Agents Actions37 (1992)
0065-4299/92/040268-05 $1.50+ 0.20/0 ~) 1992 Birkhfiuser Verlag, Basel
Effect of clonidine on experimental brain edema in the rat B. Robert, J. Y. Petit, N. Grimaud, M. Jug~ and L. Welin Department of Pharmacologyand Pharmacokinetics,Faculty of Pharmacy,Universityof Nantes, 44035 Nantes, France
Abstract Several experimental brain edema models are currently available for drug evaluation. Brain edemas are essentially vasogenic and/or cytotoxic, and eicosanoids are involved in the development of these edemas. Thus, a new model developed in our laboratory, which was obtained by phospholipase A 2 intracerebral injection was used to study the antiinfammatory effect of clonidine. The copper wire edema model was chosen as reference. Edemas were evaluated by determining the swelling and Na + and K + tissue concentrations of each hemisphere. Drugs were administered intraperitoneally. Dexamethasone was the only drug to inhibit copper wire-induced edema, whereas indomethacin and clonidine as well as dexamethasone exhibited marked antiedematous activity in our model. The effect of clonidine, which could be inhibited by prior administration of yohimbine, suggests that central %adrenergic stimulation is involved in reducing experimental brain edema. Introduction Vasogenic or cytotoxic brain edemas rapidly induce tissue alterations, causing disruption of the blood brain barrier and vasomotor paralysis followed by brain ischemia. Biochemical disturbances also occur as a result of membrane lesions, with subsequent swelling of astrocytes and decreased N a - K ATPase and oxidative phosphorylation enzymes activity [1, 2]. Eicosanoids are involved in these disturbances [3-5], essentially during the ischemic phase before edema formation [6 8]. Our experiments concerned the efficacy of certain antiinflammatory molecules in the treatment of brain edema. Clinically, only high doses of glucocorticoids can reduce brain edema and in most experimental models nonsteroidal drugs have proved rather inactive [9]. Our brain edema model allows better evaluation of the effects of some new molecules. To date, a variety of models have been
used to induce edema: physical agents, procedures involving cryogenic lesions, fi-irradiation, mechanical compression of the hemispheres or intracerebral injection of water or chemicals such as arachidonic acid, kainic acid, triethyltin and implantation of copper wires or tumor cells. Methods involving cerebral administration of enzymes implicated in infammatory reaction have not yet been experimented. Clonidine has a peripheral antiinflammatory effect on carragenin-induced rat-paw edema [10]. This effect is apparently related to activation of %adrenergic receptors [11], since it is antagonized by low doses of yohimbine. As the antihypertensive activity of clonidine involves central nervous system mechanisms, we sought to determine whether the presence of this molecule in brain tissue would induce an antiinfammatory or antiedematous effect. Two brain edema models were used: copper wire implantation
Agents Actions 37 (1992)
and intraparenchymal injection of phospholipase A 2 (PLA2).
Materials and methods Brain copper wire implantations and PLA 2 injections were performed under ketamine anesthesia (175 mg/kg) in the right hemisphere of male Wistar CF rats placed in a stereotaxic apparatus. Body temperature was maintained at 37 ~ by a heating pad. Rats (300_+ 20 g) were used for copper wire-induced edema according to the method of Levine et al. [-91. A copper wire (3ram in length, 25_+5mg) was introduced into the right hemisphere, in the frontal cortex brain structure, according to stereotaxic coordinates (A = + 3.5 ram, L = 2.5 mm with respect to the bregma). Drugs were injected intraperitoneally at 10a.m. and 5p.m. each day for a total dose of 26 tamol/kg/24 h for dexamethasone and indomethacin and 4.3 ~tmol/kg/24 h for clonidine, with treatment beginning on the day of implantation. These doses were selected in accordance with previous works [11,. 12]. Rats were anaesthetized with ether 72 h after copper wire implantation, and the entire brain was rapidly removed. An edema model developed in our laboratory was obtained by unilateral intracerebral injection of 2 IU of PLA 2 (10 IU/50gl of tris buffer 100 ~tmol/l, CaCI2 1 gmol/l solution, pH 7.4) according to the stereotaxic coordinates A -- + 2 ram, L = 2 ram, H = 3.5ram with respect to the bregma. In these conditions, the injection point was the frontal cortex-~corpus callosum junction. Macroscopic and microscopic examinations showed that PLA2 induced both vasogenic and cytotoxic lesions, with disruption of the blood brain barrier. Edema developed rapidly, with a maximal phase 24h after induction, whereas 72h were necessary with the copper wire implantation technique. Brain was quickly removed 24 h after PLA 2 injection. Drug doses were the same as those used in the copper wire-induced edema model. After removal, brain hemispheres were immediately separated, weighed and then desiccated at 110~'C for 24 h. Water content was calculated as the ratio Wf- Wd/Wr, where Wf is the fresh weight and Wd the dry weight of each hemisphere. The "swelling" (or percentage of change in weight) of the impaired right hemisphere as compared with the normal left one was determined according to the method of
269 Elliot and Jasper [13] by evaluation of the ratio 100 ( R - L ) / 1 0 0 - R , where R and L are, respectively the water content (%) of the right and left hemispheres. After desiccation, grinding and nitric acid extraction, concentrations of Na + and K + ions (mEq/kg dry tissue) in each hemisphere were evaluated by emission photometry (Hitachi 180-80 model). After Na+/K + ratios were calculated for impaired (I) and unimpaired (U) hemispheres of each rat, A (Na+/K +) was determined A(Na+/K + ) -
Na +/K + ( I ) - Na +/K + (U) Na+/K+(U)
• 100
Five groups of animals were used to study copper wire edema: (a) sham-operated rats which received either an intracerebral injection of 10~tl tris buffer for the PLAz-induced edema model or a copper wire which was then immediately removed for the other model; (b) control rats; (c) rats treated with dexamethasone; (d) rats treated with indomethacin; and (e) rats treated with clonidine. For the study of PLA2-induced edema, two other groups were added: (f) rats treated only with yohimbine; and (g) rats treated with clonidine and yohimbine (yohimbine injected 30 rain before clonidine). The cardiovascular effect of clonidine was evaluated by measurement of arterial blood pressure in rats identical to those used for edema induction. Animals were anaesthetized with ketamine 175 mg/kg i.m.), and carotid artery blood pressure was monitored through a transducer (Statham P23Db) and recorded on a polygraph. Resting blood pressure was measured in control rats and rats receiving clonidine (4.3 ~tmol/kg/24 h) according to the same time schedule as for edema treat1Tlent.
Results As brain edema did not develop in sham-operated rats, the copper wire or enzyme was obviously the direct cause of inflammatory reaction. The swelling induced by PLA2 intracerebral injection was greater than that of copper wire edema (Fig. 1) and developed more rapidly (maximal phases, respectively, 24 and 72h). Likewise, the ionic variations expressed by A(Na§ § were more marked for PLA 2 (Table 1). Moreover, PLA zinduced edema was inhibited by high doses of dexamethasone or indomethacin. Accordingly, PLAz-induced edema was chosen for experiments
Agents Actions37 (1992)
270 Table 1
Effects of differenttreatmentson variationsof A(Na+/K+) in copper wire- or PLA2-inducededemas. C
DEXA
INDO
CLO
YO
Copper wire edema
(15) 55.2_+9.0
(15) 23.5_+7.8*
(9) 44.6_+9.7
(10) 39.7_+14.5
--
PLAz-induced edema
(18) 95.3_+13.0
(15) 54.0•
(17) 24.2_+7.2***
(10) 7.9+11.4"**
(13) 57.6_+7.0*
CLO + YO
(10) 59.7_+8.8
C: control rats, DEXA: dexamethasone (26 gmol/kg/24h), INDO: indomethacin (26 gmol/kg/24h), CLO: clonidine (4.3 gmol/kg/24h), YO: yohimbine(5.2lamol/kg/24h), CLO + YO: clonidine(4.3 gmol/kg/24h)+ yohimbine(5.2lamol/kg/24h); Mean _+SEM; number of determinationsin brackets; * p < 0.05, *** p < 0.001 (Student's t-test). A(Na+/K+) calculated with Na + and K + tissueconcentrations(mEq/kg dry tissue)
(15)
(I0)
(12)
12
"[
..a
::
4
(17)
*
!!!!ii
0
-
S
C
*(10)
-
DEXA
INDO
CLO
Figure 1
Effectsof differentdrugson experimentalbrain edemain the rat. Swelling: relative percentage change in weight of the lesioned hemisphere, [] copper wire edema, Iii PLA2-inducededema; S: sham-operated rats. C: control rats, DEXA: treated with dexamethasone (26 lamol/kg/24h), INDO: treated with indomethacin (26 gmol/kg/24h), CLO: treated with clonidine (4.3~tmol/kg/24h); number of animals in brackets; **p < 0.01, 9** p < 0.001 comparativelyto control rats (Student'st-test).
to determine whether clonidine could produce cerebral antiedematous activity. Our results show that clonidine, dexamethasone and indomethacin markedly reduced swelling when edema was induced by PLA 2 injection (respectively, 2 . 6 + 1 % , 6.2+0.6% and 3.6+0.9% vs. 11.7 _+ 1.5% for control rats). The effect of clonidine was very intense despite the small injected dose (1/6th of the dexamethasone dose). Copper wire-induced edema was more difficult to inhibit since only dexamethasone showed significant activity. The variations of A(Na+/K +) were lower for the copper wire model than for PLA2-induced edema (Table 1). For the latter, the reductive effects of dexamethasone, indomethacin and clonidine were
comparable for swelling, and dexamethasone showed weak activity. The A(Na+/K +) reduction was only significant (P
0065-4299/92/040268-05 $1.50+ 0.20/0 ~) 1992 Birkhfiuser Verlag, Basel
Effect of clonidine on experimental brain edema in the rat B. Robert, J. Y. Petit, N. Grimaud, M. Jug~ and L. Welin Department of Pharmacologyand Pharmacokinetics,Faculty of Pharmacy,Universityof Nantes, 44035 Nantes, France
Abstract Several experimental brain edema models are currently available for drug evaluation. Brain edemas are essentially vasogenic and/or cytotoxic, and eicosanoids are involved in the development of these edemas. Thus, a new model developed in our laboratory, which was obtained by phospholipase A 2 intracerebral injection was used to study the antiinfammatory effect of clonidine. The copper wire edema model was chosen as reference. Edemas were evaluated by determining the swelling and Na + and K + tissue concentrations of each hemisphere. Drugs were administered intraperitoneally. Dexamethasone was the only drug to inhibit copper wire-induced edema, whereas indomethacin and clonidine as well as dexamethasone exhibited marked antiedematous activity in our model. The effect of clonidine, which could be inhibited by prior administration of yohimbine, suggests that central %adrenergic stimulation is involved in reducing experimental brain edema. Introduction Vasogenic or cytotoxic brain edemas rapidly induce tissue alterations, causing disruption of the blood brain barrier and vasomotor paralysis followed by brain ischemia. Biochemical disturbances also occur as a result of membrane lesions, with subsequent swelling of astrocytes and decreased N a - K ATPase and oxidative phosphorylation enzymes activity [1, 2]. Eicosanoids are involved in these disturbances [3-5], essentially during the ischemic phase before edema formation [6 8]. Our experiments concerned the efficacy of certain antiinflammatory molecules in the treatment of brain edema. Clinically, only high doses of glucocorticoids can reduce brain edema and in most experimental models nonsteroidal drugs have proved rather inactive [9]. Our brain edema model allows better evaluation of the effects of some new molecules. To date, a variety of models have been
used to induce edema: physical agents, procedures involving cryogenic lesions, fi-irradiation, mechanical compression of the hemispheres or intracerebral injection of water or chemicals such as arachidonic acid, kainic acid, triethyltin and implantation of copper wires or tumor cells. Methods involving cerebral administration of enzymes implicated in infammatory reaction have not yet been experimented. Clonidine has a peripheral antiinflammatory effect on carragenin-induced rat-paw edema [10]. This effect is apparently related to activation of %adrenergic receptors [11], since it is antagonized by low doses of yohimbine. As the antihypertensive activity of clonidine involves central nervous system mechanisms, we sought to determine whether the presence of this molecule in brain tissue would induce an antiinfammatory or antiedematous effect. Two brain edema models were used: copper wire implantation
Agents Actions 37 (1992)
and intraparenchymal injection of phospholipase A 2 (PLA2).
Materials and methods Brain copper wire implantations and PLA 2 injections were performed under ketamine anesthesia (175 mg/kg) in the right hemisphere of male Wistar CF rats placed in a stereotaxic apparatus. Body temperature was maintained at 37 ~ by a heating pad. Rats (300_+ 20 g) were used for copper wire-induced edema according to the method of Levine et al. [-91. A copper wire (3ram in length, 25_+5mg) was introduced into the right hemisphere, in the frontal cortex brain structure, according to stereotaxic coordinates (A = + 3.5 ram, L = 2.5 mm with respect to the bregma). Drugs were injected intraperitoneally at 10a.m. and 5p.m. each day for a total dose of 26 tamol/kg/24 h for dexamethasone and indomethacin and 4.3 ~tmol/kg/24 h for clonidine, with treatment beginning on the day of implantation. These doses were selected in accordance with previous works [11,. 12]. Rats were anaesthetized with ether 72 h after copper wire implantation, and the entire brain was rapidly removed. An edema model developed in our laboratory was obtained by unilateral intracerebral injection of 2 IU of PLA 2 (10 IU/50gl of tris buffer 100 ~tmol/l, CaCI2 1 gmol/l solution, pH 7.4) according to the stereotaxic coordinates A -- + 2 ram, L = 2 ram, H = 3.5ram with respect to the bregma. In these conditions, the injection point was the frontal cortex-~corpus callosum junction. Macroscopic and microscopic examinations showed that PLA2 induced both vasogenic and cytotoxic lesions, with disruption of the blood brain barrier. Edema developed rapidly, with a maximal phase 24h after induction, whereas 72h were necessary with the copper wire implantation technique. Brain was quickly removed 24 h after PLA 2 injection. Drug doses were the same as those used in the copper wire-induced edema model. After removal, brain hemispheres were immediately separated, weighed and then desiccated at 110~'C for 24 h. Water content was calculated as the ratio Wf- Wd/Wr, where Wf is the fresh weight and Wd the dry weight of each hemisphere. The "swelling" (or percentage of change in weight) of the impaired right hemisphere as compared with the normal left one was determined according to the method of
269 Elliot and Jasper [13] by evaluation of the ratio 100 ( R - L ) / 1 0 0 - R , where R and L are, respectively the water content (%) of the right and left hemispheres. After desiccation, grinding and nitric acid extraction, concentrations of Na + and K + ions (mEq/kg dry tissue) in each hemisphere were evaluated by emission photometry (Hitachi 180-80 model). After Na+/K + ratios were calculated for impaired (I) and unimpaired (U) hemispheres of each rat, A (Na+/K +) was determined A(Na+/K + ) -
Na +/K + ( I ) - Na +/K + (U) Na+/K+(U)
• 100
Five groups of animals were used to study copper wire edema: (a) sham-operated rats which received either an intracerebral injection of 10~tl tris buffer for the PLAz-induced edema model or a copper wire which was then immediately removed for the other model; (b) control rats; (c) rats treated with dexamethasone; (d) rats treated with indomethacin; and (e) rats treated with clonidine. For the study of PLA2-induced edema, two other groups were added: (f) rats treated only with yohimbine; and (g) rats treated with clonidine and yohimbine (yohimbine injected 30 rain before clonidine). The cardiovascular effect of clonidine was evaluated by measurement of arterial blood pressure in rats identical to those used for edema induction. Animals were anaesthetized with ketamine 175 mg/kg i.m.), and carotid artery blood pressure was monitored through a transducer (Statham P23Db) and recorded on a polygraph. Resting blood pressure was measured in control rats and rats receiving clonidine (4.3 ~tmol/kg/24 h) according to the same time schedule as for edema treat1Tlent.
Results As brain edema did not develop in sham-operated rats, the copper wire or enzyme was obviously the direct cause of inflammatory reaction. The swelling induced by PLA2 intracerebral injection was greater than that of copper wire edema (Fig. 1) and developed more rapidly (maximal phases, respectively, 24 and 72h). Likewise, the ionic variations expressed by A(Na§ § were more marked for PLA 2 (Table 1). Moreover, PLA zinduced edema was inhibited by high doses of dexamethasone or indomethacin. Accordingly, PLAz-induced edema was chosen for experiments
Agents Actions37 (1992)
270 Table 1
Effects of differenttreatmentson variationsof A(Na+/K+) in copper wire- or PLA2-inducededemas. C
DEXA
INDO
CLO
YO
Copper wire edema
(15) 55.2_+9.0
(15) 23.5_+7.8*
(9) 44.6_+9.7
(10) 39.7_+14.5
--
PLAz-induced edema
(18) 95.3_+13.0
(15) 54.0•
(17) 24.2_+7.2***
(10) 7.9+11.4"**
(13) 57.6_+7.0*
CLO + YO
(10) 59.7_+8.8
C: control rats, DEXA: dexamethasone (26 gmol/kg/24h), INDO: indomethacin (26 gmol/kg/24h), CLO: clonidine (4.3 gmol/kg/24h), YO: yohimbine(5.2lamol/kg/24h), CLO + YO: clonidine(4.3 gmol/kg/24h)+ yohimbine(5.2lamol/kg/24h); Mean _+SEM; number of determinationsin brackets; * p < 0.05, *** p < 0.001 (Student's t-test). A(Na+/K+) calculated with Na + and K + tissueconcentrations(mEq/kg dry tissue)
(15)
(I0)
(12)
12
"[
..a
::
4
(17)
*
!!!!ii
0
-
S
C
*(10)
-
DEXA
INDO
CLO
Figure 1
Effectsof differentdrugson experimentalbrain edemain the rat. Swelling: relative percentage change in weight of the lesioned hemisphere, [] copper wire edema, Iii PLA2-inducededema; S: sham-operated rats. C: control rats, DEXA: treated with dexamethasone (26 lamol/kg/24h), INDO: treated with indomethacin (26 gmol/kg/24h), CLO: treated with clonidine (4.3~tmol/kg/24h); number of animals in brackets; **p < 0.01, 9** p < 0.001 comparativelyto control rats (Student'st-test).
to determine whether clonidine could produce cerebral antiedematous activity. Our results show that clonidine, dexamethasone and indomethacin markedly reduced swelling when edema was induced by PLA 2 injection (respectively, 2 . 6 + 1 % , 6.2+0.6% and 3.6+0.9% vs. 11.7 _+ 1.5% for control rats). The effect of clonidine was very intense despite the small injected dose (1/6th of the dexamethasone dose). Copper wire-induced edema was more difficult to inhibit since only dexamethasone showed significant activity. The variations of A(Na+/K +) were lower for the copper wire model than for PLA2-induced edema (Table 1). For the latter, the reductive effects of dexamethasone, indomethacin and clonidine were
comparable for swelling, and dexamethasone showed weak activity. The A(Na+/K +) reduction was only significant (P
