- Email: [email protected]
The connection between absence-like seizures and hypothermia induced by penicillin: possible implication on other animal models of petit mal epilepsy
Brain Research 777 Ž1997. 86–94
Research report
The connection between absence-like seizures and hypothermia induced by penicillin: possible implication on other animal models of petit mal epilepsy Zoran S. Ostojic´ ) , Sabera Ruzdijic, ˇ ´ Marko Car, Ljubisav Rakic, ´ Rosica Veskov Institute for Biological Research, Department for Neurobiology and Immunology, UniÕersity of Belgrade, P.O. Box 11060, 29. NoÕembra 142, Belgrade, YugoslaÕia Accepted 29 July 1997
Abstract In this study we investigated the relationship between penicillin-induced hypothermia and petit mal epilepsy induced by this proconvulsant antibiotic. In order to find a possible dose-dependent relationship, we used two doses: 1500.000 and 1000.000 Urkg b.wt., both known as being sufficient to induce absence-like attacks with subsequent spike and wave discharges ŽSWD. in electrocorticogram ŽECoG.. Because of experimental data suggesting penicillin binding to benzodiazepine receptor recognition site, we also studied penicillin-induced changes in body temperature after diazepam pretreatment. Results of this study clearly show that penicillin in doses known to induce petit mal-like epilepsy concomitantly induces statistically significant dose-dependent decrease in body temperature. Pretreatment with diazepam completely prevents both penicillin-induced hypothermia and SWDs. On the other hand, both the diazepam and mixed diazepamq penicillin treatments did not significantly alter body temperature. These results suggest, however, that at least some of the penicillin effects described could be assigned to its binding to the benzodiazepine receptor recognition site at GABA A ionophore. This may have an important clinical implication because the inhibitory action of penicillin at the benzodiazepine receptor recognition site could account for the mechanism of penicillin-induced unspecific encephalopathies in humans. The relationship between petit mal epilepsy and hypothermia sheds new light on the action mechanisms of penicillin-induced absence seizures. q 1997 Elsevier Science B.V. Keywords: Penicillin; Diazepam; Absence seizure; Hypothermia; GABA; Thalamus
1. Introduction It has been generally accepted that brain cooling provides significant protection against ischemic brain damage during brain or heart surgery w4,11,20x. Profound hypothermia combined with total circulatory arrest has been in use for several years to correct cardiac defects in neonates and to facilitate repair due to extensive neurosurgical interventions Ža typical example is surgery for intracranial aneurysms.. Such delicate intervention would otherwise be impossible w10x. Recently, hypothermia has received increased interest because of strong evidence that mild reduction in body temperature could have a therapeutic role in reducing ischemic brain injury w9,15x. On the other hand, it is well known that many pharmacological agents induce changes in body temperature, )
Corresponding author. Fax: q38 Ž1. 11-761-433
0006-8993r97r$17.00 q 1997 Elsevier Science B.V. All rights reserved. PII S 0 0 0 6 - 8 9 9 3 Ž 9 7 . 0 1 0 0 7 - X
particularly in small laboratory animals but also in humans. Some drugs act directly on oxidative metabolic processes in brain tissue while other drugs affect the main thermoregulatory organs; thereafter, metabolic changes observed in the brain will be only an indirect consequence of these changes. Central structures of the thermoregulatory system are the preoptic Žanterior. and posterior regions of the hypothalamus, as well as the septum w1,7x. These hypothalamic neurons are responsible for both physiological and behavioral responses to temperature changes w7x. However, it was recently discovered that structures in the midline thalamus may play a role similar to that of specific hypothalamic regions in the processing of external and internal temperature information w8,41x. The role played by midline and other unspecific thalamic nuclei in the pathophysiology of absence seizures has been well documented w5,25,42x. At the same time, these and other thalamic nuclei have the unique feature that they can switch
Z.S. Ostojic´ et al.r Brain Research 777 (1997) 86–94
between two distinct modes of action in how they regulate the degree of alertness. Unspecific thalamic nuclei are responsible for generating both tonic burst firing in an awake state and phasic oscillatory activity during slowwave sleep w22x. Under pathological circumstances, spindling activity, as an electroencephalographical ŽEEG. marker of slow-wave sleep, may be transformed into spike and wave discharges ŽSWDs.. These rhythms are related in basic molecular, anatomical, and pharmacological profiles w12,16,17,24,38x. The intrinsic capacity of certain thalamic neurons, together with their possible participation in thermoregulatory processes, confirms the necessity for monitoring body temperature during the investigation of petit mal seizures in humans and experimental animals. In the last two decades numerous experimental models of petit mal-like epilepsy have been developed to serve as electrophysiological and behavioral archetype of absence seizures in humans. It has also been noted that some widely used drugs known to induce or exacerbate absence attacks Žpentylenetetrazole ŽPTZ., g-hydroxybutyrate ŽGHB., baclofen. at the same time decrease body temperature. Considerable evidence exists for the hypothermic effect of PTZ in dogs, rabbits, rats, mice w19x, and cats w3x, although it should be mentioned that the dosages employed were sufficient to induce other types of epilepsy. In 1978 it was demonstrated that another inducer ŽGHB. of petit mal-like epilepsy in experimental animals also exerts an influence upon the thermoregulatory system of monkeys; all monkeys that received GHB became profoundly hypothermic concomitantly with the EEG and behavioral changes w36x. Also, the selective GABA B agonist baclofen, when applied intraperitoneally Ži.p.. in 5–10 mgrkg doses, induced hypothermia in mice w18,23x. The next drug known to evoke absence seizures is the broad-spectrum antibiotic penicillin. High dose of parenterally applied penicillin in cats provides one of the best-suited animal models of human absence attacks w29x, and this is well known as the feline generalized penicillin epilepsy model. Later it was discovered that this antibiotic also induces absence attacks in rodents w30x. Penicillin acts as a non-competitive antagonist of GABA A receptors w13,26x, binding to the picrotoxin site at the GABA A receptor complex w26,32,40x. Other data also show penicillin binding to the benzodiazepine receptor recognition site at GABA A ionophore w2,35x. Thus far no report about the influence of convulsive doses of this agent on body temperature has been published. Concerning the relatively few and contradictory observations about the role of GABAergic receptors in thermoregulation, the primary goals of this study were to investigate the influence of high epileptogenic penicillin doses on body temperature as well as the possible parallel between petit mal-like epilepsy and decreased body temperature. We also examined penicillin-induced changes in body temperature after pretreatment with the GABA A potentiating anti-epileptic agent, diazepam, with respect to
87
experimental data that suggest penicillin binding to the benzodiazepine receptor recognition site.
2. Materials and methods 2.1. Animals The experiments were performed in male adult Ž3 months. Wistar rats with no previous drug history. Animals were freely moving and individually placed in a plexiglass recording chamber in an air-conditioned room under a 12-h lightrdark cycle, with water and standard laboratory food freely available. Room temperature was maintained at 22–248C. 2.2. Groups All rats were divided into five groups according to the drug injected. In the first Ž n s 25. and second groups Ž n s 15., dissolved crystalline penicillin was injected i.p. in doses of 1500.000 Urkg and 1000.000 Urkg b.wt., respectively. The third group Ž n s 10. was treated with a combination of 2 mgrkg diazepam and 1500.000 Urkg penicillin. Rats in the fourth group Ž n s 6. received 2 mgrkg diazepam i.p. In addition, in the fifth control Žplacebo. group Ž n s 10., rats were injected with an appropriate volume of saline solution. All statistical data were analyzed in respect to the placebo group. Five animals from each group treated with higher and lower dose of penicillin, as well as diazepam q penicillin Žfirst, second and third groups, amounting to a total of 15 animals. were surgically prepared with chronic cortical electrodes in order to detect the appearance of SWDs after administration of this drug and to reveal the suppressant effect of diazepam on the genesis and expression of penicillin-induced SWDs. 2.3. Surgical procedures Permanent epidural cortical electrodes were implanted under pentobarbital anesthesia Ž40 mgrkg, i.p.. to permit recording of the electrocorticogram ŽECoG. in freely moving animals. Two electrodes Žfine tips of stainless steel needle. were placed bilaterally over the frontal ŽA s 1.0 mm, L s 2.5–3 mm. cortex, keeping intact both the dura and cortex. Thus we avoided the appearance of latent epileptogenic foci which might be activated by high penicillin doses leading to development of focal or clonic epilepsy. An additional stainless steel screw electrode was driven into the skull above the cerebellum and served as a reference electrode. The animals were allowed to rest 1 week before control ECoG recordings were taken. All experiments were begun 1 day after the control ECoG
88
Z.S. Ostojic´ et al.r Brain Research 777 (1997) 86–94
recording. Only animals that did not show spontaneous SWDs were used. 2.4. ECoG recordings All recordings were made with the animals freely moving in shielded plexiglass recording chambers. ECoG signals were digitized with a sampling rate of 128 Hz and transmitted by shielded cable. The rats were allowed to habituate in a new environment for 30 min, and then the reference ECoG was recorded for the next 30 min prior to any drug administration. After drug injections, the ECoG was recorded throughout the experiment. For the combined injections Ždiazepam q penicillin., diazepam was injected and the ECoG was subsequently recorded for 5 min; thereafter, penicillin was given and the ECoG was continuously recorded for the next 150 min. 2.5. Temperature measurement During the experiments rats were kept in their plexiglass recording chambers in order to reduce any possible stress factors as well as baseline motor activity. All data sampling was also taken at the same time Ž10:00 h. to avoid sleep propensity. Core body temperature was measured rectally using a commercially available digital thermometer with an accuracy of "0.18C. Before measurement, the thermometer was placed in warm water Ž308C.; this simple procedure significantly decreased measuring time because the rats were kept restrained only for 8–10 s. Therefore, we prevented a rise in body temperature due to massive motor movement. Temperature alterations are shown as changes in body temperature from the basal level at t s 0. Basal levels were considered as a mean temperature computed for all animals included in the specified
group. Body temperature was continuously monitored for 5 h every 30 min after application of saline and both penicillin dose, and for 3 h every 30 min after application of diazepam and diazepam q penicillin. Exception was made during the first 30 min after application of both penicillin doses Žmeasurement of body temperature changes was performed at 15 min intervals. in order to detect the initial decrease in body temperature. Upon completion of the experiments, we performed occasional monitoring of body temperature during the next 48 h to find out the time interval needed for complete withdrawal of penicillin-induced hypothermia. 2.6. Data analysis Temperatures were presented as mean " S.E. and were analyzed using a 1-way analysis of variance ŽANOVA. for repeated measures. Sources of variation were divided into between-subjects factor ŽDrug., and within-subject effect. The latter includes the within-subject main effect ŽTime. and its interaction with the between-subjects factor ŽDrug = Time.. P - 0.05 was considered to indicate significant differences in changes in body temperature. For that purpose, we used SAS software package, general linear model procedure Žver. 6.1. with repeated statement. GreenhouseGeisser corrections were used for all repeated-measures analyses. Data were also baseline corrected at t s 0 Ž5 min before injections.. Statistical significance was tested between the placebo group and the group receiving specified drugs, or between two groups treated differentially. When significant Drug = Time interactions occurred, Tukey’s studentized range ŽHSD. tests were performed to determine the specific time points at which the drug-induced temperature response differed from that of the placebo or another groups.
Fig. 1. Changes in body temperature shown as mean " S.E. after application of two penicillin doses. Statistical differences were checked by Tukey’s studentized ŽHSD. range test between the corresponding time intervals of the penicillin Ž10 6 and 1.5 = 10 6 Urkg, respectively. and placebo sessions Ž a s 0.05, ) P - 0.05., and between that of 10 6 and 1.5 = 10 6 Urkg penicillin sessions Ž a s 0.05, q P - 0.05..
Z.S. Ostojic´ et al.r Brain Research 777 (1997) 86–94
3. Results The results of this study indicate that both doses of parenterally applied penicillin induce a dose-dependent decrease in core temperature ŽFig. 1.. Compared to placebo, application of 1500.000 Urkg penicillin induces highly significant alteration in body temperature ŽDrug = Time interaction: F10,380 s 6.76, P - 0.0001.. In addition, significant Time effect Ž F10,380 s 8.95, P - 0.0001. indicates temperature change over time. Higher penicillin dose induces an initial drop in body temperature of 0.88C Žfrom 37.43 " 0.068C to 36.63 " 0.118C; see Fig. 1. with a latency period of 15 min. Maximal level of hypothermia of y1.778C Žbody temperature: 35.66 " 0.288C. was reached 90 min after injection. In the next 90 min, a hypothermic plateau was formed at y1.748C and body temperature Ž35.69 " 0.318C. remained at this level. Since we found significant Drug = Time interaction, post hoc Tukey’s test revealed that both the initial and maximal decrease in body temperature were statistically significant, as were the all time points from 15 to 300 min ŽFig. 1.. The decrease in body temperature preceded and lasted beyond the ECoG markers of absence seizures manifested
89
by SWDs. The first epileptiform activity appeared within 25–35 min after drug application. This activity was characterized by bilaterally synchronous spike and wave complexes or single sharp waves ŽFig. 2.. The frequency of the SWDs ranged from 5 to 5.5 Hz and the duration of individual bursts varied from 2 to 7 s. Bursts of SWDs persisted almost 120 min after drug injection. On the contrary, complete restoration of body temperature in the animals having received penicillin injection with this dose was detected between 6 and 8 h. In six animals Ž24%., however, a long-lasting Ž) 24 h. and severe hypothermia Žbody temperature: 32.5–34.58C. developed. Fifteen minutes after injection of 1000.000 Urkg penicillin, mean normal body temperature Ž37.59 " 0.078C. decreased for 0.598C ŽFig. 1.. Thereafter, body temperature continued to fall and 45 min later Ž60 min postinjection period. further decreased for 0.308C Žbody temperature: 36.7 " 0.188C.. In the next 120 min, the hypothermic plateau has being formed with no or minimal change in body temperature ŽFig. 1.. Significant Drug = Time interaction also occurred Ž F10,280 s 3.72, P - 0.0001. as well as change in body temperature over Time Ž F10,280 s 5.12, P - 0.0001.. Tukey’s test revealed the presence of a statis-
Fig. 2. Generalized epileptiform activity elicited by i.p. injection of penicillin Ž1.5 = 10 6 Urkg b.wt.. manifested in typical spike and wave pattern. A: control recording Žsaline solution applied.; B: 70 min after application of penicillin. Horizontal line: 1 s, vertical line: 100 mV. Abbreviations in this and subsequent figures: LPCx, left parietal cortex; RPCx, right parietal cortex.
90
Z.S. Ostojic´ et al.r Brain Research 777 (1997) 86–94
Fig. 3. Generalized epileptiform activity elicited by i.p. injection of penicillin Ž10 6 Urkg b.wt.. manifested in typical spike and wave pattern. A: control ECoG recording Žsaline solution applied.; B: 40 min after application of penicillin. Horizontal line: 1 s, vertical line: 100 mV.
tically significant fall in body temperature at time points between 15 and 150 min. At 180 min and thereafter, however, there was no significant difference comparing
these time points with the placebo group, suggesting normalization of body temperature ŽFig. 1.. During the first 25–35 min, bilaterally synchronous SWDs with typical
Fig. 4. Changes in core temperature of each 30 min during the 180 min observation period in the control Žplacebo. and diazepam-treated conditions. There is no significant differences between the groups.
Z.S. Ostojic´ et al.r Brain Research 777 (1997) 86–94
91
Fig. 5. Changes in body temperature during the 180 min observations period between the corresponding time intervals of penicillin Ž10 6 and 1.5 = 10 6 Urkg. and paired application of diazepam Ž2 mgrkg. and penicillin Ž1.5 = 10 6 Urkg. 5 min afterward. Note significant differences between the mixed treatment and both penicillin doses. Tukey’s range test: a s 0.05, ) P - 0.05.
Fig. 6. ECoG changes observed after paired application of diazepam and penicillin 5 min afterward. A: control recording Žsaline solution applied.. B: 45 min after penicillin injection Ž50 min after diazepam pretreatment.. Note synchronized activity due to diazepam treatment and lack of penicillin-induced SWDs. Horizontal line: 1 s, vertical line: 100 mV.
Z.S. Ostojic´ et al.r Brain Research 777 (1997) 86–94
92
Fig. 7. Alteration in core temperature of each 30 min during the 180 min observation period induced by mixed treatment of diazepam Ž2 mgrkg. and penicillin.
morphology appeared in the ECoG ŽFig. 3.. The intraburst frequency, morphology and duration of individual bursts did not significantly differ from that induced by higher dose of penicillin. There was a difference, however, in the duration of discharges: in this group SWDs ceased almost 30 min earlier Žf 90 min after injection.. Time spent in hypothermia was also shortened and complete restoration of body temperature occurred 4–6 h after application of penicillin, although statistical testing revealed normalization of body temperature between 150 and 180 min after injection and subsequent lack of significant difference in relevance to the placebo treatment ŽFig. 1.. Statistical data also show that penicillin induces hypothermia in a dose-dependent manner, since we found a highly significant Drug = Time interaction Ž F10,380 s 3.88, P - 0.0001. and temperature alteration over Time Ž F10,380 s 14.07, P - 0.0001.. According to Tukey’s range test, changes in core rectal temperature induced by two penicillin doses appeared to be statistically significant at all time points between 30 and 300 min ŽFig. 1.. Table 1 Repeated-measures ANOVA about changes in core temperature Group Phigh vs. placebo Plow vs. placebo Phigh vs. Plow DPhigh vs. Phigh DPhigh vs. Plow DPhigh vs. placebo D vs. placebo
Time ŽF.
Ž P.
Time=Drug interaction ŽF. Ž P.
F10,380 s8.95 F10,280 s 5.12 F10,380 s14.07 F6,198 s8.97 F6,138 s 7.28 F6,138 s 0.97 F6,114 s 0.53
- 0.0001 - 0.0001 - 0.0001 - 0.0001 - 0.0001 ns ns
F10,380 s6.76 F10,280 s 3.72 F10,380 s 3.88 F6,198 s6.67 F6,138 s 3.18 F6,138 s1.64 F6,114 s 0.72
- 0.0001 - 0.0001 - 0.0001 - 0.0001 - 0.0059 ns ns
Abbreviations: Phigh and Plow , groups of animals that received 1.5=10 6 and 10 6 Urkg of penicillin, respectively; D, group of animals that received 2 mgrkg of diazepam; DPhigh , animals that received 1.5=10 6 Urkg of penicillin and pretreated with diazepam; ns, non-significant.
Finally, diazepam completely prevented the penicillininduced reduction of body temperature. Diazepam itself does not exert a statistically significant influence upon body temperature ŽFig. 4.. Accordingly, highly significant difference due to Drug treatment, Time and Drug = Time interaction exists for the first, second and fourth groups Žsee Table 1.. Also, Tukey’s test shows statistically significant differences at all measured time points Ž30–180 min; Fig. 5.. However, in this case, penicillin failed to induce SWDs; bilaterally synchronized rhythmic activity with a predominant frequency within the alpha band arises in the ECoG ŽFig. 6.. Moreover, according to Tukey’s test, mixed treatment with diazepam q penicillin does not significantly differ from that of placebo ŽFig. 7..
4. Discussion This study attempted to correlate hypothermia and absence epilepsy. It was shown that hypothermia clearly precedes ECoG hypersynchrony by 10–20 min. This time is probably somewhat higher because continuous monitoring of body temperature was not made between t s 0 and t s 15 min postinjection. Therefore, a critical level of hypothermia is required for development of SWDs which is reached approximately 30 min after application of penicillin. As our previous study showed, both mild Ž308C. and deep Ž14–178C. artificially induced hypothermia also evoked petit mal-like epilepsy with accompanying SWDs within the temperature range of 25–368C w28x. On the other hand, because of evidence suggesting the role of GABA B receptors in absence seizures w14,22,37,40x, as well as in hypothermia w18,23,34x, we suggest that a drop in body temperature Žfollowed by SWDs. could be consid-
Z.S. Ostojic´ et al.r Brain Research 777 (1997) 86–94
ered as a marker of GABA B receptor activation intensity. It is quite reasonable to conclude that mild-to-deep hypothermia, regardless of the means of induction, affects the syncrony of thalamic neurons and temporarily shifts them toward a burst firing mode. This effect can result either directly, through the Žmidline. thalamic nuclei, or indirectly, via the specific hypothalamic regions responsible for regulation of body temperature. However, at the moment, penicillin-induced hypothermia is the issue. How does this effect occur? According to several reports, this drug behaves as an antagonist of GABA A receptors making chloride ions unable to enter and pass through the channel w13,26,40x. Therefore, GABA A -mediated inhibition is decreased although not totally inactivated. From the point of epileptogenicity Žand also hypothermia., the question arises as to what happens with the intensity of the GABA B response after partial blockade of GABA A receptors? Misgeld et al. proposed a mutualistic relationship between two main types of GABAergic inhibition w27x. There are also findings that emphasize that inhibition mediated by GABAergic interneurons, and possibly nucleus reticularis thalami cells which reduce GABA A current, increases the amplitude of the GABA B current w14,22,39x. Therefore, penicillin could induce hypothermia and spike and waves indirectly, via the increased activity of GABA B receptors at critical thalamocortical synapses with well known consequences w14,38x. This is in agreement with contemporary studies of Horton et al. w21x and Yakimova et al. w43x who proposed the opposing interaction of GABA A and GABA B receptors in thermoregulatory responses. This could mean that GABA A and GABA B receptors regulate body temperature by adjusting each other and that reduction of one of them could result in the augmentation of the other w27x. In this way, partial blockade of GABA A inhibitory responses by penicillin induces compensatory intensifying of postsynaptic GABA B receptors activity marked by the fast decline of body temperature and subsequent SWDs. Other authors also documented an hypothermic effect after activation of GABA B receptors; for example, baclofen injected i.p. in doses of 5–10 mgrkg induces a clear hypothermic response w18,23,34x. Baclofen is regarded as responsible for in vivo epileptogenesis in animals w6x and humans w33x. This specific GABA B agonist also exacerbates the duration of SWDs in the GHB and PTZ models of absence epilepsy w38x. Also, another drug, GHB, known to induce absences after binding to GHB and GABA B receptors induces hypothermic response in monkeys w36x. Therefore, we can conclude that two main mechanisms of potentiation of GABA B receptor inhibition exist: Ž1. indirect by reducing the strength of GABA A inhibition Ž2. andror direct through specific activation of GABA B -dependent inhibition. However, in our study, hypothermia in both cases lasted beyond the seizures. SWDs disappeared approximately after 90 min Žlower penicillin dose. and 120 min Žhigher
93
penicillin dose. postinjection. These times are closely relationed with the nature of the temperature decline depicted in Fig. 1. Apparently, the tendency for the temperature to decline terminated after 60–90 min for the lower and after 90–120 min for the higher penicillin dose. During the plateau phase, body temperature remained at a relatively stable level. We hypothesize that at this time the physiological balance between GABA A and GABA B receptors is restored. In other words, GABA A receptors are released from the inhibition caused by penicillin and, as a result, absence seizures stop. Nevertheless, moderate hypothermia, although not sufficient to induce seizures, still exists and parallels the dynamics of complete penicillin elimination from the body. The results of this study clearly show that diazepam completely prevents the occurrence of penicillin-induced hypothermia and subsequent SWDs. This finding is in agreement with previous reports that penicillin binds to the benzodiazepine site at the GABA A ionophore w2,35x. Benzodiazepine-dependent mechanism of penicillin-induced hypothermia might be of great clinical importance because this could account for the mechanism of penicillin-induced non-specific encephalopathies in humans, such as confusion, memory impairment and agitation w31x. This study suggests the possibility that some thalamic nuclei known to regulate the degree of alertness at the same time represent the constituent moiety of the structures involved in regulation of body temperature. This hypothesis is in full agreement with recent findings w41x showing that certain thalamic nuclei receive and combine peripheral and central temperature inputs. In other words, because of this specific bifunctionality of certain nuclei inside the thalamus Žtemperature procession and regulation of the level of alertness., we might indirectly influence thalamocortical bursting generator through the shifting of peripheral andror central temperature information incoming to this structure. Therapeutical approaches against absence seizures in future would not be ultimately based upon pharmacological manipulation. Instead, target thalamic structures might be indirectly modified through the fine adjustment of body temperature. In this study we demonstrated that hypothermia in many cases paralleled absence seizures, so it would be of great interest to examine effects of raised body temperature on one particular model of absence seizure. Acknowledgements This study was supported by Ministry of Science and Technology of Serbia. References w1x B. Andersson, Cold defense reactions elicited by electrical stimulation within the septal area of the brain in goats, Acta Physiol. Scand. 41 Ž1957. 90–100.
94
Z.S. Ostojic´ et al.r Brain Research 777 (1997) 86–94
w2x A. Antoniadis, W.E. Muller, U. Wollert, Benzodiazepine receptor interactions may be involved in the neurotoxicity of various penicillin derivatives, Ann. Neurol. 8 Ž1980. 71–73. w3x D.B. Beleslin, D.R. Myers, Observations on the hypothermic effect of metrazol in the conscious cat, Pharmacol. Biochem. Behav. 1 Ž1973. 727–729. w4x C.P. Bailey, B.A. Cookson, D.F. Downing, Cardiac surgery under hypothermia, J. Thorac. Surg. 27 Ž1954. 73–95. w5x P.K. Banerjee, O.C. Snead, Thalamic mediodorsal and intralaminar nuclear lesions disrupt the generation of experimentally induced generalized absence-like seizures in rats, Epilep. Res. 17 Ž1994. 193–205. w6x N.G. Bowery, J.J. Maguire, G.D. Pratt, Aspects of the molecular pharmacology of GABA B receptors, Sem. Neurosci. 3 Ž1991. 241– 249. w7x J.A. Boulant, R.R. Gonzalez, The effect of skin temperature on the hypothalamic control of heat loss and heat production, Brain Res. 120 Ž1977. 367–372. w8x E. Bullit, Induction of c-Fos like protein within the lumbar spinal cord and thalamus of the rat following peripheral stimulation, Brain Res. 493 Ž1989. 391–397. w9x R. Busto, W.D. Dietrich, M.Y. Globus, Small differences in intraischemic brain temperature critically determine the extent of ischemic neuronal injury, J. Cereb. Blood Flow Metab. 7 Ž1987. 729–738. w10x D. Chyatte, J. Elefteriades, B. Kim, Profound hypothermia and circulatory arrest for aneurysm surgery, J. Neurosurg. 70 Ž1989. 489–491. w11x G.L. Clifton, Systemic hypothermia in treatment of severe brain injury: a review and update, J. Neurotrauma 12 Ž1995. 923–927. w12x V. Crunelli, S. Lightowler, C.E. Pollard, A T-type Ca2q current underlies low-threshold Ca2q potentials in cells of cat and rat lateral genicular nucleus, J. Physiol. 413 Ž1989. 543–561. w13x D.R. Curtis, C.J.A. Game, G.A.R. Johnston, R.M. McCulloch, R.M. MacLachlan, Convulsive action of penicillin, Brain Res. 43 Ž1972. 242–245. w14x A. Destexhe, D. Contreras, T.J. Sejnowski, M. Steriade, A model of spindle rhythmicity in the isolated thalamic reticular nucleus, J. Neurophysiol. 72 Ž1994. 803–818. w15x M.D. Ginsberg, L.L. Sternan, M.Y.-T. Globus, W.D. Dietrich, Therapeutic modulation of brain temperature: relevance to ischemic brain injury, Cerebr. Brain Metab. Rev. 4 Ž1992. 189–225. w16x P. Gloor ŽEd.., Electrophysiology of generalized epilepsy, Academic Press, London, 1984, pp. 107-115. w17x P. Gloor, R.G. Fariello, Generalized epilepsy: some of its cellular mechanisms differs from those of focal epilepsy, Trends Neurosci. 11 Ž1988. 63–68. w18x J.A. Gray, G.M. Goodwin, D.J. Heal, A.R. Green, Hypothermia induced by baclofen, a possible index of GABA B receptor function in mice, is enhanced by antidepressant drugs and ECS, Br. J. Pharmacol. 92 Ž1987. 863–870. w19x F. Hahn, Pharmakologie und toxikologie der schlafmittel, Dt. Med. J. 6 Ž1955. 293–300. w20x P.R. Hickey, N.P. Andersen, Deep hypothermic circulatory arrest: a review of pathophysiology and clinical experience as a basic for anesthetic management, J. Cardiol. Anesth. 1 Ž1987. 137–155. w21x R.W. Horton, R.A. LeFeuvre, N.J. Rothwell, M.J. Stock, Opposing effects of activation of central GABA-A and GABA-B receptors on brown fat thermogenesis in the rat, Neuropharmacology 27 Ž1988. 363–366. w22x J.R. Huguenard, D.A. Prince, Intrathalamic rhythmicity studied in vitro: nominal T-current modulation causes robust antioscillatory effects, J. Neurosci. 14 Ž1994. 5485–5502.
w23x H.C. Jackson, D.J. Nutt, Inhibition of baclofen-induced hypothermia in mice by the novel GABA B antagonist CGP 35348, Neuropharmacology 30 Ž1991. 535–538. w24x J. Jahnsen, R. Llinas, Ionic basis for the electroresponsiveness and oscillatory properties of guinea pig thalamic neurons in vitro, J. Physiol. 349 Ž1984. 227–247. w25x H.H. Jasper, J. Droogleever-Fortuyn, Experimental studies of the functional anatomy of petit mal epilepsy, Assoc. Res. Nerv. Ment. Dis. Proc. 26 Ž1947. 272–296. w26x R.L. Macdonald, J.L. Barker, Pentylentetrazol and penicillin are selective antagonists of GABA-mediated post-synaptic inhibition in cultured mammalian neurones, Nature 267 Ž1977. 720–721. w27x U. Misgeld, M. Bijak, W. Jarolimek, A physiological role for GABA B receptors and the effects of Baclofen in the mammalian central nervous system, Prog. Neurobiol. 46 Ž1995. 423–462. w28x Z. Ostojic, R. Veskov, Lj. Rakic, The appearance of spike and wave discharges in rat’s brain is temperature dependent, Behav. Pharmacol. 6 Ž1995. 97–98. w29x D.A. Prince, D. Farrel, ‘Centrencephalic’ spike and wave discharges following parenteral penicillin injection in the cat, Neurology 19 Ž1969. 309–310. w30x L.F. Quesney, P. Gloor, E. Kratzenberg, H. Zumstein, Pathophysiology of generalized penicillin epilepsy in the cat: the role of cortical and subcortical structures in systemic application of penicillin, Electroenceph. Clin. Neurol. 42 Ž1977. 640–655. w31x M.E. Raichle, H. Kutt, S. Louis, F. McDowell, Neurotoxicity of intravenously administered penicillin G, Arch. Neurol. 25 Ž1971. 232–239. w32x R. Ramanjaneyulu, M.K. Ticku, Interaction of pentylenetetrazol and tetrazol analogues with the picrotoxin site of the benzodiazepineGABA receptor-ionophore complex, Eur. J. Pharmacol. 98 Ž1984. 337–345. w33x J.M. Rush, F.B. Gibberd, Baclofen-induced epilepsy, J. Roy. Soc. Med. 83 Ž1990. 115–116. w34x J.S. Serrano, F.J. Minano, M. Sancibrian, J.A. Duran, Involvement of bicuculline-insensitive receptors in the hypothermic effect of GABA and its agonists, Gen. Pharmacol. 16 Ž1985. 505–508. w35x H. Shiraishi, M. Ito, T. Go, H. Mikawa, High doses of penicillin decreases w 3 Hxflunitrazepam binding sites in rat neuron primary culture, Brain Dev. 15 Ž1993. 356–361. w36x O.C. Snead, Gamma hydroxybutyrate in the monkey: I. Electroencephalographic, behavioral and pharmacokinetic studies, Neurology 28 Ž1978. 636–642. w37x O.C. Snead, Evidence for GABA B -mediated mechanisms in experimental absence seizures, Eur. J. Pharmacol. 213 Ž1992. 343–349. w38x O.C. Snead, Basic mechanisms of generalized absence seizures, Ann. Neurol. 37 Ž1995. 146–157. w39x M. Steriade, D. Contreras, Relations between cortical and thalamic cellular events during transition from sleep patterns to paroxysmal activity, J. Neurosci. 15 Ž1995. 623–642. w40x M.K. Ticku, Convulsant binding sites on the benzodiazepinerGABA receptor, in: R.W. Olsen, C.J. Venter ŽEds.., BenzodiazepinerGABA receptor and chloride channels, Alan R. Liss, New York, 1986, pp. 195-207. w41x A.K. Travis, H.J. Bockholt, M.A. Zardeto-Smith, K.A. Johnson, In vitro thermosensitivity of the midline thalamus, Brain Res. 686 Ž1995. 17–22. w42x D. Williams, A study of thalamic and cortical rhythms in petit mal, Brain 76 Ž1953. 50–69. w43x K. Yakimova, H. Sann, H.A. Schmid, F.-K. Pierau, Effects of GABA agonists and antagonists on temperature-sensitive neurones in the rat hypothalamus, J. Physiol. 494 Ž1996. 217–230.
Research report
The connection between absence-like seizures and hypothermia induced by penicillin: possible implication on other animal models of petit mal epilepsy Zoran S. Ostojic´ ) , Sabera Ruzdijic, ˇ ´ Marko Car, Ljubisav Rakic, ´ Rosica Veskov Institute for Biological Research, Department for Neurobiology and Immunology, UniÕersity of Belgrade, P.O. Box 11060, 29. NoÕembra 142, Belgrade, YugoslaÕia Accepted 29 July 1997
Abstract In this study we investigated the relationship between penicillin-induced hypothermia and petit mal epilepsy induced by this proconvulsant antibiotic. In order to find a possible dose-dependent relationship, we used two doses: 1500.000 and 1000.000 Urkg b.wt., both known as being sufficient to induce absence-like attacks with subsequent spike and wave discharges ŽSWD. in electrocorticogram ŽECoG.. Because of experimental data suggesting penicillin binding to benzodiazepine receptor recognition site, we also studied penicillin-induced changes in body temperature after diazepam pretreatment. Results of this study clearly show that penicillin in doses known to induce petit mal-like epilepsy concomitantly induces statistically significant dose-dependent decrease in body temperature. Pretreatment with diazepam completely prevents both penicillin-induced hypothermia and SWDs. On the other hand, both the diazepam and mixed diazepamq penicillin treatments did not significantly alter body temperature. These results suggest, however, that at least some of the penicillin effects described could be assigned to its binding to the benzodiazepine receptor recognition site at GABA A ionophore. This may have an important clinical implication because the inhibitory action of penicillin at the benzodiazepine receptor recognition site could account for the mechanism of penicillin-induced unspecific encephalopathies in humans. The relationship between petit mal epilepsy and hypothermia sheds new light on the action mechanisms of penicillin-induced absence seizures. q 1997 Elsevier Science B.V. Keywords: Penicillin; Diazepam; Absence seizure; Hypothermia; GABA; Thalamus
1. Introduction It has been generally accepted that brain cooling provides significant protection against ischemic brain damage during brain or heart surgery w4,11,20x. Profound hypothermia combined with total circulatory arrest has been in use for several years to correct cardiac defects in neonates and to facilitate repair due to extensive neurosurgical interventions Ža typical example is surgery for intracranial aneurysms.. Such delicate intervention would otherwise be impossible w10x. Recently, hypothermia has received increased interest because of strong evidence that mild reduction in body temperature could have a therapeutic role in reducing ischemic brain injury w9,15x. On the other hand, it is well known that many pharmacological agents induce changes in body temperature, )
Corresponding author. Fax: q38 Ž1. 11-761-433
0006-8993r97r$17.00 q 1997 Elsevier Science B.V. All rights reserved. PII S 0 0 0 6 - 8 9 9 3 Ž 9 7 . 0 1 0 0 7 - X
particularly in small laboratory animals but also in humans. Some drugs act directly on oxidative metabolic processes in brain tissue while other drugs affect the main thermoregulatory organs; thereafter, metabolic changes observed in the brain will be only an indirect consequence of these changes. Central structures of the thermoregulatory system are the preoptic Žanterior. and posterior regions of the hypothalamus, as well as the septum w1,7x. These hypothalamic neurons are responsible for both physiological and behavioral responses to temperature changes w7x. However, it was recently discovered that structures in the midline thalamus may play a role similar to that of specific hypothalamic regions in the processing of external and internal temperature information w8,41x. The role played by midline and other unspecific thalamic nuclei in the pathophysiology of absence seizures has been well documented w5,25,42x. At the same time, these and other thalamic nuclei have the unique feature that they can switch
Z.S. Ostojic´ et al.r Brain Research 777 (1997) 86–94
between two distinct modes of action in how they regulate the degree of alertness. Unspecific thalamic nuclei are responsible for generating both tonic burst firing in an awake state and phasic oscillatory activity during slowwave sleep w22x. Under pathological circumstances, spindling activity, as an electroencephalographical ŽEEG. marker of slow-wave sleep, may be transformed into spike and wave discharges ŽSWDs.. These rhythms are related in basic molecular, anatomical, and pharmacological profiles w12,16,17,24,38x. The intrinsic capacity of certain thalamic neurons, together with their possible participation in thermoregulatory processes, confirms the necessity for monitoring body temperature during the investigation of petit mal seizures in humans and experimental animals. In the last two decades numerous experimental models of petit mal-like epilepsy have been developed to serve as electrophysiological and behavioral archetype of absence seizures in humans. It has also been noted that some widely used drugs known to induce or exacerbate absence attacks Žpentylenetetrazole ŽPTZ., g-hydroxybutyrate ŽGHB., baclofen. at the same time decrease body temperature. Considerable evidence exists for the hypothermic effect of PTZ in dogs, rabbits, rats, mice w19x, and cats w3x, although it should be mentioned that the dosages employed were sufficient to induce other types of epilepsy. In 1978 it was demonstrated that another inducer ŽGHB. of petit mal-like epilepsy in experimental animals also exerts an influence upon the thermoregulatory system of monkeys; all monkeys that received GHB became profoundly hypothermic concomitantly with the EEG and behavioral changes w36x. Also, the selective GABA B agonist baclofen, when applied intraperitoneally Ži.p.. in 5–10 mgrkg doses, induced hypothermia in mice w18,23x. The next drug known to evoke absence seizures is the broad-spectrum antibiotic penicillin. High dose of parenterally applied penicillin in cats provides one of the best-suited animal models of human absence attacks w29x, and this is well known as the feline generalized penicillin epilepsy model. Later it was discovered that this antibiotic also induces absence attacks in rodents w30x. Penicillin acts as a non-competitive antagonist of GABA A receptors w13,26x, binding to the picrotoxin site at the GABA A receptor complex w26,32,40x. Other data also show penicillin binding to the benzodiazepine receptor recognition site at GABA A ionophore w2,35x. Thus far no report about the influence of convulsive doses of this agent on body temperature has been published. Concerning the relatively few and contradictory observations about the role of GABAergic receptors in thermoregulation, the primary goals of this study were to investigate the influence of high epileptogenic penicillin doses on body temperature as well as the possible parallel between petit mal-like epilepsy and decreased body temperature. We also examined penicillin-induced changes in body temperature after pretreatment with the GABA A potentiating anti-epileptic agent, diazepam, with respect to
87
experimental data that suggest penicillin binding to the benzodiazepine receptor recognition site.
2. Materials and methods 2.1. Animals The experiments were performed in male adult Ž3 months. Wistar rats with no previous drug history. Animals were freely moving and individually placed in a plexiglass recording chamber in an air-conditioned room under a 12-h lightrdark cycle, with water and standard laboratory food freely available. Room temperature was maintained at 22–248C. 2.2. Groups All rats were divided into five groups according to the drug injected. In the first Ž n s 25. and second groups Ž n s 15., dissolved crystalline penicillin was injected i.p. in doses of 1500.000 Urkg and 1000.000 Urkg b.wt., respectively. The third group Ž n s 10. was treated with a combination of 2 mgrkg diazepam and 1500.000 Urkg penicillin. Rats in the fourth group Ž n s 6. received 2 mgrkg diazepam i.p. In addition, in the fifth control Žplacebo. group Ž n s 10., rats were injected with an appropriate volume of saline solution. All statistical data were analyzed in respect to the placebo group. Five animals from each group treated with higher and lower dose of penicillin, as well as diazepam q penicillin Žfirst, second and third groups, amounting to a total of 15 animals. were surgically prepared with chronic cortical electrodes in order to detect the appearance of SWDs after administration of this drug and to reveal the suppressant effect of diazepam on the genesis and expression of penicillin-induced SWDs. 2.3. Surgical procedures Permanent epidural cortical electrodes were implanted under pentobarbital anesthesia Ž40 mgrkg, i.p.. to permit recording of the electrocorticogram ŽECoG. in freely moving animals. Two electrodes Žfine tips of stainless steel needle. were placed bilaterally over the frontal ŽA s 1.0 mm, L s 2.5–3 mm. cortex, keeping intact both the dura and cortex. Thus we avoided the appearance of latent epileptogenic foci which might be activated by high penicillin doses leading to development of focal or clonic epilepsy. An additional stainless steel screw electrode was driven into the skull above the cerebellum and served as a reference electrode. The animals were allowed to rest 1 week before control ECoG recordings were taken. All experiments were begun 1 day after the control ECoG
88
Z.S. Ostojic´ et al.r Brain Research 777 (1997) 86–94
recording. Only animals that did not show spontaneous SWDs were used. 2.4. ECoG recordings All recordings were made with the animals freely moving in shielded plexiglass recording chambers. ECoG signals were digitized with a sampling rate of 128 Hz and transmitted by shielded cable. The rats were allowed to habituate in a new environment for 30 min, and then the reference ECoG was recorded for the next 30 min prior to any drug administration. After drug injections, the ECoG was recorded throughout the experiment. For the combined injections Ždiazepam q penicillin., diazepam was injected and the ECoG was subsequently recorded for 5 min; thereafter, penicillin was given and the ECoG was continuously recorded for the next 150 min. 2.5. Temperature measurement During the experiments rats were kept in their plexiglass recording chambers in order to reduce any possible stress factors as well as baseline motor activity. All data sampling was also taken at the same time Ž10:00 h. to avoid sleep propensity. Core body temperature was measured rectally using a commercially available digital thermometer with an accuracy of "0.18C. Before measurement, the thermometer was placed in warm water Ž308C.; this simple procedure significantly decreased measuring time because the rats were kept restrained only for 8–10 s. Therefore, we prevented a rise in body temperature due to massive motor movement. Temperature alterations are shown as changes in body temperature from the basal level at t s 0. Basal levels were considered as a mean temperature computed for all animals included in the specified
group. Body temperature was continuously monitored for 5 h every 30 min after application of saline and both penicillin dose, and for 3 h every 30 min after application of diazepam and diazepam q penicillin. Exception was made during the first 30 min after application of both penicillin doses Žmeasurement of body temperature changes was performed at 15 min intervals. in order to detect the initial decrease in body temperature. Upon completion of the experiments, we performed occasional monitoring of body temperature during the next 48 h to find out the time interval needed for complete withdrawal of penicillin-induced hypothermia. 2.6. Data analysis Temperatures were presented as mean " S.E. and were analyzed using a 1-way analysis of variance ŽANOVA. for repeated measures. Sources of variation were divided into between-subjects factor ŽDrug., and within-subject effect. The latter includes the within-subject main effect ŽTime. and its interaction with the between-subjects factor ŽDrug = Time.. P - 0.05 was considered to indicate significant differences in changes in body temperature. For that purpose, we used SAS software package, general linear model procedure Žver. 6.1. with repeated statement. GreenhouseGeisser corrections were used for all repeated-measures analyses. Data were also baseline corrected at t s 0 Ž5 min before injections.. Statistical significance was tested between the placebo group and the group receiving specified drugs, or between two groups treated differentially. When significant Drug = Time interactions occurred, Tukey’s studentized range ŽHSD. tests were performed to determine the specific time points at which the drug-induced temperature response differed from that of the placebo or another groups.
Fig. 1. Changes in body temperature shown as mean " S.E. after application of two penicillin doses. Statistical differences were checked by Tukey’s studentized ŽHSD. range test between the corresponding time intervals of the penicillin Ž10 6 and 1.5 = 10 6 Urkg, respectively. and placebo sessions Ž a s 0.05, ) P - 0.05., and between that of 10 6 and 1.5 = 10 6 Urkg penicillin sessions Ž a s 0.05, q P - 0.05..
Z.S. Ostojic´ et al.r Brain Research 777 (1997) 86–94
3. Results The results of this study indicate that both doses of parenterally applied penicillin induce a dose-dependent decrease in core temperature ŽFig. 1.. Compared to placebo, application of 1500.000 Urkg penicillin induces highly significant alteration in body temperature ŽDrug = Time interaction: F10,380 s 6.76, P - 0.0001.. In addition, significant Time effect Ž F10,380 s 8.95, P - 0.0001. indicates temperature change over time. Higher penicillin dose induces an initial drop in body temperature of 0.88C Žfrom 37.43 " 0.068C to 36.63 " 0.118C; see Fig. 1. with a latency period of 15 min. Maximal level of hypothermia of y1.778C Žbody temperature: 35.66 " 0.288C. was reached 90 min after injection. In the next 90 min, a hypothermic plateau was formed at y1.748C and body temperature Ž35.69 " 0.318C. remained at this level. Since we found significant Drug = Time interaction, post hoc Tukey’s test revealed that both the initial and maximal decrease in body temperature were statistically significant, as were the all time points from 15 to 300 min ŽFig. 1.. The decrease in body temperature preceded and lasted beyond the ECoG markers of absence seizures manifested
89
by SWDs. The first epileptiform activity appeared within 25–35 min after drug application. This activity was characterized by bilaterally synchronous spike and wave complexes or single sharp waves ŽFig. 2.. The frequency of the SWDs ranged from 5 to 5.5 Hz and the duration of individual bursts varied from 2 to 7 s. Bursts of SWDs persisted almost 120 min after drug injection. On the contrary, complete restoration of body temperature in the animals having received penicillin injection with this dose was detected between 6 and 8 h. In six animals Ž24%., however, a long-lasting Ž) 24 h. and severe hypothermia Žbody temperature: 32.5–34.58C. developed. Fifteen minutes after injection of 1000.000 Urkg penicillin, mean normal body temperature Ž37.59 " 0.078C. decreased for 0.598C ŽFig. 1.. Thereafter, body temperature continued to fall and 45 min later Ž60 min postinjection period. further decreased for 0.308C Žbody temperature: 36.7 " 0.188C.. In the next 120 min, the hypothermic plateau has being formed with no or minimal change in body temperature ŽFig. 1.. Significant Drug = Time interaction also occurred Ž F10,280 s 3.72, P - 0.0001. as well as change in body temperature over Time Ž F10,280 s 5.12, P - 0.0001.. Tukey’s test revealed the presence of a statis-
Fig. 2. Generalized epileptiform activity elicited by i.p. injection of penicillin Ž1.5 = 10 6 Urkg b.wt.. manifested in typical spike and wave pattern. A: control recording Žsaline solution applied.; B: 70 min after application of penicillin. Horizontal line: 1 s, vertical line: 100 mV. Abbreviations in this and subsequent figures: LPCx, left parietal cortex; RPCx, right parietal cortex.
90
Z.S. Ostojic´ et al.r Brain Research 777 (1997) 86–94
Fig. 3. Generalized epileptiform activity elicited by i.p. injection of penicillin Ž10 6 Urkg b.wt.. manifested in typical spike and wave pattern. A: control ECoG recording Žsaline solution applied.; B: 40 min after application of penicillin. Horizontal line: 1 s, vertical line: 100 mV.
tically significant fall in body temperature at time points between 15 and 150 min. At 180 min and thereafter, however, there was no significant difference comparing
these time points with the placebo group, suggesting normalization of body temperature ŽFig. 1.. During the first 25–35 min, bilaterally synchronous SWDs with typical
Fig. 4. Changes in core temperature of each 30 min during the 180 min observation period in the control Žplacebo. and diazepam-treated conditions. There is no significant differences between the groups.
Z.S. Ostojic´ et al.r Brain Research 777 (1997) 86–94
91
Fig. 5. Changes in body temperature during the 180 min observations period between the corresponding time intervals of penicillin Ž10 6 and 1.5 = 10 6 Urkg. and paired application of diazepam Ž2 mgrkg. and penicillin Ž1.5 = 10 6 Urkg. 5 min afterward. Note significant differences between the mixed treatment and both penicillin doses. Tukey’s range test: a s 0.05, ) P - 0.05.
Fig. 6. ECoG changes observed after paired application of diazepam and penicillin 5 min afterward. A: control recording Žsaline solution applied.. B: 45 min after penicillin injection Ž50 min after diazepam pretreatment.. Note synchronized activity due to diazepam treatment and lack of penicillin-induced SWDs. Horizontal line: 1 s, vertical line: 100 mV.
Z.S. Ostojic´ et al.r Brain Research 777 (1997) 86–94
92
Fig. 7. Alteration in core temperature of each 30 min during the 180 min observation period induced by mixed treatment of diazepam Ž2 mgrkg. and penicillin.
morphology appeared in the ECoG ŽFig. 3.. The intraburst frequency, morphology and duration of individual bursts did not significantly differ from that induced by higher dose of penicillin. There was a difference, however, in the duration of discharges: in this group SWDs ceased almost 30 min earlier Žf 90 min after injection.. Time spent in hypothermia was also shortened and complete restoration of body temperature occurred 4–6 h after application of penicillin, although statistical testing revealed normalization of body temperature between 150 and 180 min after injection and subsequent lack of significant difference in relevance to the placebo treatment ŽFig. 1.. Statistical data also show that penicillin induces hypothermia in a dose-dependent manner, since we found a highly significant Drug = Time interaction Ž F10,380 s 3.88, P - 0.0001. and temperature alteration over Time Ž F10,380 s 14.07, P - 0.0001.. According to Tukey’s range test, changes in core rectal temperature induced by two penicillin doses appeared to be statistically significant at all time points between 30 and 300 min ŽFig. 1.. Table 1 Repeated-measures ANOVA about changes in core temperature Group Phigh vs. placebo Plow vs. placebo Phigh vs. Plow DPhigh vs. Phigh DPhigh vs. Plow DPhigh vs. placebo D vs. placebo
Time ŽF.
Ž P.
Time=Drug interaction ŽF. Ž P.
F10,380 s8.95 F10,280 s 5.12 F10,380 s14.07 F6,198 s8.97 F6,138 s 7.28 F6,138 s 0.97 F6,114 s 0.53
- 0.0001 - 0.0001 - 0.0001 - 0.0001 - 0.0001 ns ns
F10,380 s6.76 F10,280 s 3.72 F10,380 s 3.88 F6,198 s6.67 F6,138 s 3.18 F6,138 s1.64 F6,114 s 0.72
- 0.0001 - 0.0001 - 0.0001 - 0.0001 - 0.0059 ns ns
Abbreviations: Phigh and Plow , groups of animals that received 1.5=10 6 and 10 6 Urkg of penicillin, respectively; D, group of animals that received 2 mgrkg of diazepam; DPhigh , animals that received 1.5=10 6 Urkg of penicillin and pretreated with diazepam; ns, non-significant.
Finally, diazepam completely prevented the penicillininduced reduction of body temperature. Diazepam itself does not exert a statistically significant influence upon body temperature ŽFig. 4.. Accordingly, highly significant difference due to Drug treatment, Time and Drug = Time interaction exists for the first, second and fourth groups Žsee Table 1.. Also, Tukey’s test shows statistically significant differences at all measured time points Ž30–180 min; Fig. 5.. However, in this case, penicillin failed to induce SWDs; bilaterally synchronized rhythmic activity with a predominant frequency within the alpha band arises in the ECoG ŽFig. 6.. Moreover, according to Tukey’s test, mixed treatment with diazepam q penicillin does not significantly differ from that of placebo ŽFig. 7..
4. Discussion This study attempted to correlate hypothermia and absence epilepsy. It was shown that hypothermia clearly precedes ECoG hypersynchrony by 10–20 min. This time is probably somewhat higher because continuous monitoring of body temperature was not made between t s 0 and t s 15 min postinjection. Therefore, a critical level of hypothermia is required for development of SWDs which is reached approximately 30 min after application of penicillin. As our previous study showed, both mild Ž308C. and deep Ž14–178C. artificially induced hypothermia also evoked petit mal-like epilepsy with accompanying SWDs within the temperature range of 25–368C w28x. On the other hand, because of evidence suggesting the role of GABA B receptors in absence seizures w14,22,37,40x, as well as in hypothermia w18,23,34x, we suggest that a drop in body temperature Žfollowed by SWDs. could be consid-
Z.S. Ostojic´ et al.r Brain Research 777 (1997) 86–94
ered as a marker of GABA B receptor activation intensity. It is quite reasonable to conclude that mild-to-deep hypothermia, regardless of the means of induction, affects the syncrony of thalamic neurons and temporarily shifts them toward a burst firing mode. This effect can result either directly, through the Žmidline. thalamic nuclei, or indirectly, via the specific hypothalamic regions responsible for regulation of body temperature. However, at the moment, penicillin-induced hypothermia is the issue. How does this effect occur? According to several reports, this drug behaves as an antagonist of GABA A receptors making chloride ions unable to enter and pass through the channel w13,26,40x. Therefore, GABA A -mediated inhibition is decreased although not totally inactivated. From the point of epileptogenicity Žand also hypothermia., the question arises as to what happens with the intensity of the GABA B response after partial blockade of GABA A receptors? Misgeld et al. proposed a mutualistic relationship between two main types of GABAergic inhibition w27x. There are also findings that emphasize that inhibition mediated by GABAergic interneurons, and possibly nucleus reticularis thalami cells which reduce GABA A current, increases the amplitude of the GABA B current w14,22,39x. Therefore, penicillin could induce hypothermia and spike and waves indirectly, via the increased activity of GABA B receptors at critical thalamocortical synapses with well known consequences w14,38x. This is in agreement with contemporary studies of Horton et al. w21x and Yakimova et al. w43x who proposed the opposing interaction of GABA A and GABA B receptors in thermoregulatory responses. This could mean that GABA A and GABA B receptors regulate body temperature by adjusting each other and that reduction of one of them could result in the augmentation of the other w27x. In this way, partial blockade of GABA A inhibitory responses by penicillin induces compensatory intensifying of postsynaptic GABA B receptors activity marked by the fast decline of body temperature and subsequent SWDs. Other authors also documented an hypothermic effect after activation of GABA B receptors; for example, baclofen injected i.p. in doses of 5–10 mgrkg induces a clear hypothermic response w18,23,34x. Baclofen is regarded as responsible for in vivo epileptogenesis in animals w6x and humans w33x. This specific GABA B agonist also exacerbates the duration of SWDs in the GHB and PTZ models of absence epilepsy w38x. Also, another drug, GHB, known to induce absences after binding to GHB and GABA B receptors induces hypothermic response in monkeys w36x. Therefore, we can conclude that two main mechanisms of potentiation of GABA B receptor inhibition exist: Ž1. indirect by reducing the strength of GABA A inhibition Ž2. andror direct through specific activation of GABA B -dependent inhibition. However, in our study, hypothermia in both cases lasted beyond the seizures. SWDs disappeared approximately after 90 min Žlower penicillin dose. and 120 min Žhigher
93
penicillin dose. postinjection. These times are closely relationed with the nature of the temperature decline depicted in Fig. 1. Apparently, the tendency for the temperature to decline terminated after 60–90 min for the lower and after 90–120 min for the higher penicillin dose. During the plateau phase, body temperature remained at a relatively stable level. We hypothesize that at this time the physiological balance between GABA A and GABA B receptors is restored. In other words, GABA A receptors are released from the inhibition caused by penicillin and, as a result, absence seizures stop. Nevertheless, moderate hypothermia, although not sufficient to induce seizures, still exists and parallels the dynamics of complete penicillin elimination from the body. The results of this study clearly show that diazepam completely prevents the occurrence of penicillin-induced hypothermia and subsequent SWDs. This finding is in agreement with previous reports that penicillin binds to the benzodiazepine site at the GABA A ionophore w2,35x. Benzodiazepine-dependent mechanism of penicillin-induced hypothermia might be of great clinical importance because this could account for the mechanism of penicillin-induced non-specific encephalopathies in humans, such as confusion, memory impairment and agitation w31x. This study suggests the possibility that some thalamic nuclei known to regulate the degree of alertness at the same time represent the constituent moiety of the structures involved in regulation of body temperature. This hypothesis is in full agreement with recent findings w41x showing that certain thalamic nuclei receive and combine peripheral and central temperature inputs. In other words, because of this specific bifunctionality of certain nuclei inside the thalamus Žtemperature procession and regulation of the level of alertness., we might indirectly influence thalamocortical bursting generator through the shifting of peripheral andror central temperature information incoming to this structure. Therapeutical approaches against absence seizures in future would not be ultimately based upon pharmacological manipulation. Instead, target thalamic structures might be indirectly modified through the fine adjustment of body temperature. In this study we demonstrated that hypothermia in many cases paralleled absence seizures, so it would be of great interest to examine effects of raised body temperature on one particular model of absence seizure. Acknowledgements This study was supported by Ministry of Science and Technology of Serbia. References w1x B. Andersson, Cold defense reactions elicited by electrical stimulation within the septal area of the brain in goats, Acta Physiol. Scand. 41 Ž1957. 90–100.
94
Z.S. Ostojic´ et al.r Brain Research 777 (1997) 86–94
w2x A. Antoniadis, W.E. Muller, U. Wollert, Benzodiazepine receptor interactions may be involved in the neurotoxicity of various penicillin derivatives, Ann. Neurol. 8 Ž1980. 71–73. w3x D.B. Beleslin, D.R. Myers, Observations on the hypothermic effect of metrazol in the conscious cat, Pharmacol. Biochem. Behav. 1 Ž1973. 727–729. w4x C.P. Bailey, B.A. Cookson, D.F. Downing, Cardiac surgery under hypothermia, J. Thorac. Surg. 27 Ž1954. 73–95. w5x P.K. Banerjee, O.C. Snead, Thalamic mediodorsal and intralaminar nuclear lesions disrupt the generation of experimentally induced generalized absence-like seizures in rats, Epilep. Res. 17 Ž1994. 193–205. w6x N.G. Bowery, J.J. Maguire, G.D. Pratt, Aspects of the molecular pharmacology of GABA B receptors, Sem. Neurosci. 3 Ž1991. 241– 249. w7x J.A. Boulant, R.R. Gonzalez, The effect of skin temperature on the hypothalamic control of heat loss and heat production, Brain Res. 120 Ž1977. 367–372. w8x E. Bullit, Induction of c-Fos like protein within the lumbar spinal cord and thalamus of the rat following peripheral stimulation, Brain Res. 493 Ž1989. 391–397. w9x R. Busto, W.D. Dietrich, M.Y. Globus, Small differences in intraischemic brain temperature critically determine the extent of ischemic neuronal injury, J. Cereb. Blood Flow Metab. 7 Ž1987. 729–738. w10x D. Chyatte, J. Elefteriades, B. Kim, Profound hypothermia and circulatory arrest for aneurysm surgery, J. Neurosurg. 70 Ž1989. 489–491. w11x G.L. Clifton, Systemic hypothermia in treatment of severe brain injury: a review and update, J. Neurotrauma 12 Ž1995. 923–927. w12x V. Crunelli, S. Lightowler, C.E. Pollard, A T-type Ca2q current underlies low-threshold Ca2q potentials in cells of cat and rat lateral genicular nucleus, J. Physiol. 413 Ž1989. 543–561. w13x D.R. Curtis, C.J.A. Game, G.A.R. Johnston, R.M. McCulloch, R.M. MacLachlan, Convulsive action of penicillin, Brain Res. 43 Ž1972. 242–245. w14x A. Destexhe, D. Contreras, T.J. Sejnowski, M. Steriade, A model of spindle rhythmicity in the isolated thalamic reticular nucleus, J. Neurophysiol. 72 Ž1994. 803–818. w15x M.D. Ginsberg, L.L. Sternan, M.Y.-T. Globus, W.D. Dietrich, Therapeutic modulation of brain temperature: relevance to ischemic brain injury, Cerebr. Brain Metab. Rev. 4 Ž1992. 189–225. w16x P. Gloor ŽEd.., Electrophysiology of generalized epilepsy, Academic Press, London, 1984, pp. 107-115. w17x P. Gloor, R.G. Fariello, Generalized epilepsy: some of its cellular mechanisms differs from those of focal epilepsy, Trends Neurosci. 11 Ž1988. 63–68. w18x J.A. Gray, G.M. Goodwin, D.J. Heal, A.R. Green, Hypothermia induced by baclofen, a possible index of GABA B receptor function in mice, is enhanced by antidepressant drugs and ECS, Br. J. Pharmacol. 92 Ž1987. 863–870. w19x F. Hahn, Pharmakologie und toxikologie der schlafmittel, Dt. Med. J. 6 Ž1955. 293–300. w20x P.R. Hickey, N.P. Andersen, Deep hypothermic circulatory arrest: a review of pathophysiology and clinical experience as a basic for anesthetic management, J. Cardiol. Anesth. 1 Ž1987. 137–155. w21x R.W. Horton, R.A. LeFeuvre, N.J. Rothwell, M.J. Stock, Opposing effects of activation of central GABA-A and GABA-B receptors on brown fat thermogenesis in the rat, Neuropharmacology 27 Ž1988. 363–366. w22x J.R. Huguenard, D.A. Prince, Intrathalamic rhythmicity studied in vitro: nominal T-current modulation causes robust antioscillatory effects, J. Neurosci. 14 Ž1994. 5485–5502.
w23x H.C. Jackson, D.J. Nutt, Inhibition of baclofen-induced hypothermia in mice by the novel GABA B antagonist CGP 35348, Neuropharmacology 30 Ž1991. 535–538. w24x J. Jahnsen, R. Llinas, Ionic basis for the electroresponsiveness and oscillatory properties of guinea pig thalamic neurons in vitro, J. Physiol. 349 Ž1984. 227–247. w25x H.H. Jasper, J. Droogleever-Fortuyn, Experimental studies of the functional anatomy of petit mal epilepsy, Assoc. Res. Nerv. Ment. Dis. Proc. 26 Ž1947. 272–296. w26x R.L. Macdonald, J.L. Barker, Pentylentetrazol and penicillin are selective antagonists of GABA-mediated post-synaptic inhibition in cultured mammalian neurones, Nature 267 Ž1977. 720–721. w27x U. Misgeld, M. Bijak, W. Jarolimek, A physiological role for GABA B receptors and the effects of Baclofen in the mammalian central nervous system, Prog. Neurobiol. 46 Ž1995. 423–462. w28x Z. Ostojic, R. Veskov, Lj. Rakic, The appearance of spike and wave discharges in rat’s brain is temperature dependent, Behav. Pharmacol. 6 Ž1995. 97–98. w29x D.A. Prince, D. Farrel, ‘Centrencephalic’ spike and wave discharges following parenteral penicillin injection in the cat, Neurology 19 Ž1969. 309–310. w30x L.F. Quesney, P. Gloor, E. Kratzenberg, H. Zumstein, Pathophysiology of generalized penicillin epilepsy in the cat: the role of cortical and subcortical structures in systemic application of penicillin, Electroenceph. Clin. Neurol. 42 Ž1977. 640–655. w31x M.E. Raichle, H. Kutt, S. Louis, F. McDowell, Neurotoxicity of intravenously administered penicillin G, Arch. Neurol. 25 Ž1971. 232–239. w32x R. Ramanjaneyulu, M.K. Ticku, Interaction of pentylenetetrazol and tetrazol analogues with the picrotoxin site of the benzodiazepineGABA receptor-ionophore complex, Eur. J. Pharmacol. 98 Ž1984. 337–345. w33x J.M. Rush, F.B. Gibberd, Baclofen-induced epilepsy, J. Roy. Soc. Med. 83 Ž1990. 115–116. w34x J.S. Serrano, F.J. Minano, M. Sancibrian, J.A. Duran, Involvement of bicuculline-insensitive receptors in the hypothermic effect of GABA and its agonists, Gen. Pharmacol. 16 Ž1985. 505–508. w35x H. Shiraishi, M. Ito, T. Go, H. Mikawa, High doses of penicillin decreases w 3 Hxflunitrazepam binding sites in rat neuron primary culture, Brain Dev. 15 Ž1993. 356–361. w36x O.C. Snead, Gamma hydroxybutyrate in the monkey: I. Electroencephalographic, behavioral and pharmacokinetic studies, Neurology 28 Ž1978. 636–642. w37x O.C. Snead, Evidence for GABA B -mediated mechanisms in experimental absence seizures, Eur. J. Pharmacol. 213 Ž1992. 343–349. w38x O.C. Snead, Basic mechanisms of generalized absence seizures, Ann. Neurol. 37 Ž1995. 146–157. w39x M. Steriade, D. Contreras, Relations between cortical and thalamic cellular events during transition from sleep patterns to paroxysmal activity, J. Neurosci. 15 Ž1995. 623–642. w40x M.K. Ticku, Convulsant binding sites on the benzodiazepinerGABA receptor, in: R.W. Olsen, C.J. Venter ŽEds.., BenzodiazepinerGABA receptor and chloride channels, Alan R. Liss, New York, 1986, pp. 195-207. w41x A.K. Travis, H.J. Bockholt, M.A. Zardeto-Smith, K.A. Johnson, In vitro thermosensitivity of the midline thalamus, Brain Res. 686 Ž1995. 17–22. w42x D. Williams, A study of thalamic and cortical rhythms in petit mal, Brain 76 Ž1953. 50–69. w43x K. Yakimova, H. Sann, H.A. Schmid, F.-K. Pierau, Effects of GABA agonists and antagonists on temperature-sensitive neurones in the rat hypothalamus, J. Physiol. 494 Ž1996. 217–230.