139
Brain Research, 546 (1991) 139-142 © 1991 Elsevier Science Publishers B.V. 0006-8993/91/$03.50 A D ONIS 0006899391246073
BRES 24607
Short Communications
Corticotropin releasing hormone, interleukin-la, and tumor necrosis factor-a share characteristics of stress mediators Mauro Bianchi, Paola Sacerdote, Luisa Locatelli, Paolo Mantegazza and Alberto E. Panerai Department of Pharmacology, University of Milano, Milan (Italy)
(Accepted 8 January 1991) Key words: Corticotropin releasing hormone; Interleukin-1; Tumor necrosis factor; Stress; Naloxone; Analgesia
Interleukin-la and tumor necrosis factor-a induce an increase in pain thresholds in the rat. We now show that also the corticotropin releasing hormone induces an analgesic effect that, similarly to what is observed with the two cytokines, is not reversible by naloxone. Moreover, we also show that after the administration of interleukin-la, tumor necrosis factor-a, and corticotropin releasing hormone, naioxone becomes analgesic itself. A similar observation was also made in the human and the experimental animal after exposure to stressful conditions. The results obtained suggest that the two cytokines share with corticotropin releasing hormone some characteristics of stress mediators. A 3 min footshock, a stressful condition for the rat, induces an increase of analgesic thresholds that is not reversed by the opiate antagonist naloxone 9'18. Interestingly, after the fading of analgesia, naloxone becomes analgesic by itself 13"14. An analgesic effect of naloxone has been shown also in the human, e.g. after surgical stress 5'~5. These experimental evidences suggest that responses to naloxone might be modified following stress. Corticotropin releasing hormone (CRH) is widely recognized as the main mediator of cognitive stress, and initiates a cascade of events that involves primarily the opioid peptide fl-endorphin (BE) and adrenocorticotropin (ACTH) 4. However, some non-cognitive stressful conditions, such as infections, have been shown to be recognized by the brain 3. It has been recently proposed that cytokines such as interleukin-la (IL-1), and tumor necrosis factor-a (TNF) might initiate, in response to non-cognitive stress, a cascade of events similar to that initiated by C R H a'3. Consistently with this hypothesis, both IL-1 and TNF can induce BE, and A C T H synthesis and release, as well as analgesia 1'2'12. We previously showed that the time pattern of the analgesic responses after the intracerebroventricular (i.c.v.) injection of IL- 1, and TNF is comparable to that of naloxone nonreversible analgesia induced by footshock 9"18. We wondered whether CRH might induce a naloxone non-reversible analgesia similar to that observed after footshock, or TNF and IL-1 administration. Moreover, in order to further evaluate the hypothesis that these cytokines, similarly to CRH, behave as stress mediators,
we investigated whether naloxone could induce analgesia after TNF, IL-1, or C R H administration. Sprague-Dawley CD male rats, 10 in each experimental group, had polyethylene cannulae inserted in their lateral ventricles under pentobarbital anesthesia (45.0 mg/kg), with a method previously described 7. Experiments were conducted at least 5 days after surgery. Rat C R H (Peninsula, St. Helens, G.B.) at the dose of 500.0 ng; human rlL-1 (generous gift of Dr. Lomedico, Hoffmann La Roche, Nutley, NJ) at the dose of 5.0 ng; murine rTNF (Biogen, Gent, Belgium) at the dose of 1.0 ng were administered i.c.v. Saline was administered i.c.v. in control animals. Naloxone (S.I.EA.C., Como, Italy) at a dose of 10 mg/kg or saline were administered intraperitoneally. All experiments were conducted in the morning by the same person who was not aware of the treatments, nor the goals of the study. Changes in pain thresholds were measured by the hot plate test 1° and expressed as percentage of the Maximal Possible Effect (% MPE). Maximal Possible Effect expresses the equation [(TL-BL)/(ML-BL)] x 100, where BL is the mean basal latency measured before the first treatment was applied (5.5-7.5 s); TL is the test latency measured after treatments; ML is the maximal latency accepted (15 s), chosen in order to avoid tissue damage to the foot pads. Statistical analysis of results was performed by the Kruskall-Wallis analysis of variance for non-parametric values 9.
Correspondence: A.E. Panerai, Department of Pharmacology, University of Milano, Via Vanvitelli 32, 20129, Milan, Italy.
140 In a first experiment, CRH, IL-1 or TNF was administered together with naloxone or saline. Nociceptive thresholds were evaluated by the hot plate test at 1, 3, 5, 10, 15, 20, 25, and 30 min after IL-1, TNF, and 1, 3, 5, 10, 15, 20, 25, 30, 35, 40, 45, and 50 min after CRH administration. In a second experiment, naloxone or saline were administered 10 min after IL-1 or TNF, and 20 min after CRH. Nociceptive thresholds were measured at the time intervals as in Exp. 1. In order to evaluate whether the exposure to hot plate played a role in the increase of nociceptive thresholds induced by naloxone, in a further experiment, rats were administered the same treatments as in Exp. 2, but were placed on a plate similar to that used for the hot plate test, maintained at room temperature (22 °C). Nociceptive thresholds were measured on a hot plate starting at the time of naloxone administration. In order to evaluate possible modifications of naloxone binding, rats administered saline, CRH 500 ng, TNF 1.0 rig, or IL-1 5.0 ng i.c.v., were decapitated 5 and 10 min after TNF or IL-1 administration, and 10 and 20 min after CRH. Hindbrain and midbraln were dissected and placed in cold phosphosaline buffer (PBS), pH 7.4. Tissues were placed in 9.0 ml/g cold 50 mM HEPES buffer pH 7.4, homogenized, centrifuged twice at 20,500 g for 10 min at 4 °C. Pellets were resuspended at room temperature in 9.0 ml/g PBS pH 7.4 for 15 min in order to get rid of endogenous ligands. Thereafter samples were centrifuged and pellets were frozen at -80 °C. For the binding assay, 100 /~! membrane aliquots containing 500-700 mg proteins were incubated in a total volume of 140/tl of ice-cold PBS pH 7.4 containing 17 ag/ml PMSF (phenyl-methyl-sulphonyl-fluoride). In saturation studies, [3H]naloxone (spec. act. 60 Ci/mmol) was added in final concentrations ranging from 0.03 to 2.5 nM in the presence or absence of 10-6 M naloxone. For displacement studies, different concentrations of cold naloxone (from 10-3 to 10-12 M) were added to a fixed amount of [3H]naloxone (2 nmol). After incubation for 60 min at 0 °C, the reaction was terminated by 1 ml ice-cold Tris-HCl 50 mM containing 1% bovine serum albumine (BSA), and samples filtered on GF-C Whatman filters. Statistical analysis of binding studies was performed by the E B D A / L I G A N D program s. The upper and middle panels of Fig. 1 (TNF), Fig. 2 (IL-1) and Fig. 3 (CRH) show that the cytokines and CRH induce a clear-cut and short-lasting increase in nociceptive thresholds. Naloxone, when administered together with the cytokines or CRH, did not block the increase of nociceptive thresholds, but maintained it elevated for a longer period, as is shown in the upper panel of Figs. 1, 2 and 3. Naloxone administered when
the nociceptive thresholds had returned to basal values (i.e. 10 min after TNF and IL-I, and 20 rain after CRH administration) induced a new increase of thresholds that lasted for at least 15 min, while saline was ineffective, as is shown in the middle panel of Figs. 1, 2 and 3. The lower panel of Figs. 1, 2 and 3 shows that also when rats were not placed on the heated hot plate after CRH, IL-1 or TNF administration, naloxone still induces an increase of nociceptive thresholds. Neither the Bma x n o r the K d of opiate receptors changed after the administration of CRH, TNF or IL-1 (data not shown). The results presented suggest that CRH, tL-1, and
IOO o A
80
O/'$k
m
T
oo
,o / / \
I
20 L p o
Brain Research, 546 (1991) 139-142 © 1991 Elsevier Science Publishers B.V. 0006-8993/91/$03.50 A D ONIS 0006899391246073
BRES 24607
Short Communications
Corticotropin releasing hormone, interleukin-la, and tumor necrosis factor-a share characteristics of stress mediators Mauro Bianchi, Paola Sacerdote, Luisa Locatelli, Paolo Mantegazza and Alberto E. Panerai Department of Pharmacology, University of Milano, Milan (Italy)
(Accepted 8 January 1991) Key words: Corticotropin releasing hormone; Interleukin-1; Tumor necrosis factor; Stress; Naloxone; Analgesia
Interleukin-la and tumor necrosis factor-a induce an increase in pain thresholds in the rat. We now show that also the corticotropin releasing hormone induces an analgesic effect that, similarly to what is observed with the two cytokines, is not reversible by naloxone. Moreover, we also show that after the administration of interleukin-la, tumor necrosis factor-a, and corticotropin releasing hormone, naioxone becomes analgesic itself. A similar observation was also made in the human and the experimental animal after exposure to stressful conditions. The results obtained suggest that the two cytokines share with corticotropin releasing hormone some characteristics of stress mediators. A 3 min footshock, a stressful condition for the rat, induces an increase of analgesic thresholds that is not reversed by the opiate antagonist naloxone 9'18. Interestingly, after the fading of analgesia, naloxone becomes analgesic by itself 13"14. An analgesic effect of naloxone has been shown also in the human, e.g. after surgical stress 5'~5. These experimental evidences suggest that responses to naloxone might be modified following stress. Corticotropin releasing hormone (CRH) is widely recognized as the main mediator of cognitive stress, and initiates a cascade of events that involves primarily the opioid peptide fl-endorphin (BE) and adrenocorticotropin (ACTH) 4. However, some non-cognitive stressful conditions, such as infections, have been shown to be recognized by the brain 3. It has been recently proposed that cytokines such as interleukin-la (IL-1), and tumor necrosis factor-a (TNF) might initiate, in response to non-cognitive stress, a cascade of events similar to that initiated by C R H a'3. Consistently with this hypothesis, both IL-1 and TNF can induce BE, and A C T H synthesis and release, as well as analgesia 1'2'12. We previously showed that the time pattern of the analgesic responses after the intracerebroventricular (i.c.v.) injection of IL- 1, and TNF is comparable to that of naloxone nonreversible analgesia induced by footshock 9"18. We wondered whether CRH might induce a naloxone non-reversible analgesia similar to that observed after footshock, or TNF and IL-1 administration. Moreover, in order to further evaluate the hypothesis that these cytokines, similarly to CRH, behave as stress mediators,
we investigated whether naloxone could induce analgesia after TNF, IL-1, or C R H administration. Sprague-Dawley CD male rats, 10 in each experimental group, had polyethylene cannulae inserted in their lateral ventricles under pentobarbital anesthesia (45.0 mg/kg), with a method previously described 7. Experiments were conducted at least 5 days after surgery. Rat C R H (Peninsula, St. Helens, G.B.) at the dose of 500.0 ng; human rlL-1 (generous gift of Dr. Lomedico, Hoffmann La Roche, Nutley, NJ) at the dose of 5.0 ng; murine rTNF (Biogen, Gent, Belgium) at the dose of 1.0 ng were administered i.c.v. Saline was administered i.c.v. in control animals. Naloxone (S.I.EA.C., Como, Italy) at a dose of 10 mg/kg or saline were administered intraperitoneally. All experiments were conducted in the morning by the same person who was not aware of the treatments, nor the goals of the study. Changes in pain thresholds were measured by the hot plate test 1° and expressed as percentage of the Maximal Possible Effect (% MPE). Maximal Possible Effect expresses the equation [(TL-BL)/(ML-BL)] x 100, where BL is the mean basal latency measured before the first treatment was applied (5.5-7.5 s); TL is the test latency measured after treatments; ML is the maximal latency accepted (15 s), chosen in order to avoid tissue damage to the foot pads. Statistical analysis of results was performed by the Kruskall-Wallis analysis of variance for non-parametric values 9.
Correspondence: A.E. Panerai, Department of Pharmacology, University of Milano, Via Vanvitelli 32, 20129, Milan, Italy.
140 In a first experiment, CRH, IL-1 or TNF was administered together with naloxone or saline. Nociceptive thresholds were evaluated by the hot plate test at 1, 3, 5, 10, 15, 20, 25, and 30 min after IL-1, TNF, and 1, 3, 5, 10, 15, 20, 25, 30, 35, 40, 45, and 50 min after CRH administration. In a second experiment, naloxone or saline were administered 10 min after IL-1 or TNF, and 20 min after CRH. Nociceptive thresholds were measured at the time intervals as in Exp. 1. In order to evaluate whether the exposure to hot plate played a role in the increase of nociceptive thresholds induced by naloxone, in a further experiment, rats were administered the same treatments as in Exp. 2, but were placed on a plate similar to that used for the hot plate test, maintained at room temperature (22 °C). Nociceptive thresholds were measured on a hot plate starting at the time of naloxone administration. In order to evaluate possible modifications of naloxone binding, rats administered saline, CRH 500 ng, TNF 1.0 rig, or IL-1 5.0 ng i.c.v., were decapitated 5 and 10 min after TNF or IL-1 administration, and 10 and 20 min after CRH. Hindbrain and midbraln were dissected and placed in cold phosphosaline buffer (PBS), pH 7.4. Tissues were placed in 9.0 ml/g cold 50 mM HEPES buffer pH 7.4, homogenized, centrifuged twice at 20,500 g for 10 min at 4 °C. Pellets were resuspended at room temperature in 9.0 ml/g PBS pH 7.4 for 15 min in order to get rid of endogenous ligands. Thereafter samples were centrifuged and pellets were frozen at -80 °C. For the binding assay, 100 /~! membrane aliquots containing 500-700 mg proteins were incubated in a total volume of 140/tl of ice-cold PBS pH 7.4 containing 17 ag/ml PMSF (phenyl-methyl-sulphonyl-fluoride). In saturation studies, [3H]naloxone (spec. act. 60 Ci/mmol) was added in final concentrations ranging from 0.03 to 2.5 nM in the presence or absence of 10-6 M naloxone. For displacement studies, different concentrations of cold naloxone (from 10-3 to 10-12 M) were added to a fixed amount of [3H]naloxone (2 nmol). After incubation for 60 min at 0 °C, the reaction was terminated by 1 ml ice-cold Tris-HCl 50 mM containing 1% bovine serum albumine (BSA), and samples filtered on GF-C Whatman filters. Statistical analysis of binding studies was performed by the E B D A / L I G A N D program s. The upper and middle panels of Fig. 1 (TNF), Fig. 2 (IL-1) and Fig. 3 (CRH) show that the cytokines and CRH induce a clear-cut and short-lasting increase in nociceptive thresholds. Naloxone, when administered together with the cytokines or CRH, did not block the increase of nociceptive thresholds, but maintained it elevated for a longer period, as is shown in the upper panel of Figs. 1, 2 and 3. Naloxone administered when
the nociceptive thresholds had returned to basal values (i.e. 10 min after TNF and IL-I, and 20 rain after CRH administration) induced a new increase of thresholds that lasted for at least 15 min, while saline was ineffective, as is shown in the middle panel of Figs. 1, 2 and 3. The lower panel of Figs. 1, 2 and 3 shows that also when rats were not placed on the heated hot plate after CRH, IL-1 or TNF administration, naloxone still induces an increase of nociceptive thresholds. Neither the Bma x n o r the K d of opiate receptors changed after the administration of CRH, TNF or IL-1 (data not shown). The results presented suggest that CRH, tL-1, and
IOO o A
80
O/'$k
m
T
oo
,o / / \
I
20 L p o
