Neuroendocrinology 19: 339-351 (1975)
Effects of Stress on Plasma Corticosterone and Growth Hormone Levels in Rats with Median Eminence-Pituitary Islands' R.W. R ice 2. J. K r o n in g an d V. C r it c h l o w Department of Anatomy, Medical School, University of Oregon Health Sciences Center, Portland, Oreg.
Key Words. Stress responses • Corticosterone • Growth hormone • Hypothalamus ■ Median eminence • Pituitary Abstract. The effects of stress on plasma corticosterone and growth hormone (GH) were determined in adult female rats 24 h after forebrain removal in order to study the anatomi cal substrate essential for stress-induced alteration of pituitary function. Rats with median eminence-pituitary islands (MEP1) showed significant increments in plasma corticosterone levels 15 min after immobilization-blood withdrawal stress. Rats with basal hypothalamicpituitary islands (BHP1) showed comparable increments, suggesting that the presence of hypothalamic tissue in addition to median eminence did not enhance the steroid response. A subsequent study demonstrated that MEP1 rats are capable of a 2nd response. 6 h after the first. MEPI rats also showed a significant 50% increase in plasma GH levels 15 min after immobilization-blood withdrawal. In a final experiment, MEPI rats showed corticosterone but not GH responses to immobilization applied without blood withdrawal. These data offer further evidence that some stressful stimuli can alter pituitary functions under short term conditions in the absence of hypothalamic or other forebrain neurons.
Considerable evidence suggests that certain stressful stimuli are capable of activating the pituitary-adrenal system in rats with surgically-interrupted central neural connections of the medial basal hypothalamus (MBH). Thus, rats with chronically-isolated MBH show significant increases in plasma corticosterone levels following exposure to a variety of stressful stimuli [H alász at a!., 1967; F eldm an et al„ 1968, 1970; P alk a et al., 1969; G reer 1 Supported by USPHS grant AM 16794 from the National Institutes of Health. 2 Supported by USPHS postdoctoral fellowship IF 22 AM 00337 from the National Institutes of Health.
Downloaded by: Nagoya University 133.6.82.173 - 1/10/2019 10:14:42 PM
Received: June 17th, 1975; revised MS accepted: October 28th, 1975.
340
R ice/ K rön ing/C r itch low
Downloaded by: Nagoya University 133.6.82.173 - 1/10/2019 10:14:42 PM
et al., 1970; M akara et al., 1970], Results from acute studies utilizing partial forebrain removal consistently indicate that the isolated MBH-pituitary unit can support pituitary-adrenal responses to combined ether and leg break [M atsuda el al., 1963], ether-blood withdrawal or immobilization [D u n n and C r it c h l o w , 1969a], Hypothalamic structures, exclusive of those con tained in the median eminence, may not be required for such responses to some stimuli. M atsuda el al. [1964] reported that rats with median eminencepituitary islands (MEPI) showed elevated corticosterone levels following ether-leg break stress and suggested that the median eminence-pituitary unit constituted the minimal substrate for this response. Similarly, M akara et al. [1970] reported that rats with MEPI showed corticosterone responses to i.p. injections of histamine and insulin. The median eminence, however, does not appear essential in all situations. Ablation of the medial hypothalamus, including median eminence, is compatible with acute corticosterone responses to injection of E. coli endotoxin or large doses of formaldehyde [S ta rk et al., 1973/74], despite pituitary infarction. Furthermore, W itorsch and B rodish [1972] report that removal of the entire forebrain, including median emi nence, does not block corticosterone responses to certain combined stress procedures. These findings suggest that central neural pathways to the MBH, and perhaps the MBH itself, are not essential to the triggering of ACTH secretion by some noxious stimuli. The mechanisms involved in such pi tuitary activation are unknown, and their importance and role in the physio logical control of pituitary-adrenal function are undetermined. This laboratory was previously unable to elicit plasma corticosterone re sponses to ether or immobilization stress from pituitary islands in rats sub jected to total forebrain removal [D u n n and C r it c h l o w , 1969a]. Subsequent unpublished observations, however, suggested that acute corticosterone re sponses to immobilization-blood withdrawal persisted when the median emi nence was retained in connection with the pituitary; the magnitude of such responses was similar to that seen earlier in island preparations containing basal hypothalamic tissue [D u n n and C r it c h l o w , 1969a], The purpose of the present studies was to further explore the potential of the median emi nence-pituitary unit to support acute changes in ACTH secretion in response to stress and to assess the contribution of MBH tissue to such responses. Changes in plasma corticosterone levels were used as an index of changes in ACTH secretion. Because immobilization acutely activates the pituitaryadrenal system in rats with basal hypothalamic islands [D u n n and C r it c h l o w , 1969a], this stress procedure was used throughout the experiments. A secondary aim was to determine whether stress-induced alterations in pi-
Stress and Median Eminence-Pituitary islands
341
tuitary secretion of growth hormone (GH) persist in these preparations. In this species, stress causes an acute, marked decrease in circulating GH levels [S c h a lc h and R e ic h l in , 1966; G arcia and G e sc h w in d , 1968].
Materials and Methods
Downloaded by: Nagoya University 133.6.82.173 - 1/10/2019 10:14:42 PM
Adult female rats (Charles River, CD), weighing between 200 and 280 g, were used. All rats were maintained in animal quarters under conditions of controlled lighting (05.00-18.00 h) and temperature (26 ± 2 C) for at least 3 weeks prior to experimentation. Except where specified, Purina Lab Chow and tap water were available ad libitum. The 1st experiment was designed to assess pituitary-adrenal responses to immobiliza tion-blood withdrawal in MEPI rats and to compare these responses with those obtained in BHPI rats and in intact controls. Animals were studied approximately 24 h after surgery. Forebrain removal and brain stem transection were performed under ether anesthesia as described by M atsuda et al. [1963] and D unn and C ritchlow [1969a). In BHPI prepara tions, the brain stem was transected at the level of the superior colliculus and most of the forebrain was removed by aspiration, leaving tissue in the ventromedial-arcuate region. In MEPI rats, the entire forebrain was removed except for the median eminence which was retained in connection with the pituitary. The rats were placed in a 95% 0 2-5% C 0 2 environment following surgery until 17.00 h. Rectal temperatures were monitored periodi cally and cold water immersion was applied when temperatures exceeded 39 C. Because rats with forebrains removed did not eat spontaneously, food was withheld from intact controls and all rats were given i.p. injections of 5 ml of 5% dextrose-saline at 17.00 h. Approximately 24 h after surgery, rats were subjected to the previously described double bleeding procedure [Z immermann and C ritchlow , 1967] to assess 'non-stress’ and stress corticosterone levels, i.c., concentrations before and after immobilization-blood with drawal. This procedure consisted of rapidly immobilizing the rats for 3 min in the supine position, exposing the external jugular vein following procainization of the overlying skin and removing 0.5 ml of blood within 3 min of initial handling. These initial blood samples were treated as described below for determining ‘non-stress’ plasma corticosterone levels. 15 min after initiation of immobilization, a 2nd blood sample was obtained following decapitation for determining stress levels of corticosterone. Blood was collected in EDTArinsed syringes or beakers and centrifuged. Plasma was separated for determination of corticosterone concentrations by the microfiuorometric method of G lick et al. [1964], Because of the results obtained in the above experiment, a 2nd study was performed to determine whether MEPI rats are capable of responding to a 2nd immobilization-blood withdrawal stress. Beginning at 09.00 h, approximately 24 h after surgery, the animals were immobilized for 3 min and "non-stress’ blood samples were obtained as described above. 13 min after initiation of stress, rats were immobilized again and a 2nd or stress sample was collected from the jugular vein at 15 min. The rats were then placed in a holding cage and studied 6 h later; blood samples were collected within 3 min of initiation of immobilization and decapitation was performed at 15 min. Some of the blood samples from MEPI rats in the above experiments were also used for determining 'non-stress’ and stress GH levels. In our experience (unpublished observa tions), significant changes in plasma GH concentration do not occur prior to 3 min after
342
R ice/K roning /C ritchlow
initiation of stress and nearly maximal responses are obtained at 15 min. The blood was chilled immediately after collection and centrifuged. Aliquots of plasma were rapidly frozen and stored at 15 C for measurement of radioimmunoassayable GH, using the NIH kit and the methods described by Schindler el at. [1972], The sensitivity of the assay was approximately 0.2 ng and the coefficient of variability for plasma pools was 13% between and 7% within assays. A final study was designed to determine whether the corticosterone and GH responses observed in MEPI rats could be elicited by immobilization per se, i.e., without the stimuli associated with blood sampling. Surgery was performed as described above. Approximately 24 h later, half the animals were decapitated within 20 sec of handling and blood was collected for determination of 'non-stress' levels of both corticosterone and GH. The remaining animals were immobilized for 3 min and decapitated at 15 min. Blood was collected for measuring stress levels of both hormones. Treatments were assigned and data analyzed according to a randomized design. Twoway analysis of variance (ANOV) for repeated measures [W iner, 1962] and the paired /-test were used to analyze data from experiments involving both 'non-stress' and stress blood samples from individual rats. Comparison of stress-induced increments was per formed by one-way ANOV and Newman-Keuls’ test. Sludent’s /-test was used in the last experiment which involved single data points from individual animals. Probabilities less than 0.05 were considered statistically significant. Heads were fixed in 10% formalin following decapitation, decalcified in formic acidsodium citrate solution and subsequently processed for histological examination of in situ pituitary and hypothalamus. Data were discarded from rats with pituitary infarcts and from MEPI rats with remnants of hypothalamic tissue dorsal to the median eminence.
As shown in figure 1, the ‘non-stress’ corticosterone levels were higher (p
Effects of Stress on Plasma Corticosterone and Growth Hormone Levels in Rats with Median Eminence-Pituitary Islands' R.W. R ice 2. J. K r o n in g an d V. C r it c h l o w Department of Anatomy, Medical School, University of Oregon Health Sciences Center, Portland, Oreg.
Key Words. Stress responses • Corticosterone • Growth hormone • Hypothalamus ■ Median eminence • Pituitary Abstract. The effects of stress on plasma corticosterone and growth hormone (GH) were determined in adult female rats 24 h after forebrain removal in order to study the anatomi cal substrate essential for stress-induced alteration of pituitary function. Rats with median eminence-pituitary islands (MEP1) showed significant increments in plasma corticosterone levels 15 min after immobilization-blood withdrawal stress. Rats with basal hypothalamicpituitary islands (BHP1) showed comparable increments, suggesting that the presence of hypothalamic tissue in addition to median eminence did not enhance the steroid response. A subsequent study demonstrated that MEP1 rats are capable of a 2nd response. 6 h after the first. MEPI rats also showed a significant 50% increase in plasma GH levels 15 min after immobilization-blood withdrawal. In a final experiment, MEPI rats showed corticosterone but not GH responses to immobilization applied without blood withdrawal. These data offer further evidence that some stressful stimuli can alter pituitary functions under short term conditions in the absence of hypothalamic or other forebrain neurons.
Considerable evidence suggests that certain stressful stimuli are capable of activating the pituitary-adrenal system in rats with surgically-interrupted central neural connections of the medial basal hypothalamus (MBH). Thus, rats with chronically-isolated MBH show significant increases in plasma corticosterone levels following exposure to a variety of stressful stimuli [H alász at a!., 1967; F eldm an et al„ 1968, 1970; P alk a et al., 1969; G reer 1 Supported by USPHS grant AM 16794 from the National Institutes of Health. 2 Supported by USPHS postdoctoral fellowship IF 22 AM 00337 from the National Institutes of Health.
Downloaded by: Nagoya University 133.6.82.173 - 1/10/2019 10:14:42 PM
Received: June 17th, 1975; revised MS accepted: October 28th, 1975.
340
R ice/ K rön ing/C r itch low
Downloaded by: Nagoya University 133.6.82.173 - 1/10/2019 10:14:42 PM
et al., 1970; M akara et al., 1970], Results from acute studies utilizing partial forebrain removal consistently indicate that the isolated MBH-pituitary unit can support pituitary-adrenal responses to combined ether and leg break [M atsuda el al., 1963], ether-blood withdrawal or immobilization [D u n n and C r it c h l o w , 1969a], Hypothalamic structures, exclusive of those con tained in the median eminence, may not be required for such responses to some stimuli. M atsuda el al. [1964] reported that rats with median eminencepituitary islands (MEPI) showed elevated corticosterone levels following ether-leg break stress and suggested that the median eminence-pituitary unit constituted the minimal substrate for this response. Similarly, M akara et al. [1970] reported that rats with MEPI showed corticosterone responses to i.p. injections of histamine and insulin. The median eminence, however, does not appear essential in all situations. Ablation of the medial hypothalamus, including median eminence, is compatible with acute corticosterone responses to injection of E. coli endotoxin or large doses of formaldehyde [S ta rk et al., 1973/74], despite pituitary infarction. Furthermore, W itorsch and B rodish [1972] report that removal of the entire forebrain, including median emi nence, does not block corticosterone responses to certain combined stress procedures. These findings suggest that central neural pathways to the MBH, and perhaps the MBH itself, are not essential to the triggering of ACTH secretion by some noxious stimuli. The mechanisms involved in such pi tuitary activation are unknown, and their importance and role in the physio logical control of pituitary-adrenal function are undetermined. This laboratory was previously unable to elicit plasma corticosterone re sponses to ether or immobilization stress from pituitary islands in rats sub jected to total forebrain removal [D u n n and C r it c h l o w , 1969a]. Subsequent unpublished observations, however, suggested that acute corticosterone re sponses to immobilization-blood withdrawal persisted when the median emi nence was retained in connection with the pituitary; the magnitude of such responses was similar to that seen earlier in island preparations containing basal hypothalamic tissue [D u n n and C r it c h l o w , 1969a], The purpose of the present studies was to further explore the potential of the median emi nence-pituitary unit to support acute changes in ACTH secretion in response to stress and to assess the contribution of MBH tissue to such responses. Changes in plasma corticosterone levels were used as an index of changes in ACTH secretion. Because immobilization acutely activates the pituitaryadrenal system in rats with basal hypothalamic islands [D u n n and C r it c h l o w , 1969a], this stress procedure was used throughout the experiments. A secondary aim was to determine whether stress-induced alterations in pi-
Stress and Median Eminence-Pituitary islands
341
tuitary secretion of growth hormone (GH) persist in these preparations. In this species, stress causes an acute, marked decrease in circulating GH levels [S c h a lc h and R e ic h l in , 1966; G arcia and G e sc h w in d , 1968].
Materials and Methods
Downloaded by: Nagoya University 133.6.82.173 - 1/10/2019 10:14:42 PM
Adult female rats (Charles River, CD), weighing between 200 and 280 g, were used. All rats were maintained in animal quarters under conditions of controlled lighting (05.00-18.00 h) and temperature (26 ± 2 C) for at least 3 weeks prior to experimentation. Except where specified, Purina Lab Chow and tap water were available ad libitum. The 1st experiment was designed to assess pituitary-adrenal responses to immobiliza tion-blood withdrawal in MEPI rats and to compare these responses with those obtained in BHPI rats and in intact controls. Animals were studied approximately 24 h after surgery. Forebrain removal and brain stem transection were performed under ether anesthesia as described by M atsuda et al. [1963] and D unn and C ritchlow [1969a). In BHPI prepara tions, the brain stem was transected at the level of the superior colliculus and most of the forebrain was removed by aspiration, leaving tissue in the ventromedial-arcuate region. In MEPI rats, the entire forebrain was removed except for the median eminence which was retained in connection with the pituitary. The rats were placed in a 95% 0 2-5% C 0 2 environment following surgery until 17.00 h. Rectal temperatures were monitored periodi cally and cold water immersion was applied when temperatures exceeded 39 C. Because rats with forebrains removed did not eat spontaneously, food was withheld from intact controls and all rats were given i.p. injections of 5 ml of 5% dextrose-saline at 17.00 h. Approximately 24 h after surgery, rats were subjected to the previously described double bleeding procedure [Z immermann and C ritchlow , 1967] to assess 'non-stress’ and stress corticosterone levels, i.c., concentrations before and after immobilization-blood with drawal. This procedure consisted of rapidly immobilizing the rats for 3 min in the supine position, exposing the external jugular vein following procainization of the overlying skin and removing 0.5 ml of blood within 3 min of initial handling. These initial blood samples were treated as described below for determining ‘non-stress’ plasma corticosterone levels. 15 min after initiation of immobilization, a 2nd blood sample was obtained following decapitation for determining stress levels of corticosterone. Blood was collected in EDTArinsed syringes or beakers and centrifuged. Plasma was separated for determination of corticosterone concentrations by the microfiuorometric method of G lick et al. [1964], Because of the results obtained in the above experiment, a 2nd study was performed to determine whether MEPI rats are capable of responding to a 2nd immobilization-blood withdrawal stress. Beginning at 09.00 h, approximately 24 h after surgery, the animals were immobilized for 3 min and "non-stress’ blood samples were obtained as described above. 13 min after initiation of stress, rats were immobilized again and a 2nd or stress sample was collected from the jugular vein at 15 min. The rats were then placed in a holding cage and studied 6 h later; blood samples were collected within 3 min of initiation of immobilization and decapitation was performed at 15 min. Some of the blood samples from MEPI rats in the above experiments were also used for determining 'non-stress’ and stress GH levels. In our experience (unpublished observa tions), significant changes in plasma GH concentration do not occur prior to 3 min after
342
R ice/K roning /C ritchlow
initiation of stress and nearly maximal responses are obtained at 15 min. The blood was chilled immediately after collection and centrifuged. Aliquots of plasma were rapidly frozen and stored at 15 C for measurement of radioimmunoassayable GH, using the NIH kit and the methods described by Schindler el at. [1972], The sensitivity of the assay was approximately 0.2 ng and the coefficient of variability for plasma pools was 13% between and 7% within assays. A final study was designed to determine whether the corticosterone and GH responses observed in MEPI rats could be elicited by immobilization per se, i.e., without the stimuli associated with blood sampling. Surgery was performed as described above. Approximately 24 h later, half the animals were decapitated within 20 sec of handling and blood was collected for determination of 'non-stress' levels of both corticosterone and GH. The remaining animals were immobilized for 3 min and decapitated at 15 min. Blood was collected for measuring stress levels of both hormones. Treatments were assigned and data analyzed according to a randomized design. Twoway analysis of variance (ANOV) for repeated measures [W iner, 1962] and the paired /-test were used to analyze data from experiments involving both 'non-stress' and stress blood samples from individual rats. Comparison of stress-induced increments was per formed by one-way ANOV and Newman-Keuls’ test. Sludent’s /-test was used in the last experiment which involved single data points from individual animals. Probabilities less than 0.05 were considered statistically significant. Heads were fixed in 10% formalin following decapitation, decalcified in formic acidsodium citrate solution and subsequently processed for histological examination of in situ pituitary and hypothalamus. Data were discarded from rats with pituitary infarcts and from MEPI rats with remnants of hypothalamic tissue dorsal to the median eminence.
As shown in figure 1, the ‘non-stress’ corticosterone levels were higher (p
