg-hr)
NONE CLD ESK IMM
41 72 81 103
34 69 96 100
38 42 56 60
PFA (/xEq-hr)
NONE CLD ESK IMM
151 268 161 313
202 308 275 344
245 296 372 267
PGL (mg-hr)
NONE CLD ESK IMM
321 323 327 431
324 321 311 336
308 370 284 321
Values were determined as described in the Method section.
in individual cages (similar to controls) for an additional 1.0 or 2.0 hr prior to killing.
Statistical Analysis Data is presented as mean-+S.E.M, for individual groups. Statistical analyses was by appropriate ANOVA with post hoc comparisons by the two tail Dunnett's test; the criteria for significance was set at p<0.05 for all comparisons. The construct of Area Under the Curve (AUC) has been utilized to make some of the comparisons. These values were obtained as described by Gibaldi and Perrier [8]. Baseline values were determined from the control data; experimental data were plotted as obtained for the time points. Time segments were divided into: 0-2 hours of stress (onset), 2-4 hours of stress (maintenance), and 0-2 hours post stress termination (recovery). RESULTS
The time course of response for PCS for each of the three stressful stimuli is presented in Fig. 1. All of the stressful stimuli evoked a positive response peaking at 0.5 hr and persisting at significantly elevated values throughout the 4 hr of application of each stressful stimuli. By 1 hr after termination, values for PCS in CLD and ESK animals had returned to normal, while those of IMM rats were still significantly elevated. IMM also resulted in higher values during stress. The comparative potencies of the three stressful stimuli may be estimated from the area under the curve (AUC) data obtained from these time-course experiments and presented in Table 1. During the onset and maintenance periods, these values totaled 203, 177 and 141/~g-hr for IMM, ESK, and CLD, respectively, as compared to 75/~g-hr in the control rats.
CLD
........ o ........
ESK
---A----
IMM
2OO
Phase of Stressful Stimulus Response Parameter
•- - o
.
E o_
/
0
~ k~
15o
_o
i
.
.
.
~/ . . . . .
~
"'~IA
~
•
....
""
....-" ~ I / ~ .'" ,,."
3
\ ~
50
O~
I I
|
I
2 3 STRESSFUL STIMULI
b/ 4"
I I RECOVERY
I 7' (HOURS}
FIG. 2. Response of plasma fatty acids (PFA) to stressful stimuli. Each point is the mean of values obtained from 6 rats exposed to each stressful stimulus, except that t =0 points are the mean of values obtained from 12 rats (2 killed at each corresponding time point) as described in the Method section. Solid symbols represent values that differ significantly from t=0 values for each stressful stimulus (0<0.05).
The time course of responses for PFA is presented in Fig. 2. Each of the stressful stimuli evoked a different pattern of response in this parameter. For example, CLD significantly elevated PFA values for the entire duration of the stressexposure; this elevation persisted for at least two hours post cessation of the stress. The PFA response to IMM was significantly elevated during the 4 hr application of the stressful stimulus, persisted for 1 hr post termination, and returned to normal values in the next hr. The response of PFA levels to ESK was unique. No response was seen until the 4 hr time point, but this significant elevation persisted throughout the 2 hr post-stress period. AUC values for PFA (Table 1) reflected the response to each of the stressful stimuli. For the onset phase,/zEq-hr values were 313,268, and 161 for IMM, CLD, and ESK, respectively, with a control value of 151. For the maintenance phase, the order was similar; 344, 308, and 275 for IMM, CLD, and ESK, respectively, with a control value of 202/zEq-hr. In the post-stress phase, only ESK rats deviated from control values. Figure 3 presents the time-course of data for PGL in response to the stimuli; 3 differential effects were seen. The most significant effect was that of IMM which evoked a marked elevation of PGL over the first 2 hours (the onset phase). In contrast, ESK caused a small, but statistically significant elevation of PGL levels at 1 and 2 hr of stress, and CLD evoked small but significant elevations of PGL at 0.5 and 4 hr of stress. These effects are borne out by the AUC values in Table 1, where the only deviations from control were those seen with IMM during the onset phase. DISCUSSION
In order to best consider the characteristics of timeresponse curves obtained, a triphasic system has been used.
598 The first 2 hours of exposure to each stressful stimulus is considered as the " o n s e t " phase, during which the animal is exposed to a novel situation. Presumably, little adaptation of biochemical mechanisms occurs during this period. The next 2 hours of exposure are considered the "maintenance" phase, and presumably sees the initiation of adaptation phenomena. Finally, the 2 hours after cessation of the stress period (the " r e c o v e r y " phase) offers the biological systems of the organism an opportunity to return to normal. Using this approach, the responsiveness of the hypothalamo-pituitary-adrenocortical system, as reflected in the PCS levels, showed the least degree of specificity in terms of differential responding to the 3 stressful stimuli. Some differences in the magnitude and duration of the PCS response were observed, presumably reflecting differential severity o f the stressful stimuli, but, in general, these data would support some previously advocated contentions that PCS is a non-specific response [21,27]. However, the differential decrements in PCS levels over time seen with the three stressors are in agreement with the findings of Bassett and Cairncross [1] and Keim and Sigg [12]. The somewhat longer persistence of the PCS levels in the animals exposed to IMM can be explained on the basis of a greater magnitude of hypothalamo-pituitary activity during the 4 hr stress exposure [14]. A contrasting situation is observed in the case of the sympathetic nervous system-adipose tissue axis, in which stimulation is manifested by increased lipolytic activity with consequent elevation o f P F A levels. Three different response patterns were seen. Animals exposed to the E S K stress showed no change in P F A levels until the 4 hr stress point; the elevated levels seen at this time were maintained throughout the 2 hr of the recovery phase. In contrast, both CLD and IMM stress caused significant elevations in P F A levels from 30 minutes of stress through the In'st hour of the recovery phase; however, by the second hour of the recovery phase, the P F A levels in the IMM rats had fallen back to levels near those o f controls, while those of the CLD rats remained somewhat elevated from the value of the controls. A confounding variable for CLD may be an increased utilization of P F A by the rat in order to facilitate maintenance of a normothermia in the face of a severe temperature gradient between animal and environment [2], additionally, one must consider the fact that all animals were without food for 6 hours [28]. The unique response to E S K , a response similar to that seen by Mallow and Witt [22] may reflect the fact that once adipose tissue lipase is activated by increased sympathetic nervous activity, the enzyme has a sufficiently lengthy half-life in the activated form to result in a persistent lipolysis [2,28]. The delayed onset o f this response is, however, inexplicable at this time. Finally, the effects of the three stressful stimuli on PGL levels yields results that tend to support the concept of specific response patterns to specific stresses as well. Rats exposed to C L D stress show minimal changes in PGL values; the increases observed at 0.5 and 4.0 hr are statistically significant although they represent less than a 10% increase above baseline. In contrast, exposure of rats to E S K leads to
ODIO A N D M A I C K E L
250[.-
/,,&~
'\,,,
,,¢
~ 2OO ~-
~o
CLD
....... [] .......
EsK
---"¢~--- IMM
",,
'
+
" .............. q" .....
" ......... o ................ o
g <
I00
g.
50
Oo
I
I
I
I
.~//
2 3 ,~ STRESSFUL STIMULI
I
I RECOVERY
I
2 (HOURS)
FIG. 3. Response of plasmaglucose (PGL) to stressful stimuli. Each point is the mean of values obtained from 6 rats exposed to each stressful stimulus, except that t=0 points are the mean of values obtained from 12 rats (2 killed at each corresponding time point) as described in the Method section. Solid symbols represent values that differ significantly from t=0 values for each stressful stimulus (p<0.05).
significant elevations in PGL levels at 1.0 and 2.0 hr; these values are about 20% above control levels. The effects of exposure to IMM are unique, as PGL levels are elevated by more than 50% at the 0.5 hr time point, become even more elevated at 1.0 hour and do not return to normal values until 1.0 hr post termination of the stress session. This responsivity to IMM probably reflects both an increased activity of the sympatho-adrenal medullary system [16, 17, 32] and a reduced utilization of PGL by the immobilized, and therefore inactive, rat [13]. The CLD rats would not be expected to show an increased PGL since they are utilizing glucose at a higher rate [23]; the E S K rats presumably have a more moderate response since they are more active (jumping in response to the shock) despite increased sympathomedullary output [3]. Similarly, some portion of the increased PGL levels after 0.5-2.0 hr of IMM may reflect a decreased utilization due to decreased muscular activity [13]. Thus, it seems rather apparent that the three stressful stimuli are capable of eliciting clearly differential response patterns even though these may be slightly confounded by other factors influencing the biochemical parameters being measured. This should not be surprising since the stimuli represent different blends o f continuity (CLD, IMM) and discontinuity (ESK), of psychological components (ESK, IMM) and physiological components (CLD), of activity (CLD, ESK) and inactivity (IMM).
REFERENCES 1. Bassett, J. R. and J. D. Cairncross. Time course for plasma 2. Brodie, B. B., R. P. Maickel and D. N. Stem. Autonomic nerv1l-hydroxycorticosteroid elevation in rats during stress. Pharous system and adipose tissue. In: Handbook of Physiology: macol Biochem Behav 3: 139--142, 1975. Adipose Tissue, edited by A. E. Renold and G. F. Cahill. Baltimore: Waverly Press, 1965, pp. 583-600.
RESPONSES TO STRESSFUL
STIMULI
3. DaPrada, M., L. Pieri and G. B. Picotti. Effect of midazolam (a water soluble benzodiazepine) on stress-induced increase of plasma catecholamines. In: Catecholamines and Stress: Recent Advances, edited by E. Usdin, R. Ketnansky and I. J. Kopin. New York: Elsevier/North Holland, 1980, pp. 231-236. 4. Duncombe, W. G. The colorimetric microdetermination of nonesterified fatty acids in plasma. Clin Chim Acta 9: 122-125, 1964. 5. Ehrentheil, O. F., L. J. Reyna, C. J. Adams, T. J. Giovanniello and E. T. Chert. Blood sugar response to electric shock stress in normal, adrenalectomized and hypophysectomized white rats. Diabetes 16: 325-330, 1967. 6. Froberg, J., C. O. Karisson, L. Levi and L. Lindberg. Physiological and biochemical stress reactions induced by psychosocial stimuli. In: Society, Stress and Disease, vol 1, edited by L. Levi. London: Oxford University Press, 1971, pp. 280-295. 7. Ganong, W. F. Review of Medical Physiology. San Francisco: Lange Medical Publications, 1975. 8. Gibaldi, M. and D. Perrier. Pharmacokinetics. New York: Marcel Dekker, 1975. 9. Guillemin, R., G. W. Clayton, H. S. Lipscomb and J. Smith. Fluorimetric measurement of rat plasma and adrenal corticosterone concentrations. J Lab Clin Med 53: 830-832, 1959. 10. Henry, J. P. Controversy in the stress field. In: Catecholamines and Stress, Recent Advances, edited by E. Usdin, R. Kvetnansky and I. J. Kopin. New York: Elsevier/North Holland, 1980, pp. 557-571. 11. Kadish, A. H., R. L. Little and J. C. Sternberg. A new and rapid method for the determination of glucose by measurement of the rate of oxygen consumption. Clin Chem 14:116-131, 1968. 12. Keim, K. L. and E. B. Sigg. Physiological and biochemical concomitants of restraint stress in rats. Pharmacol Biochem Behav 4: 289-297, 1976. 13. Koslowsld, S. Panel discussion on stress theory. In: Catecholamines and Stress: Recent Advances, edited by E. Usdin and I. J. Kopin. New York: Elsevier/North Holland, 1980, p. 854. 14. Krieger, D. T. The hypothalamus and neuroendocrinology. In: Neuroendocrinology, edited by D. T. Krieger and J. C. Hughes. Suncerland: Sinaicer, 1980, pp. 3-12. 15. Kvetnansky, R. Biosynthesis of adrenal catecholamines during adaptation of rats to immobifization stress. Adv Exp IVied Biol 32: 603--617, 1972. 16. Kvetnansky, R., R. McCarty, N. B. Thoa, C. R. Lake and I. J. Kopin. Sympatho-adrenal responses of spontaneously hypertensive rats to immobilization stress. Am J Physiol 236: H 457-H 462, 1979. 17. Kvetnansky, R., V. K. Weise, N. B. Thoa and I. J. Kopin. Effects of chronic guanethidine treatment and adrenal medullectomy on plasma levels of catecholamines and corticosterone in forcibly immobilized rats. J Pharmacol Exp Ther 209: 287-291, 1979.
599 18. Lenox, R. H., G. T. Kant, G. R. Sessions, L. L. Pennington, E. A. Maugey and 3. L. Meyerhoff. Specific hormonal and neurochemical responses to different stressors. Neuroendocrinology 30: 300-308, 1980. 19. Maickel, R. P., N. Matussek, D. N. Stern and B. B. Brodie. The sympathetic nervous system as a homeostatiG mechanism. I. Absolute need for sympathetic nervous function in body temperature maintenance by cold-exposed rats. J Pharmacol Exp Ther 157: 103-110, 1967. 20. Maickel, R. P., D. N. Stern, E. Takabatake and B. Brodie. The sympathetic nervous syst.em as a homeostatic mechanism. II. Effect of adrenocortical hormones on body temperature maintenance of cold-exposed rats. J Pharmacol Exp Ther 157:111116, 1967. 21. Makara, O. B., M. Palkovits and J. Szenthagotai. The endocrine hypothalamus and the response to stress. In: Selye's Guide to Stress Research, edited by 3. Selye. New York: Van Nostrand Reinhold Co., 1980, pp. 280-337. 22. Mallow, S. and P. N. Witt. Effect of stress and tranquilization on plasma free fatty acid levels in the rat. J Pharmacol Exp Ther 132: 126-130, 1966. 23. Manara, L., E. Costa, D. N. Stern and R. P. Maickel. Effect of chemical sympathectomy on oxygen consumption by the coldexposed rat. Int J Neuropharmacol 4: 301-307, 1965. 24. Mason, 3. W. A re-evaluation of the concept of "nonspecificity" in stress theory. J Psychiatr Res 8: 323-333, 1971. 25. Mason, J. W. Specificity in the organization of neuroendocrine response profiles. In: Frontiers in Neurology and Neuroscience Research, edited by P. Seeman and G. M. Brown. Toronto: University of Toronto Press, 1974, pp. 68-80. 26. Selye, H. A syndrome produced by diverse noxious agents. Nature 138: 320, 1936. 27. Selye, H. The Physiology and Pathology of Exposure to Stress. Montreal: Acta Inc. Medical Publishers, 1950. 28. Stern, D. N. and R. P. Malckel. Studies on starvation-induced hypermobifization of free fatty acids (FFA). Life Sci 11: 872877, 1963. 29. Tepperman, J. Metabolic and Endocrine Physiology. Chicago: Yearbook Biomedical Pubfishers, 1981. 30. Ursin, H. Personality, activation and somatic health: a new psychosomatic theory. In: Coping and Health, edited by S. Levine and H. Ursin. New York: Plenum Press, 1980. 31. Vernikos-Danellis, J. and P. Heybach. Psychophysiologic mechanisms regulating the hypothalamic-pituitary-adrenal response to stress. In: Selye's Guide to Stress Research, edited by J. Selye. New York: Van Nostrand Reinhold Co., 1980, pp. 206-251. 32. Wannarka, G. L., H. P. Fletcher and R. P. Maickel. Centrally mediated drug-induced hyperglycemia in mice. Neuropharmacology 22: 341-346, 1983.
0031-9384/85 $3.00 + .00
Comparative Biochemical Responses of Rats to Different Stressful S t i m u l i 1 MAURICIO
R. O D I O 2 A N D R O G E R P. M A I C K E L
Department o f Pharmacology and Toxicology, School o f Pharmacy and Pharmacal Sciences Purdue University, West Lafayette, I N 47907 R e c e i v e d 16 J a n u a r y 1984 ODIO, M. R. AND R. P. MAICKEL. Comparative biochemical responses of rats to different stressful stimuli. PHYSIOL BEHAV 34(4) 595-599, 1985.--Adult male rats were exposed to single applications of one of three stressful stimuli (low environmental temperature, immobilization, random footshock) for periods up to 4 hours and plasma levels of corticosterone (PCS), fatty acids (PFA), and glucose (PGL) were determined at various points during the stress exposure and 1 and 2 hours post-exposure. The levels of PCS were increased by all 3 stressful stimuli in a similar temporal pattern, with the greatest magnitude of effect seen for immobilization and the least for cold exposure. The time courses of increased PFA levels were similar for immobilization and cold exposure; the response to foot shock was delayed in onset by 2 hours. The PGL response was minimal for cold exposure and foot shock, but showed a marked elevation during the first 2 hours of immobilization. The results suggest that the response pattern obtained is characteristic of the stressful stimulus employed, with PCS showing the least degree of specificity. Stressful stimuli Plasma fatty acids
Cold exposure Immobilization Plasma glucose
Random footshock
Plasma corticosterone
ferential patterns of responses by Lenox et al. [18], who showed varied responses of brain cGMP, pituitary c A M P and circulating prolactin levels in rats immobilized or exposed to cold. Keim and Sigg [12] found that the temporal pattern o f responses, i.e., changes in hypothalamic N E and PCS observed in rats exposed to a novel environment, 2 mA electric footshock, or restraint, were not similar. Even as subtle a difference as whether footshock was applied to rats according to a fixed or variable schedule resulted in differential time courses for PCS responses [1]. While single time point measurements may suggest that the hypothalamicpituitary adrenocortical axis responds in a non-specific manner to different stressful stimuli, examination o f temporal response "profiles" supports an opposite conclusion. Thus, Vernikos-Dannellis and Heyback [31] stress the importance of time-course measurements to describe response phenomena, rather than single time point values. In terms o f the overall " r e s p o n s e " to stressful stimuli, two recent reports may be o f particular relevance. Ursin [30] examined the relationship between various types o f personalities and the involvement o f response networks in the experience o f parachute jumping. The results show three response groups: one with a predominantly adrenal medullary (E plus NE) response, a second with a predominantly adrenocortical (PCS) response, and a third with a response mainly in plasma fatty acids (PFA). Henry [10] reviewed this and other data and concluded that differential responses
S E L Y E [26] first reported reactions o f higher organisms to the application o f noxious or threatening stimuli, and later characterized these responses to stress as the General Adaptation Syndrome [27]. Numerous subsequent publications have contradicted and supported various aspects o f the Selye hypothesis. In 1971, Mason [24] reported that initial exposure to virtually any unfamiliar stimulus evoked an increase in pituitary-adrenocortical activity as measured by elevated levels of plasma corticosterone (PCS), while animals acclimated to the various experimental procedures demonstrated differential PCS responses to the stressors. Forberg et al. [6] reported elevated urinary excretion of epinephrine (E) and norepinephrine (NE) in normal human volunteers viewing either pleasant, comic movies or fearful, rage-eliciting films. These questions of the nonspecificity of stress responses are interesting since the pituitaryadrenocortical, adrenal medullary and post-ganglionic sympathetic nervous systems are all presumably included in the overall " r e s p o n s e " to stress as defined by Selye's G.A.S. [27]. Mason [25] concluded that stimulus-specific responses o f the organism could be confounded by biochemical phenomena evoked by the heightened state of arousal induced in many experimental conditions; thus, the "stimulus-independent" response o f Selye could be the result of cognitive-emotional factors that would confound the experimental results. Different stressful stimuli have been shown to elicit dif-
~Research supported in part by a LASPAU-University of Costa Pica Graduate Scholarship and by a Purdue University David Ross XR Fellowship. 2Taken in part from a thesis submitted in partial fulfillment of the requirements for the Ph.D. degree, Purdue University, 1983.
595
596
ODIO A N D M A I C K E L
such as PCS or plasma E or NE could be correlated with both the type of stimulus and the background and experience of the experimental subjects. In the present report, plasma corticosterone (PCS) levels have been selected to characterize the hypothalamopituitary-adrenocortical system [29]; plasma fatty acid (PFA) levels have been selected to characterize the sympathetic nervous-adipose tissue system [2]; and plasma glucose (PGL) levels have been selected to characterize the adrenomedullary-liver system [7,32]. While no one of these parameters represents the optimal choice, characterization of the time-response curves resulting from activation by several stressful stimuli may yield a fingerprint of the stress response for each stimulus [31]. Three stressful stimuli, sufficiently different in nature to enhance the probability of producing stimulus-specific response patterns, were used. Exposure to an environmental temperature of 4°C (CLD) was selected as a primarily physiological stressor; all three response systems are clearly involved [19,20]; physical immobilization (IMM) was selected as a primarily psychological stressor [15,16]; again involving all three response systems [15,16]; and inescapable random electric footshock (ESK) was selected as a third stressful stimulus, one which combined physiological (slight discomfort) with psychological (an element of anticipation) characteristics and also differed from CLD and IMM in being discontinuous. E S K has been reported to induce responses in PCS [1], P F A [22], and P G L [5].
CLD
6O
J4k-'~L .... - - a k
8
!m - ....
,,,, "~\
!
....... o ........
ESK
---¢~----
IMM
............. ~:,,.
O
-',
if) c~ O c3
k.
X
03 0_ 10
%
I I
1 I 2 5 STRESSFUL STIMULI
Lt/ 4" /:
I I RECOVERY
I 2 (HOURS) ._
FIG. 1. Response of plasma corticosterone (PCS) to stressful stimuli. Each point in the mean of values obtained from 6 rats exposed to each stressful stimulus, except that t=0 points are the mean of values obtained from 12 rats (2 killed at each corresponding time point) as described in the Method section. Solid symbols represent values that differ significantly from t=0 values for each stressful stimulus (0<0.05).
METHOD
Subjects
Adult, male Sprague-Dawley-derived rats (total N = 144) weighing 200-225 g at delivery were purchased from Murphy Breeding Laboratories, Plainfield, IN. They were group housed (6/cage) in stainless steel cag_es (60x 57 x 50 era) on an ad lib diet of Wayne " L a b l o x " and tap water in an animal room maintained at 21-25°C with a 14/10 lighting cycle (on at 0600, off at 2000 hr). The rats were allowed a 9 day acclimation period prior to experimental use.
Chemicals
All chemicals used in the assay procedures were analytical reagent quality purchased from commercial suppliers. Heparin sodium was purchased from Abbott Laboratories, North Chicago, IL. Chemical standards used were: corticosterone and palmitic acid (Sigma Chemical Co., St. Louis, MO) and glucose (Beckman Instruments, Chicago, IL). Procedures
Prior to collection of blood samples, 50 ml beakers and 13 ml disposable borosilicate glass culture tubes were rinsed with a solution containing 125 U heparin sodium/ml 0.9% NaCI. The tubes were emptied, inverted for 10 min, then chilled in ice. Animals were killed by decapitation and trunk blood was collected into a chilled beaker, transferred to a chilled tube and centrifuged (2,000 rpm for 15 min) at 40°C. Following centrifugatmn, the tuoes were again placed in ice while four aliquots of plasma were transferred to sterile, plastic stoppered cryotubes for storage at -70°C until assay. Prior to assay, each tube was thawed at room temperature and mixed by gentle agitation to insure a homogeneous sample. Each cryotube was used for a single assay (i.e.,
PCS, PFA, or PGL) with replicate sampling; the fourth tube was maintained as a spare. Plasma corticosterone (PCS) was determined by the fluorometric method of Guillemin et al. [9] with activation and emission wavelengths of 467 and 525 nm, respectively. Plasma fatty acids (PFA) were determined by the colorimetric method of Duneombe [4] using the absorbance peak at 440 nm. Plasma glucose (PGL) was determined directly in a Beckman Model I Glucose Analyzer, using the glucose oxidase procedure o f Kadish et aL [11]. Cold exposure (CLD) was applied by placing individual rats in stainless steel cages (27x20x 17 cm) with wire mesh floors located in a walk in cold room maintained at 2.5-5.5°C with continual air circulation. Immobilization (IMM) was applied by placing individual rats in a metal cage (30x 15 x 17 cm) with screen wire sides and a moveable upper lid capable of continuous (thumbscrew) adjustment in the vertical plane. Pressure was applied to the animal until movement was completely restricted; at this point respiration was not impeded and no physical damage was produced. Electric footshock stress (ESK) was applied by placing individual animals in Lehigh Valley operant chambers; the grid floor was connected to a Grason Stadler shock generator supplying 0.5 mA shocks with a duration of 2.0 sec on a randomized schedule of 6 shocks per 60 rain. Control animals were placed in individual plastic cages (27x 18x 13 cm) for time periods comparable to the stressed animals. No food or water was available to the animals during the stress periods. Each test group consisted of 6 rats exposed to a given stress and 2 additional rats as controls. The 6 sets of controls were not significantly different from each other; consequently, they were pooled to yield a total N=12. Stressful stimuli were applied for 0.5, 1.0, 2.0 or 4.0 hr; additional groups were exposed to stressful stimuli for 4 hr, then placed
RESPONSES TO S T R E S S F U L STIMULI
597
TABLE 1 AREA UNDER THE CURVE (AUC) VALUES FOR RESPONSES TO STRESSFUL STIMULI IN RATS
Stressful Stimulus
Onset
Maintenance
Recovery
PCS 0zg-hr)
NONE CLD ESK IMM
41 72 81 103
34 69 96 100
38 42 56 60
PFA (/xEq-hr)
NONE CLD ESK IMM
151 268 161 313
202 308 275 344
245 296 372 267
PGL (mg-hr)
NONE CLD ESK IMM
321 323 327 431
324 321 311 336
308 370 284 321
Values were determined as described in the Method section.
in individual cages (similar to controls) for an additional 1.0 or 2.0 hr prior to killing.
Statistical Analysis Data is presented as mean-+S.E.M, for individual groups. Statistical analyses was by appropriate ANOVA with post hoc comparisons by the two tail Dunnett's test; the criteria for significance was set at p<0.05 for all comparisons. The construct of Area Under the Curve (AUC) has been utilized to make some of the comparisons. These values were obtained as described by Gibaldi and Perrier [8]. Baseline values were determined from the control data; experimental data were plotted as obtained for the time points. Time segments were divided into: 0-2 hours of stress (onset), 2-4 hours of stress (maintenance), and 0-2 hours post stress termination (recovery). RESULTS
The time course of response for PCS for each of the three stressful stimuli is presented in Fig. 1. All of the stressful stimuli evoked a positive response peaking at 0.5 hr and persisting at significantly elevated values throughout the 4 hr of application of each stressful stimuli. By 1 hr after termination, values for PCS in CLD and ESK animals had returned to normal, while those of IMM rats were still significantly elevated. IMM also resulted in higher values during stress. The comparative potencies of the three stressful stimuli may be estimated from the area under the curve (AUC) data obtained from these time-course experiments and presented in Table 1. During the onset and maintenance periods, these values totaled 203, 177 and 141/~g-hr for IMM, ESK, and CLD, respectively, as compared to 75/~g-hr in the control rats.
CLD
........ o ........
ESK
---A----
IMM
2OO
Phase of Stressful Stimulus Response Parameter
•- - o
.
E o_
/
0
~ k~
15o
_o
i
.
.
.
~/ . . . . .
~
"'~IA
~
•
....
""
....-" ~ I / ~ .'" ,,."
3
\ ~
50
O~
I I
|
I
2 3 STRESSFUL STIMULI
b/ 4"
I I RECOVERY
I 7' (HOURS}
FIG. 2. Response of plasma fatty acids (PFA) to stressful stimuli. Each point is the mean of values obtained from 6 rats exposed to each stressful stimulus, except that t =0 points are the mean of values obtained from 12 rats (2 killed at each corresponding time point) as described in the Method section. Solid symbols represent values that differ significantly from t=0 values for each stressful stimulus (0<0.05).
The time course of responses for PFA is presented in Fig. 2. Each of the stressful stimuli evoked a different pattern of response in this parameter. For example, CLD significantly elevated PFA values for the entire duration of the stressexposure; this elevation persisted for at least two hours post cessation of the stress. The PFA response to IMM was significantly elevated during the 4 hr application of the stressful stimulus, persisted for 1 hr post termination, and returned to normal values in the next hr. The response of PFA levels to ESK was unique. No response was seen until the 4 hr time point, but this significant elevation persisted throughout the 2 hr post-stress period. AUC values for PFA (Table 1) reflected the response to each of the stressful stimuli. For the onset phase,/zEq-hr values were 313,268, and 161 for IMM, CLD, and ESK, respectively, with a control value of 151. For the maintenance phase, the order was similar; 344, 308, and 275 for IMM, CLD, and ESK, respectively, with a control value of 202/zEq-hr. In the post-stress phase, only ESK rats deviated from control values. Figure 3 presents the time-course of data for PGL in response to the stimuli; 3 differential effects were seen. The most significant effect was that of IMM which evoked a marked elevation of PGL over the first 2 hours (the onset phase). In contrast, ESK caused a small, but statistically significant elevation of PGL levels at 1 and 2 hr of stress, and CLD evoked small but significant elevations of PGL at 0.5 and 4 hr of stress. These effects are borne out by the AUC values in Table 1, where the only deviations from control were those seen with IMM during the onset phase. DISCUSSION
In order to best consider the characteristics of timeresponse curves obtained, a triphasic system has been used.
598 The first 2 hours of exposure to each stressful stimulus is considered as the " o n s e t " phase, during which the animal is exposed to a novel situation. Presumably, little adaptation of biochemical mechanisms occurs during this period. The next 2 hours of exposure are considered the "maintenance" phase, and presumably sees the initiation of adaptation phenomena. Finally, the 2 hours after cessation of the stress period (the " r e c o v e r y " phase) offers the biological systems of the organism an opportunity to return to normal. Using this approach, the responsiveness of the hypothalamo-pituitary-adrenocortical system, as reflected in the PCS levels, showed the least degree of specificity in terms of differential responding to the 3 stressful stimuli. Some differences in the magnitude and duration of the PCS response were observed, presumably reflecting differential severity o f the stressful stimuli, but, in general, these data would support some previously advocated contentions that PCS is a non-specific response [21,27]. However, the differential decrements in PCS levels over time seen with the three stressors are in agreement with the findings of Bassett and Cairncross [1] and Keim and Sigg [12]. The somewhat longer persistence of the PCS levels in the animals exposed to IMM can be explained on the basis of a greater magnitude of hypothalamo-pituitary activity during the 4 hr stress exposure [14]. A contrasting situation is observed in the case of the sympathetic nervous system-adipose tissue axis, in which stimulation is manifested by increased lipolytic activity with consequent elevation o f P F A levels. Three different response patterns were seen. Animals exposed to the E S K stress showed no change in P F A levels until the 4 hr stress point; the elevated levels seen at this time were maintained throughout the 2 hr of the recovery phase. In contrast, both CLD and IMM stress caused significant elevations in P F A levels from 30 minutes of stress through the In'st hour of the recovery phase; however, by the second hour of the recovery phase, the P F A levels in the IMM rats had fallen back to levels near those o f controls, while those of the CLD rats remained somewhat elevated from the value of the controls. A confounding variable for CLD may be an increased utilization of P F A by the rat in order to facilitate maintenance of a normothermia in the face of a severe temperature gradient between animal and environment [2], additionally, one must consider the fact that all animals were without food for 6 hours [28]. The unique response to E S K , a response similar to that seen by Mallow and Witt [22] may reflect the fact that once adipose tissue lipase is activated by increased sympathetic nervous activity, the enzyme has a sufficiently lengthy half-life in the activated form to result in a persistent lipolysis [2,28]. The delayed onset o f this response is, however, inexplicable at this time. Finally, the effects of the three stressful stimuli on PGL levels yields results that tend to support the concept of specific response patterns to specific stresses as well. Rats exposed to C L D stress show minimal changes in PGL values; the increases observed at 0.5 and 4.0 hr are statistically significant although they represent less than a 10% increase above baseline. In contrast, exposure of rats to E S K leads to
ODIO A N D M A I C K E L
250[.-
/,,&~
'\,,,
,,¢
~ 2OO ~-
~o
CLD
....... [] .......
EsK
---"¢~--- IMM
",,
'
+
" .............. q" .....
" ......... o ................ o
g <
I00
g.
50
Oo
I
I
I
I
.~//
2 3 ,~ STRESSFUL STIMULI
I
I RECOVERY
I
2 (HOURS)
FIG. 3. Response of plasmaglucose (PGL) to stressful stimuli. Each point is the mean of values obtained from 6 rats exposed to each stressful stimulus, except that t=0 points are the mean of values obtained from 12 rats (2 killed at each corresponding time point) as described in the Method section. Solid symbols represent values that differ significantly from t=0 values for each stressful stimulus (p<0.05).
significant elevations in PGL levels at 1.0 and 2.0 hr; these values are about 20% above control levels. The effects of exposure to IMM are unique, as PGL levels are elevated by more than 50% at the 0.5 hr time point, become even more elevated at 1.0 hour and do not return to normal values until 1.0 hr post termination of the stress session. This responsivity to IMM probably reflects both an increased activity of the sympatho-adrenal medullary system [16, 17, 32] and a reduced utilization of PGL by the immobilized, and therefore inactive, rat [13]. The CLD rats would not be expected to show an increased PGL since they are utilizing glucose at a higher rate [23]; the E S K rats presumably have a more moderate response since they are more active (jumping in response to the shock) despite increased sympathomedullary output [3]. Similarly, some portion of the increased PGL levels after 0.5-2.0 hr of IMM may reflect a decreased utilization due to decreased muscular activity [13]. Thus, it seems rather apparent that the three stressful stimuli are capable of eliciting clearly differential response patterns even though these may be slightly confounded by other factors influencing the biochemical parameters being measured. This should not be surprising since the stimuli represent different blends o f continuity (CLD, IMM) and discontinuity (ESK), of psychological components (ESK, IMM) and physiological components (CLD), of activity (CLD, ESK) and inactivity (IMM).
REFERENCES 1. Bassett, J. R. and J. D. Cairncross. Time course for plasma 2. Brodie, B. B., R. P. Maickel and D. N. Stem. Autonomic nerv1l-hydroxycorticosteroid elevation in rats during stress. Pharous system and adipose tissue. In: Handbook of Physiology: macol Biochem Behav 3: 139--142, 1975. Adipose Tissue, edited by A. E. Renold and G. F. Cahill. Baltimore: Waverly Press, 1965, pp. 583-600.
RESPONSES TO STRESSFUL
STIMULI
3. DaPrada, M., L. Pieri and G. B. Picotti. Effect of midazolam (a water soluble benzodiazepine) on stress-induced increase of plasma catecholamines. In: Catecholamines and Stress: Recent Advances, edited by E. Usdin, R. Ketnansky and I. J. Kopin. New York: Elsevier/North Holland, 1980, pp. 231-236. 4. Duncombe, W. G. The colorimetric microdetermination of nonesterified fatty acids in plasma. Clin Chim Acta 9: 122-125, 1964. 5. Ehrentheil, O. F., L. J. Reyna, C. J. Adams, T. J. Giovanniello and E. T. Chert. Blood sugar response to electric shock stress in normal, adrenalectomized and hypophysectomized white rats. Diabetes 16: 325-330, 1967. 6. Froberg, J., C. O. Karisson, L. Levi and L. Lindberg. Physiological and biochemical stress reactions induced by psychosocial stimuli. In: Society, Stress and Disease, vol 1, edited by L. Levi. London: Oxford University Press, 1971, pp. 280-295. 7. Ganong, W. F. Review of Medical Physiology. San Francisco: Lange Medical Publications, 1975. 8. Gibaldi, M. and D. Perrier. Pharmacokinetics. New York: Marcel Dekker, 1975. 9. Guillemin, R., G. W. Clayton, H. S. Lipscomb and J. Smith. Fluorimetric measurement of rat plasma and adrenal corticosterone concentrations. J Lab Clin Med 53: 830-832, 1959. 10. Henry, J. P. Controversy in the stress field. In: Catecholamines and Stress, Recent Advances, edited by E. Usdin, R. Kvetnansky and I. J. Kopin. New York: Elsevier/North Holland, 1980, pp. 557-571. 11. Kadish, A. H., R. L. Little and J. C. Sternberg. A new and rapid method for the determination of glucose by measurement of the rate of oxygen consumption. Clin Chem 14:116-131, 1968. 12. Keim, K. L. and E. B. Sigg. Physiological and biochemical concomitants of restraint stress in rats. Pharmacol Biochem Behav 4: 289-297, 1976. 13. Koslowsld, S. Panel discussion on stress theory. In: Catecholamines and Stress: Recent Advances, edited by E. Usdin and I. J. Kopin. New York: Elsevier/North Holland, 1980, p. 854. 14. Krieger, D. T. The hypothalamus and neuroendocrinology. In: Neuroendocrinology, edited by D. T. Krieger and J. C. Hughes. Suncerland: Sinaicer, 1980, pp. 3-12. 15. Kvetnansky, R. Biosynthesis of adrenal catecholamines during adaptation of rats to immobifization stress. Adv Exp IVied Biol 32: 603--617, 1972. 16. Kvetnansky, R., R. McCarty, N. B. Thoa, C. R. Lake and I. J. Kopin. Sympatho-adrenal responses of spontaneously hypertensive rats to immobilization stress. Am J Physiol 236: H 457-H 462, 1979. 17. Kvetnansky, R., V. K. Weise, N. B. Thoa and I. J. Kopin. Effects of chronic guanethidine treatment and adrenal medullectomy on plasma levels of catecholamines and corticosterone in forcibly immobilized rats. J Pharmacol Exp Ther 209: 287-291, 1979.
599 18. Lenox, R. H., G. T. Kant, G. R. Sessions, L. L. Pennington, E. A. Maugey and 3. L. Meyerhoff. Specific hormonal and neurochemical responses to different stressors. Neuroendocrinology 30: 300-308, 1980. 19. Maickel, R. P., N. Matussek, D. N. Stern and B. B. Brodie. The sympathetic nervous system as a homeostatiG mechanism. I. Absolute need for sympathetic nervous function in body temperature maintenance by cold-exposed rats. J Pharmacol Exp Ther 157: 103-110, 1967. 20. Maickel, R. P., D. N. Stern, E. Takabatake and B. Brodie. The sympathetic nervous syst.em as a homeostatic mechanism. II. Effect of adrenocortical hormones on body temperature maintenance of cold-exposed rats. J Pharmacol Exp Ther 157:111116, 1967. 21. Makara, O. B., M. Palkovits and J. Szenthagotai. The endocrine hypothalamus and the response to stress. In: Selye's Guide to Stress Research, edited by 3. Selye. New York: Van Nostrand Reinhold Co., 1980, pp. 280-337. 22. Mallow, S. and P. N. Witt. Effect of stress and tranquilization on plasma free fatty acid levels in the rat. J Pharmacol Exp Ther 132: 126-130, 1966. 23. Manara, L., E. Costa, D. N. Stern and R. P. Maickel. Effect of chemical sympathectomy on oxygen consumption by the coldexposed rat. Int J Neuropharmacol 4: 301-307, 1965. 24. Mason, 3. W. A re-evaluation of the concept of "nonspecificity" in stress theory. J Psychiatr Res 8: 323-333, 1971. 25. Mason, J. W. Specificity in the organization of neuroendocrine response profiles. In: Frontiers in Neurology and Neuroscience Research, edited by P. Seeman and G. M. Brown. Toronto: University of Toronto Press, 1974, pp. 68-80. 26. Selye, H. A syndrome produced by diverse noxious agents. Nature 138: 320, 1936. 27. Selye, H. The Physiology and Pathology of Exposure to Stress. Montreal: Acta Inc. Medical Publishers, 1950. 28. Stern, D. N. and R. P. Malckel. Studies on starvation-induced hypermobifization of free fatty acids (FFA). Life Sci 11: 872877, 1963. 29. Tepperman, J. Metabolic and Endocrine Physiology. Chicago: Yearbook Biomedical Pubfishers, 1981. 30. Ursin, H. Personality, activation and somatic health: a new psychosomatic theory. In: Coping and Health, edited by S. Levine and H. Ursin. New York: Plenum Press, 1980. 31. Vernikos-Danellis, J. and P. Heybach. Psychophysiologic mechanisms regulating the hypothalamic-pituitary-adrenal response to stress. In: Selye's Guide to Stress Research, edited by J. Selye. New York: Van Nostrand Reinhold Co., 1980, pp. 206-251. 32. Wannarka, G. L., H. P. Fletcher and R. P. Maickel. Centrally mediated drug-induced hyperglycemia in mice. Neuropharmacology 22: 341-346, 1983.