Journal of Cerebral Blood Flow and Metabolism 12:962-970 © 1992 The International Society of Cerebral Blood Flow and Metabolism Published by Raven Press, Ltd., New York
Stimulation of the Fastigial Nucleus Enhances EEG Recovery and Reduces Tissue Damage After Focal Cerebral Ischemia Fangyi Zhang and Costantino Iadecola Department of Neurology, University of Minnesota Medical School, Minneapolis, Minnesota, U.S.A.
Summary: Stimulation of the cerebellar fastigial nucleus (FN) increases CBF but not metabolism and reduces the tissue damage resulting from focal cerebral ischemia. This effect may result from enhancing CBF in the isch emic tissue without increasing local metabolic demands. To test this hypothesis, we studied whether the reduction in tissue damage is restricted to the neocortex, a region in which the CBF increase is independent of metabolism, and whether stimulation of the dorsal medullary reticular formation (DMRF), a treatment that increases both cere bral metabolism and CBF, also protects the brain from ischemia. In halothane-anesthetized Sprague-Dawley rats, the middle cerebral artery (MCA) was occluded ei ther proximally or distally to the lenticulostriate branches. The FN or DMRF were then stimulated for 1 h (50--100 f.LA; 50 Hz; 1 s onl 1 s oft). Twenty-four hours later, the infarct volume was determined. FN stimulation substantially reduced the size of the infarct, an effect that was greater with distal (-69 ± 8%; n 6; p < 0 .001 ;
mean ± SD) than with proximal ( - 38 ± 8%; n 8; p < 0.001) MCA occlusion. The reduction occurred only in neocortex (-43 ± 9%; p < 0.00 1) and not in striatum (-16 ± 21%; p > 0.05). Stimulation of the FN also en
Studies over the past several decades have dem onstrated that there are central neural networks that have a profound influence on the circulation and metabolism of the brain (see Iadecola (19920) for review). One such pathway involves the area of the rostral cerebellar fastigial nucleus (FN). Electrical stimulation of this region increases CBF globally through neural projections entirely contained within the brain (Nakai et al., 1982; Iadecola et al., 1983c). The increases in CBF are cholinergically mediated
(Iadecola et al., 1986) and are greatest in cerebral cortex (240% of control), wherein they are not as sociated with increased cerebral glucose utilization (CGU) (Nakai et aI., 1983). Another pathway in volves a restricted region of the dorsal medullary reticular formation (DMRF). Excitation of this re gion is able to increase CBF globally (Iadecola et aI., 19830). However, at variance with the FN, the increases in CBF evoked from the DMRF are asso ciated with, and presumably secondary to, in creased CGU (secondary metabolic vasodilation) (Iadecola et al., 1983b). It has been demonstrated recently that stimula tion of the FN reduces the tissue damage resulting from occlusion of the middle cerebral artery (MCA) in rat (Reis et al., 1991). FN stimulation may "pro tect" the ischemic tissue from infarction by enhanc ing CBF in the ischemic territory without increasing
=
hanced recovery of EEG amplitude in the ischemic cortex (+48%; p < 0.003). DMRF stimulation (n 7) did not affect the stroke size or EEG recovery (p > 0.05). Thus, stimulation of the FN, but not the DMRF, attenuates the damage reSUlting from focal ischemia. The finding that the protection is limited to the neocortex and that the amount of tissue salvaged is greater after distal occlusions suggests that the FN, unlike the DMRF, may rescue tis sue from infarction by enhancing collateral flow without increasing local energy requirements. Activation of se lected neural networks may modulate the expression of the damage resulting from focal cerebral ischemia. Key Words: Cerebellum-Medullary reticular formation Middle cerebral artery-Cerebral blood flow-Rat. =
=
Received April 4, 1992; final revision received June 8, 1992; accepted June 29, 1992. Address correspondence and reprint requests to Dr. C. Iadecola at Department of Neurology, University of Minnesota Medical School, Box 295 UMHC, 420 Delaware Street S.E., Minneapolis, MN 55455, U.S.A. Abbreviations used: AP, arterial pressure; CGU, cerebral glu cose utilization; DMRF, dorsal medullary reticular formation; FN, fastigial nucleus; MCA, middle cerebral artery.
962
BRAINSTEM STIMULATION AND FOCAL ISCHEMIA
963
local metabolic demands. Based on this hypothesis, the reduction in tissue damage should be restricted to the neocortex, a region in which the CBF in crease is independent of metabolism (Nakai et aI., 1983). Furthermore, stimulation of the DMRF, a treatment that increases CBF and metabolism, should not exert any protective effect. In this study, therefore, we sought to investigate whether the re duction in tissue damage exerted by FN stimulation is restricted to the cortex and whether also stimu lation of the DMRF reduces the extent of ischemic damage resulting from MCA occlusion. We also sought to determine whether activation of these re gions improves the functional state of the tissue af ter ischemia, as indicated by the recovery of spon taneous electrical activity. We shall demonstrate that stimulation of the FN, but not the DMRF, en hances the recovery of the EEG and reduces, only in neocortex, the tissue damage resulting from focal ischemia. Portions of this study have been pub lished in abstract form (Zhang and Iadecola, 1992).
METHODS Methods for anesthesia, surgical preparation of ani mals, and brain stimulation have been described in detail in previous publications (Nakai et ai., 1982; Iadecola et aI., 1983a,b,c, 1986) and will only be summarized.
General preparation of animal Studies were conducted on 40 male Sprague-Dawley rats weighing 280-350 g. Animals were anesthetized with halothane in 100% oxygen (5% induction, 2-3% during surgery, 1% maintenance) administered through a facial mask. The tail artery was cannulated for continuous mon itoring of the arterial pressure (AP) and heart rate, and rats were placed in a stereotaxic frame (Kopt). Body tem perature was maintained at 37 ± 0.5°C by a thermostati cally controlled lamp connected to a rectal probe. Small volumes of arterial blood (50 J.ll) were sampled at various times for measurement of P a02' P aco2' and pH by a blood-gas analyzer (Corning, model 178).
Brain stimulation and recording of the electroencephalogram· As described in detail elsewhere (Nakai et aI., 1982; Iadecola et al. 1983a), the FN or the DMRF was stimu lated with cathodal square-wave pulses (0.5 ms) through monopolar electrodes (tip diameter: 150 J.lm). The anode was a clip attached to the neck muscles. In all animals, stimulated sites were histologically verified in Nissl stained sections (Fig. 1). The EEG was recorded monopolarly through a stain less steel screw inserted to lie on the dura at a site 2 mm lateral and 2 mm caudal to the bregma ipsilaterally to the occluded MCA. The reference electrode was a clip at tached to the scalp muscles. A 30 s EEG epoch was dig itized from the record by a flat-bed scanner (Hewlett Packard). The X, Y coordinates of the tracing were de-
FIG. 1. Location of stimulated sites (black dots) in the region of the fastigial nucleus (FN) (upper panel) or dorsal medul lary reticular formation (DMRF) (lower panel). Py, pyramidal tract; SQ, superior olive; STN, spinal trigeminal nucleus; STT, spinal trigeminal tract; ve, vestibular complex; VII, fa cial nerve nucleus.
termined using a computer program for vectorizing TIFF files (UnGraph, Biosoft). The mean amplitude of the dominant frequency was then determined using wave analysis software (Superscope, GW Instruments). To minimize the confounding effects on the EEG of changes in AP or of fluctuations in the level of anesthesia, the AP was tightly controlled (see below) and the concentration of halothane was maintained constant throughout the pe riod of ischemia and brain stimulation.
Occlusion of the middle cerebral artery The MCA was exposed using an approach modified from that of Tamura et al. (1981). Briefly, a skin incision was made between the orbit and external auditory me atus. The coronoid process of the mandible was removed to expose the inferotemporal fossa. For proximal occlu sions (Fig. 2, site A), a 3-4 mm hole was drilled at a site
J Cereb Blood Flow Metab, Vol. /2, No.6, 1992
F. ZHANG AND C. IADECOLA
964
Int. Cerebral Vein
image analysis system (MCID, MI, Imaging Research Inc.). Brain sections were digitized and the borders of the lesion were traced using a mouse. The volume of the infarct (mm3) was computed on line as the product of the cross-sectional area and distance between sections. To reconstruct the rostrocaudal extent of the lesion, the cross-sectional area of the stroke was plotted as a func tion of the distance from the interaural line (Mohamed et aI., 1985).
Experimental protocol
FIG. 2. Location of sites at which the middle cerebral artery (MeA) was occluded. For proximal occlusions (site A), the MeA was cauterized from the origin of the lenticulostriate branches to the inferior edge of the olfactory tract and sev ered proximally to the lenticulostriate branches. For distal occlusions (site B), the MeA was cauterized from the inferior cerebral vein to the superior edge of the olfactory tract and severed at the level of the inferior cerebral vein. superior and lateral to the foramen ovale to expose the proximal portion of the MCA. The MCA was elevated and cauterized just proximally to the origin of the lentic ulostriate branches. The occlusion was extended 2-3 mm distally and the MCA was transected. For distal occlu sions (Fig. 2, site B), the MCA was elevated and cauter ized just proximally to the inferior cerebral vein (Fig. 2, site B). The occlusion was extended up to the superior edge of the olfactory tract and the MCA was transected. In most rats, distal or proximal MCA occlusions resulted in sustained reductions in the amplitude of the EEG (6080%). These animals were found to have large infarcts involving mostly the cortex (distal occlusion) or cortex and striatum (proximal occlusion). Rats in which MCA occlusion did not substantially decrease the EEG ampli tude «40%) or in which the EEG recovered spontane ously suffered small cortical infarcts or infarcts restricted to the striatum (Salgado et aI., 1989). These animals were not included in the study.
Determination of infarct volumes Twenty-four hours after MCA occlusion, rats were deeply anesthetized with 5% halothane and mounted on a stereotaxic frame. The coronal plane corresponding to the interauralline was marked by inserting an electrode at the stereotaxic coordinates of the interaural line. This mark was then used as a reference to reconstruct the rostrocaudal extent of the lesion. Rats were then killed by decapitation and their brains were rapidly removed and frozen in isopentane at -30°C. Coronal sections (30 ""m) were cut in a cryostat (-20°C), collected at 300 ""m in tervals, and stained with thionin using conventionalmeth ods. The boundaries of the infarct were usually well de marcated on thionin-stained sections. When the borders of the lesion were not clear, they were verified micro scopically. The infarct volume was determined using an
J Cereb Blood Flow Me/ab, Vol. 12, No.6, 1992
Rats were surgically prepared and the MCA was ex posed as described above. In animals in which the FN was stimulated, an electrode was inserted into the cere bellum at a site 5 mm rostral to the obex and 0.8 mm lateral to the midline. The most active pressor site in the FN was then localized using procedures described in de tail elsewhere (Nakai et aI., 1982). In animals in which the DMRF was stimulated, the electrode was inserted through the cerebellum at a site 1 mm lateral to midline and 3 mm anterior to the obex (Iadecola et aI., 1983a,b). Once the most active site in the FN or DMRF was local ized, procedures for occlusion of the MCA were begun. The stereotaxic frame was rotated to place the animal in a lateral position and the MCA occluded as described above. Occurrence of neocortical ischemia was indicated by a 60-80% reduction in the amplitude of the EEG. Three to 5 min after observing the reduction in EEG am plitude, stimulation of the FN or DMRF was begun. As described in detail elsewhere (Nakai et aI., 1982; Iadecola et aI., 1983a,b), the FN or DMRF were stimulated with intermittent trains of stimuli (I s on/I s off; 50 Hz). During the first 2-4 min, the current intensity was gradually in creased up to five times the threshold current (i.e., the stimulus current producing a 10 mm Hg elevation in the AP, usually 10-20 ""A). At the same time, the evoked elevations in the AP were offset by slow removal of small amounts (1-2 ml) of arterial blood. By this procedure, the AP did not differ between stimulated and unstimulated rats (Table 1) and the maximal AP in stimulated animals was 115 ± 10 mm Hg. The FN or DMRF was stimulated for 1 h. At the end of the stimulation period, catheters were removed and wounds infiltrated with lidocaine and sutured. Animals were allowed to recover and carefully observed during the postoperative period for the occur rence of seizures. Twenty-four hours later, rats were killed for determination of the infarct size. Unstimulated rats and rats that underwent DMRF stimulation tended to have a Peo2 higher than that of awake, behaving rats (e.g., LeDoux et al., 1983) (Table 1). This degree of hy percapnia is commonly observed in spontaneously breathing halothane-anesthetized rats undergoing MCA occlusion (e.g., Brint et a!., 1988; Reis et a!., 1991; Maiese et a!., 1992). The lower Peo2 of rats in which the FN was stimulated (Table 1) can be attributed to the in crease in the ventilatory rate resulting from stimulation of the fastigial pressor area (Onai et a!., 1987). Six groups of rats were studied (Table I). The first group (n 6) consisted of rats in which the MCA was exposed and not occluded. The second group (n 7) consisted of rats in which the MCA was occluded but no brain stimulation was carried out. The third and fourth groups consisted of rats in which the MCA was occluded proximally and the FN (n 8) or the DMRF (n 7) was stimulated. The fifth group (n 6) consisted of unstim ulated rats in which the MCA was occluded distally. The =
=
=
=
=
965
BRAINSTEM STIMULATION AND FOCAL ISCHEMIA
TABLE 1. Body weight, arterial pressure, and blood gases of rats with and without stimulation of FN or DMRF after
MeA occlusion Proximal occlusion
Sham
MAP (mm Hg) Pe o2 (mm Hg) P02 (mm Hg) pH
n
FN stimulation
Unstimulated
Unstimulated
FN stimulation
322 ± 18
323 ± 22
322 ± 15
313 ± 24
317 ± 29
82 ± 9
90 ± 9
93 ± 5
87 ± 6
86 ± 7
95 ± 4
63 ± 3
56 ± 5
41 ± 6a
52 ± 5
56 ± 5
42 ± 7a
321 ± 31
BW
Distal occlusion DMRF stimulation
b
205 ± 53 7.34 ± 0.03 6
247 ± 74 7.49 ± 0.03 7
252 ± 71 7.48 ± 0.03 8
318 ± 50 7.38 ± 0.04a 7
236 ± 58 7.41 ± O.03a 6
250 ± 27 7.49 ± 0.04 6
Values are mean ± SD. BW: body weight; MAP: mean arterial pressure. MAP and blood gases values represent the average of three measurements taken at 5, 30, and 60 min after MCA occlusion. a p < 0.05 from sham and unstimulated groups (analysis of variance and Tukey's test). b p < 0.05 from stimulated groups.
sixth group (n 6) consisted of rats in which the MCA was occluded distally and the FN was stimulated. =
Data analysis Data are expressed as means ± SD. Comparisons among multiple groups were statistically evaluated by analysis of variance and Tukey's test (Systat, Inc.) (Ta bles 1 and 2 and Fig. 6). Two group comparisons were evaluated by the unpaired t test (Fig. 3). For all statistical procedures differences were considered significant for p < 0.05.
RESULTS
et aI., 1992). When the cross-sectional area of the stroke was plotted as a function of the distance from the interaural line (Fig. 4), the largest area of infarc tion occurred approximately 6 mm rostral to the interaural line. Distal occlusions of the MeA (Fig. 2, site B) (n 6) resulted in infarcts involving al most exclusively the cerebral cortex (Table 2). The size of the lesion and the rostrocaudal distribution of its cross-sectional area were comparable to those of neocortical infarcts resulting from proximal MeA occlusion (Fig. 4; Table 2). =
Effect of proximal or distal MeA occlusion on size
Effect of stimulation of the FN or DMRF on the
and regional distribution of the infarct
size of the infarct resulting from MeA occlusion
In rats in which the MeA was exposed and ele vated but not occluded (sham occlusion; n 6), a small infarct restricted to the cortex adjacent to the surgical site was observed (24 ± 14 mm3). Occlu sion of the MeA proximal to the origin of the len ticulostriate arteries (Fig. 2, site A; n 7) resulted in infarcts involving neocortex and striatum (Table 2, Fig. 3). The regional distribution of the infarct was comparable to that previously described in the rat (Fig. 3) (Tamura et aI., 1981; Park et aI., 1988; Ginsberg and Busto, 1989; Reis et aI., 1991; Maiese =
=
With FN stimulation (n 8), the size of the total infarct resulting from proximal MeA occlusion was reduced by 38 ± 8% (p < 0.001 from the unstimu lated group; analysis of variance and Tukey's test) (Table 2). In the neocortex, the volume of the in farct was reduced by 43 ± 9% (p < 0.001) while in the striatum the stroke size was not affected (p > 0.05). The regions of cerebral cortex spared from infarction included the dorsomedial portion of the frontal, parietal, and occipital cortex (Fig. 4). No effect was observed below the rhinal fissure. In rats =
TABLE 2. Effect of FN or DMRF stimulation on the volume of the infarct resulting from occlusion of the middle
cerebral artery Proximal occlusion
Total Cortex Striatum
n
Distal occlusion
Unstimulated
FN stimulation
DMRF stimulation
Unstimulated
FN stimulation
330 ± 41 250 ± 45 57 ± 8 7
205 ± na 144 ± 65a 49 ± 9 8
322 ± 52 241 ± 49 65 ± 12 7
223 ± 28 207 ± 25 10 ± 14 6
73 ± 18* 64 ± 14* 5 ± 6 6
Values are mean ± SD. a p < 0.001 (analysis of variance and Tukey's test).
J Cereb Blood Flow Metab, Vol. /2, No.6. 1992
966
F. ZHANG AND C. IADECOLA
NEOCORTEX 60 C
A N
E E -
Unstimulated FN Stimulation *
40
*
*
as
�
as
-
20
o a-
as
-
c:
O��WW����
15 12 9 6 3 o ·3 Distance from interaural line (mm)
STRIATUM 20 UNSTIMULATED
FN STIM
*
DMRFSTIM
FIG. 3. Distribution of the cross-sectional area of the infarct
resulting from proximal occlusion of the middle cerebral ar tery (MCA) with and without (left column) stimulation of the fastigial nucleus (FN) or dorsal medullary reticular formation (DMRF). Thionin-stained sections were projected on paper using a microprojector, and the area of necrosis was traced. After MCA occlusion, stimulation of the FN (middle column), but not the DMRF (right column), leads to smaller infarcts. The area rescued from infarction involves the dorsomedial aspect of the frontal, parietal, and occipital cortex. FN stim ulation does not affect stroke size in subcortical regions.
in which the MeA was occluded more distally (Fig. 2, site B), FN stimulation (n 6) reduced the total infarct size by 69 ± 8%, a reduction significantly larger than that observed after proximal occlusion (p < 0.001). In contrast to the FN, stimulation of the DMRF did not influence the volume of the in farct resulting from proximal MeA occlusion (Table 2, Fig. 4) (p > 0. 05). The rostrocaudal distribution of the cross-sectional area of the infarct in animals with or without FN stimulation is illustrated in Fig. 4. The greatest reductions in the total cross sectional area occurred between �O and 7 mm ros tral to the interaural line (Fig. 3) (p < 0.05). In the striatum, the area was slightly reduced only at the rostrocaudal level, where the stroke was largest (Fig. 3, p < 0. 05). The effect of FN stimulation on the cross-sectional area of infarcts resulting from distal MeA occlusion was similar to that observed after proximal occlusion. These observations indi cate that, in agreement with the study of Reis et al. =
J
Cereb Blood Flow Metab, Vol. 12, No.6, 1992
N
E E as Q)
r
15
10
a-
as
-
0 a-
as c:
-
5
o
I � I I ���__�������...�.�I-L�
-3
o
6
3
9
12
15
Distance from interaural line (mm) FIG. 4. Cross-sectional area of the infarct resulting from proximal MCA occlusion plotted as a function of the distance from the interaural line. Note that the area of stroke in the cortex (upper panel) is smaller in FN-stimulated animals, par ticularly in the caudal regions of the infarct. In striatum (lower panel), a small reduction was observed only at the site where the infarct was largest. Asterisks indicate p values of less than 0.01 (t test).
(1991), the largest amount of tissue rescued from infarction involves the dorsomedial and caudal por tions of the infarct. Effect of FN and DMRF stimulation on the reduction in EEG amplitude after MeA occlusion
In unstimulated rats (n 7), proximal MeA oc clusion reduced the EEG amplitude by 73 ± 8% (p =
BRAINSTEM STIMULATION AND FOCAL ISCHEMIA
967
AFTER OCCLUSION
BEFORE OCCLUSION
Smin
60 min
�
UNSTIM. FIG. 5. Effect of stimulation of the fastigial nucleus (FN) or dorsal medullary reticular formation (DMRF) on the reduction in EEG amplitude resulting from MeA occlusion. Five minutes after MeA occlusion, the EEG amplitude is reduced equally in all groups (middle panel). FN or DMRF stimulation was initiated at this time. Note that FN stimulation (right panel, middle trace) enhances the recovery of the EEG amplitude after 60 min of ischemia.
FN STIM.
DMRF STIM.
�
� 1
-- L..____
FN or DMRF stim
�0.5mv 2 sec < 0.001; Fig. 5). Sixty minutes later, the EEG am plitude recovered by 13 ± 12% (Figs. 5 and 6). With FN stimulation (n = 8), the initial EEG depression was not different from that of unstimulated controls (72 ± 5%; p > 0.05 from the unstimulated group; Fig. 5). However, 60 min after occlusion, the EEG recovery was significantly greater than that of the unstimulated group (Fig. 6; p < 0.003). Stimulation of the DMRF did not influence the changes in am plitude of the EEG resulting from MCA occlusion (Figs. 5 and 6; p > 0.05 from the unstimulated group). Thus, stimulation of the FN, but not the DMRF, enhances the recovery of the EEG after neocortical ischemia.
the functional impairment and tissue damage result ing from cerebral cortical ischemia. Therefore, the present study confirms and extends the results of Reis et al. (1991), who first described that FN stim ulation reduces the size of the infarct resulting from MCA occlusion. The effect of FN stimulation on the size of the infarct is unlikely to be a consequence of the lower peo2 of the rats in which the FN was stimulated since differences in peo2 comparable to those seen between stimulated and unstimulated rats do not appear to influence the tissue damage resulting from focal ischemia (Maiese et al., 1992), nor can the
100
DISCUSSION
This study sought to determine whether activa tion of the FN or the DMRF, regions associated with intracerebral pathways that have profound ef fects on CBF and/or CGU, could influence the func tional impairment and tissue damage resulting from focal cerebral ischemia. We found that electrical stimulation of the FN is able to revert the depres sion in EEG amplitude and to reduce the size of the hemispheric infarct resulting from proximal or dis tal MCA occlusions. Analysis of the regional distri bution of the effect revealed that the brain tissue rescued from infarction is restricted to the neocor tex, as in the striatum the size of the infarct is not affected. In contrast to the FN, stimulation of the DMRF did not influence the EEG changes and the size of the infarct resulting from MCA occlusion. These findings indicate that excitation of the fasti gial pressor area, but not the DMRF, ameliorates
�
Mean±SD !!lI Unstimulated III DMRF Stimulation • FN stimulation * p
-
�o
o Q) a: " w w
50 25 0 '--
__-"-'
FIG. 6. Effect of fastigial nucleus (FN) or dorsal medullary
reticular formation (DMRF) stimulation on the recovery of the EEG amplitude after MeA occlusion. EEG recovery is the difference between the EEG amplitude at 5 and 60 min, nor malized with respect to the preocclusion amplitude and ex pressed as a percentage of the EEG amplitude at 5 min. Stim ulation of the FN but not DMRF enhances the EEG recovery at 60 min after MeA occlusion.
J
Cereb Blood Flow Metab, Vol. /2, No.6, 1992
968
F. ZHANG AND C. IADECOLA
protective effect exerted by FN stimulation be a consequence of the controlled hemorrhage per formed to prevent hypertension. Removal of 1-2 ml of blood does not affect the resting CBF and pro duces only minimal hemodilution (ladecola et aI., 1983a). These changes cannot account for the sub stantial reduction in infarct size elicited by FN stim ulation. Furthermore, similar amounts of blood were removed during DMRF stimulation, a treat ment that does not affect the magnitude of ischemic damage. It is also unlikely that the effect of FN stimulation on stroke size is a nonspecific conse quence of brain stimulation as electrical stimulation of the DMRF (present study) or of the dentate nu cleus (Reis et aI., 1991) does not influence the size of the infarct. Therefore, the reduction in stroke size produced by FN stimulation appears to be a direct consequence of activation of neural pathways originating or passing through the region of the ros tral FN. The reduction in stroke size elicited by stimula tion of the FN is regionally selective as it occurs in the neocortex and not in the striatum. This finding is of interest in view of the fact that the mechanisms mediating the cerebrovascular actions of FN stim ulation in the neocortex and striatum are substan tially different. While in the neocortex the increases in CBF are independent of increased metabolic ac tivity, in the striatum the vasodilation is associated with, and presumably secondary to, increased me tabolism (Nakai et aI., 1983). Therefore, in the neo cortex, FN stimulation may increase substrate de livery without enhancing the energy requirements of the tissue. An excess of substrates may result in a greater resistance to energy failure during isch emia. In support of this hypothesis is our observa tion that stimulation of the DMRF, a treatment that increases CBF by increasing the metabolic de mands of the tissue (Iadecola et aI., 1983b), does not ameliorate ischemic damage. The regional selectivity of the effect of FN stim ulation could also result from differences in the pat tern of vascular supply in the neocortex and stria tum. The blood supply of the striatum originates largely from terminal arteries (Yamori et aI., 1976). Thus, the potential for development of a compen satory collateral circulation in the striatum is less than in the neocortex. The striatum is therefore more vulnerable to ischemic damage and could be less sensitive to the protective effects of FN stim ulation. These observations may indicate that FN stimulation improves the outcome of focal ischemia by enhancing the development of a compensatory collateral circulation. This hypothesis is supported by our finding that the magnitude of the tissue J
Cereb Blood Flow Metab, Vol. 12, No.6, 1992
rescued from infarction is greater with distal than with proximal MCA occlusions. Occlusion of the MCA proximal to the lenticulostriate branches cuts off blood supply in branches arising proximal to the olfactory tract and supplying the lateral aspect of the frontal lobe (Yamori et aI., 1976; Reike et aI., 1981). More distal MCA occlusions spare these branches, thereby increasing the potential for the development of a collateral circulation in the neo cortex. Thus, FN stimulation, unlike hypercapnia, may be effective in increasing CBF to the ischemic cortex, thereby preventing the occurrence of a "steal phenomenon" (Jones et aI., 1989). However, further studies are required to determine whether the protective effect exerted by FN stimulation de pends upon enhanced collateral flow to the isch emic territory. On the other hand, the possibility that the protec tive effect of FN stimulation on ischemic damage is independent of CBF and is mediated by local re lease of endogenous "cytoprotective" agents, neu rotransmitters (Reis et aI., 1991), cannot be ex cluded. In this regard, the observation that the cere brovasodilation elicited by FN stimulation is cholinergically mediated (ladecola et aI., 1986) sug gests a possible neuroprotective role for acetylcho line. Indeed, cholinergic agonists have been pro posed to ameliorate cerebral ischemia. However, these agents seem to mediate their effect through an increase in CBF to the ischemic territory (Scremin and Scremin, 1986). In addition, the recent finding that acetylcholine mediates the cerebrovasodilation elicited by FN stimulation through nitric oxide (NO) (ladecola, 1992b) raises the possibility that NO is involved in protecting the brain tissue from ischemic cell death. While there have been reports suggesting that NO contributes to ischemic neuro nal death (Dawson et aI., 1991; Nowicki et aI., 1991), preliminary studies suggest that arginine, the substrate for NO synthesis, ameliorates brain dam age in focal ischemia (M. Moskowitz, personal communication, 1992). The latter finding would in dicate that NO protects the brain from ischemia. Thus, the precise role of NO in ischemic damage remains to be elucidated. Whatever the mechanisms of the salvage, FN stimulation, like nimodipine, glutamate receptor an tagonists, or imidazole receptor agonists (Mohamed et aI., 1985; Park et aI., 1988; Maiese et aI., 1992), is able to rescue from infarction only the tissue lo cated at the border of the ischemic territory. Thus, the protection from ischemic damage exerted by FN stimulation may be effective only on the so called "ischemic penumbra," the transitional zone between ischemic core and normal tissue in which
BRAINSTEM STIMULATION AND FOCAL ISCHEMIA
neurons are electrically silent but still viable (see Hakim, 1987 for review). Consistent with this hy pothesis is our observation that FN stimulation en hances the recovery of the EEG amplitude after MCA occlusion. This finding may reflect regaining of spontaneous electrical activity in viable neurons rendered electrically silent by the ischemic insult. The origin of the neural pathways mediating the effect of FN stimulation on ischemic damage has not been elucidated. We have previously demon strated that the increases in CBF and AP elicited by electrical stimulation of the FN are mediated by excitation of fibers passing through or terminating in the area (Chida et al., 1989). Indeed, activation of FN neurons by local microinjection of excitatory amino acid produces decreases in the AP, CBF, and CGU (Chida et al., 1986,1989). Thus, if the mecha nisms by which FN stimulation limits the extent of ischemic damage depend upon enhancement of CBF to the ischemic area, the effect is more likely to be initiated by excitation of fibers of passage than from local FN neurons. The source of these neural processes has not yet been determined. Miura and Takayama (1988) have suggested that these fibers originate in the laterodorsal portion of the dorsal parabrachial nucleus of the pons. This hypothesis is supported by recent studies from our laboratory in which the distribution of neural activity evoked by FN stimulation was mapped using c-fos expression as a marker of active cells (Xu et aI., 1992). We found that one of the largest increases in the num ber of Fos-positive cells occurred bilaterally in the dorsal parabrachial nucleus, a region that does not receive monosynaptic projections from the FN (An drezik et aI., 1984). This finding is consistent with the hypothesis that electrical stimulation of the re gion of the FN activates reciprocal projections be tween the parabrachial nuclei as they cross the mid line via the subfastigial bundle (Miura and Takayama, 1988). In conclusion, this study demonstrates that exci tation of neural pathways associated with the FN, but not the DMRF, ameliorates functional impair ment and tissue damage resulting from MCA occlu sion. Our findings suggest that FN stimulation may ameliorate ischemic damage by enhancing blood flow to the ischemic brain without increasing the local metabolic demands. Thus, the present study supports the hypothesis that there are selected neu ral networks that can modulate the expression of the functional impairment and structural damage re sulting from focal cerebral ischemia. Acknowledgment: This work was supported by Grants in-Aid from the American Heart Association (Minne-
969
sota). F.Z. is a Fellow of the American Heart Association (Minnesota).
REFERENCES Andrezik JA, Dormer KJ, Foreman RD, Person RJ (1984) Fas tigial nucleus projections to the brain stem in beagles: Path ways for autonomic regulation. Neuroscience 11:497-507 Brint S, Jacewicz M, Kiessling M, Tanabe J, Pulsinelli W (1988) Focal brain ischemia in the rat: Methods for reproducible neocortical infarction using tandem occlusion of the distal middle cerebral and ipsilateral common carotid arteries. J Cereb Blood Flow Metab 8:474--485 Chida K, Iadecola C, Underwood M, Reis DJ (1986) A novel vasodepressor response elicited from the rat cerebellar fas tigial nucleus: The fastigial depressor response. Brain Res 370:378-382 Chida K, Iadecola C, Reis DJ (1989) Global reduction in cerebral blood flow and metabolism elicited from intrinsic neurons of fastigial nucleus. Brain Res 500: 177-192 Dawson VL, Dawson TM, London ED, Bredt DS, Snyder SH (1991) Nitric oxide mediates glutamate neurotoxicity in pri mary cortical cultures. Proc Natl Acad Sci USA 88:63686371 Ginsberg MD, Busto R (1989) Rodent models of cerebral isch emia. Stroke 20: 1627-1642 Hakim AM (1987) The cerebral ischemic penumbra. Can J Neu ral Sci 14:557-559 Iadecola C (l992a) Intrinsic and extrinsic neural regulation of the cerebral circulation. In: Stimulated Cerebral Blood Flow (Schmiedek P, Einhiiupl K, Kirsch CoM, eds), Heidelberg, Springer-Verlag, pp 19-36 Iadecola C (1992b) Nitric oxide participates in the cerebrova sodilation elicited from cerebellar fastigial nucleus. Am J Physiol (in press) Iadecola C, Nakai M, Arbit E, Reis DJ (l983a) Global cerebral vasodilation elicited by focal electrical stimulation within the dorsal medullary reticular formation in anesthetized rat. J Cereb Blood Flow Metab 3:270--279 Iadecola C, Nakai M, Mraovitch S, Ruggiero DA, Tucker LW, Reis DJ (l983b) Global increase in cerebral metabolism and blood flow produced by focal electrical stimulation of dorsal medullary reticular formation in rat. Brain Res 272;101-114 Iadecola C, Mraovitch S, Meeley M, Reis D (l983c) Lesions of the basal forebrain in rat selectively impair the cortical va sodilation elicited from cerebellar fastigial nucleus. Brain Res 279:41-52 Iadecola C, Underwood MD, Reis DJ (1986) Muscarinic cholin ergic receptors mediate the cerebrovasodilation elicited by stimulation of the cerebellar fastigial nucleus in rat. Brain Res 368:375-379 Jones SC, Bose B, Furlan AJ, Friel HT, Easley KA, Meredith MP, Little JR (1989) co2 reactivity and heterogeneity of ce rebral blood flow in ischemic, border zone, and normal cor tex. Am J Physiol 26: H473-H482 LeDoux JE, Thompson ME, Iadecola C, Tucker LW, Reis DJ (1983) Local cerebral blood flow increases during auditory and emotional processing in the conscious rat. Science 221: 576-578 Maiese K, Pek L, Berger SB, Reis DJ (1992) Reduction in focal cerebral ischemia by agents acting at imidazole receptors. J Cereb Blood Flow Metab 12:53--63 Miura M, Takayama K (1988) The site of the origin of the so called fastigial pressor response. Brain Res 473:352-358 Mohamed AA, Gotoh 0, Graham Dl, Osborne KA, McCulloch J, Mendelow AD, Teasdale GM, Harper AM (1985) Effect of pretreatment with the calcium antagonist nimodipine on lo cal cerebral blood flow and histopathology after middle ce rebral artery occlusion. Ann Neural 18:705-711 Nakai M, Iadecola C, Reis DJ (1982) Global cerebral vasodila-
J
Cereb Blood Flow Metab, Vol. 12, No.6, 1992
970
F. ZHANG AND C. IADECOLA
tion by stimulation of rat fastigial cerebellar nucleus. Am J Physiol 243: H226-H235 Nakai M, Iadecola C, Ruggiero DA, Tucker LW, Reis DJ (1983) Electrical stimulation of cerebellar fastigial nucleus in creases cerebrocortical blood flow without change in local metabolism: Evidence for an intrinsic system in brain for primary vasodilation. Brain Res 260:35-49 Nowicki JP, Duval D, Poignet H, Scatton B (1991) Nitric oxide mediates neuronal death after focal cerebral ischemia in the mouse. Eur J Pharmacol 204:339-340 Onai T, Takayama K, Miura M (1987) Projections to areas of the nucleus tractus solitarii related to circulatory and respiratory responses in cats. J Auton Nerv Syst 18: 163-175 Park CK, Nehls DG, Graham DI, Teasdale GM, McCulloch J (1988) The glutamate antagonist MK-801 reduces focal isch emic brain damage in the rat. Ann Neurol 24:543-551 Reike GK, Bowers DE, Penn P (1981) Vascular supply pattern to rat caudatoputamen and globus pallidus: Scanning electron microscopic study of vascular endocasts of stroke-prone vessels. Stroke 12:840-847 Reis DJ, Berger SB, Underwood MD, Khayata M (1991) Elec trical stimulation of cerebellar fastigial nucleus reduces isch-
J
Cereb Blood Flow Metab, Vol. ]2, No.6, 1992
emic infarction elicited by middle cerebral artery occlusion in rat. J Cereb Blood Flow Metab 11:810-818 Salgado AV, Jones SC, Furlan AJ, Korfali E, Marshall SA, Lit tle JR (1989) Bimodal treatment with nimodipine and low molecular-weight dextran for focal cerebral ischemia in the rat. Ann Neurol 26:621-627 Scremin OU, Scremin AME (1986) Physostigmine induced re versal of ischemia following acute middle cerebral artery occlusion in the rat. Stroke 17:1004-1009 Tamura A, Graham DI, McCulloch J, Teasdale GM (1981) Focal cerebral ischaemia in the rat: 1. Description of technique and early neuropathological consequences following middle ce rebral artery occlusion. J Cereb Blood Flow Metab 1:53-69 Xu X, Zhang F, Iadecola C (1992) Activation of the fastigial pressor area increases Fos-like immunoreactivity in brain stem autonomic nuclei. Soc Neurosci Abstr 18: 1177 Yamori Y, Horie R, Handa H, Sato M, Fukase M (1976) Patho genetic similarity of strokes in stroke-prone spontaneously hypertensive rats and humans. Stroke 7: 46-53 Zhang F, Iadecola C (1992) Fastigial nucleus stimulation reduces ischemic damage in neocortex but not striatum. [Abstract] Fed Proc 6:A1262
Stimulation of the Fastigial Nucleus Enhances EEG Recovery and Reduces Tissue Damage After Focal Cerebral Ischemia Fangyi Zhang and Costantino Iadecola Department of Neurology, University of Minnesota Medical School, Minneapolis, Minnesota, U.S.A.
Summary: Stimulation of the cerebellar fastigial nucleus (FN) increases CBF but not metabolism and reduces the tissue damage resulting from focal cerebral ischemia. This effect may result from enhancing CBF in the isch emic tissue without increasing local metabolic demands. To test this hypothesis, we studied whether the reduction in tissue damage is restricted to the neocortex, a region in which the CBF increase is independent of metabolism, and whether stimulation of the dorsal medullary reticular formation (DMRF), a treatment that increases both cere bral metabolism and CBF, also protects the brain from ischemia. In halothane-anesthetized Sprague-Dawley rats, the middle cerebral artery (MCA) was occluded ei ther proximally or distally to the lenticulostriate branches. The FN or DMRF were then stimulated for 1 h (50--100 f.LA; 50 Hz; 1 s onl 1 s oft). Twenty-four hours later, the infarct volume was determined. FN stimulation substantially reduced the size of the infarct, an effect that was greater with distal (-69 ± 8%; n 6; p < 0 .001 ;
mean ± SD) than with proximal ( - 38 ± 8%; n 8; p < 0.001) MCA occlusion. The reduction occurred only in neocortex (-43 ± 9%; p < 0.00 1) and not in striatum (-16 ± 21%; p > 0.05). Stimulation of the FN also en
Studies over the past several decades have dem onstrated that there are central neural networks that have a profound influence on the circulation and metabolism of the brain (see Iadecola (19920) for review). One such pathway involves the area of the rostral cerebellar fastigial nucleus (FN). Electrical stimulation of this region increases CBF globally through neural projections entirely contained within the brain (Nakai et al., 1982; Iadecola et al., 1983c). The increases in CBF are cholinergically mediated
(Iadecola et al., 1986) and are greatest in cerebral cortex (240% of control), wherein they are not as sociated with increased cerebral glucose utilization (CGU) (Nakai et aI., 1983). Another pathway in volves a restricted region of the dorsal medullary reticular formation (DMRF). Excitation of this re gion is able to increase CBF globally (Iadecola et aI., 19830). However, at variance with the FN, the increases in CBF evoked from the DMRF are asso ciated with, and presumably secondary to, in creased CGU (secondary metabolic vasodilation) (Iadecola et al., 1983b). It has been demonstrated recently that stimula tion of the FN reduces the tissue damage resulting from occlusion of the middle cerebral artery (MCA) in rat (Reis et al., 1991). FN stimulation may "pro tect" the ischemic tissue from infarction by enhanc ing CBF in the ischemic territory without increasing
=
hanced recovery of EEG amplitude in the ischemic cortex (+48%; p < 0.003). DMRF stimulation (n 7) did not affect the stroke size or EEG recovery (p > 0.05). Thus, stimulation of the FN, but not the DMRF, attenuates the damage reSUlting from focal ischemia. The finding that the protection is limited to the neocortex and that the amount of tissue salvaged is greater after distal occlusions suggests that the FN, unlike the DMRF, may rescue tis sue from infarction by enhancing collateral flow without increasing local energy requirements. Activation of se lected neural networks may modulate the expression of the damage resulting from focal cerebral ischemia. Key Words: Cerebellum-Medullary reticular formation Middle cerebral artery-Cerebral blood flow-Rat. =
=
Received April 4, 1992; final revision received June 8, 1992; accepted June 29, 1992. Address correspondence and reprint requests to Dr. C. Iadecola at Department of Neurology, University of Minnesota Medical School, Box 295 UMHC, 420 Delaware Street S.E., Minneapolis, MN 55455, U.S.A. Abbreviations used: AP, arterial pressure; CGU, cerebral glu cose utilization; DMRF, dorsal medullary reticular formation; FN, fastigial nucleus; MCA, middle cerebral artery.
962
BRAINSTEM STIMULATION AND FOCAL ISCHEMIA
963
local metabolic demands. Based on this hypothesis, the reduction in tissue damage should be restricted to the neocortex, a region in which the CBF in crease is independent of metabolism (Nakai et aI., 1983). Furthermore, stimulation of the DMRF, a treatment that increases CBF and metabolism, should not exert any protective effect. In this study, therefore, we sought to investigate whether the re duction in tissue damage exerted by FN stimulation is restricted to the cortex and whether also stimu lation of the DMRF reduces the extent of ischemic damage resulting from MCA occlusion. We also sought to determine whether activation of these re gions improves the functional state of the tissue af ter ischemia, as indicated by the recovery of spon taneous electrical activity. We shall demonstrate that stimulation of the FN, but not the DMRF, en hances the recovery of the EEG and reduces, only in neocortex, the tissue damage resulting from focal ischemia. Portions of this study have been pub lished in abstract form (Zhang and Iadecola, 1992).
METHODS Methods for anesthesia, surgical preparation of ani mals, and brain stimulation have been described in detail in previous publications (Nakai et ai., 1982; Iadecola et aI., 1983a,b,c, 1986) and will only be summarized.
General preparation of animal Studies were conducted on 40 male Sprague-Dawley rats weighing 280-350 g. Animals were anesthetized with halothane in 100% oxygen (5% induction, 2-3% during surgery, 1% maintenance) administered through a facial mask. The tail artery was cannulated for continuous mon itoring of the arterial pressure (AP) and heart rate, and rats were placed in a stereotaxic frame (Kopt). Body tem perature was maintained at 37 ± 0.5°C by a thermostati cally controlled lamp connected to a rectal probe. Small volumes of arterial blood (50 J.ll) were sampled at various times for measurement of P a02' P aco2' and pH by a blood-gas analyzer (Corning, model 178).
Brain stimulation and recording of the electroencephalogram· As described in detail elsewhere (Nakai et aI., 1982; Iadecola et al. 1983a), the FN or the DMRF was stimu lated with cathodal square-wave pulses (0.5 ms) through monopolar electrodes (tip diameter: 150 J.lm). The anode was a clip attached to the neck muscles. In all animals, stimulated sites were histologically verified in Nissl stained sections (Fig. 1). The EEG was recorded monopolarly through a stain less steel screw inserted to lie on the dura at a site 2 mm lateral and 2 mm caudal to the bregma ipsilaterally to the occluded MCA. The reference electrode was a clip at tached to the scalp muscles. A 30 s EEG epoch was dig itized from the record by a flat-bed scanner (Hewlett Packard). The X, Y coordinates of the tracing were de-
FIG. 1. Location of stimulated sites (black dots) in the region of the fastigial nucleus (FN) (upper panel) or dorsal medul lary reticular formation (DMRF) (lower panel). Py, pyramidal tract; SQ, superior olive; STN, spinal trigeminal nucleus; STT, spinal trigeminal tract; ve, vestibular complex; VII, fa cial nerve nucleus.
termined using a computer program for vectorizing TIFF files (UnGraph, Biosoft). The mean amplitude of the dominant frequency was then determined using wave analysis software (Superscope, GW Instruments). To minimize the confounding effects on the EEG of changes in AP or of fluctuations in the level of anesthesia, the AP was tightly controlled (see below) and the concentration of halothane was maintained constant throughout the pe riod of ischemia and brain stimulation.
Occlusion of the middle cerebral artery The MCA was exposed using an approach modified from that of Tamura et al. (1981). Briefly, a skin incision was made between the orbit and external auditory me atus. The coronoid process of the mandible was removed to expose the inferotemporal fossa. For proximal occlu sions (Fig. 2, site A), a 3-4 mm hole was drilled at a site
J Cereb Blood Flow Metab, Vol. /2, No.6, 1992
F. ZHANG AND C. IADECOLA
964
Int. Cerebral Vein
image analysis system (MCID, MI, Imaging Research Inc.). Brain sections were digitized and the borders of the lesion were traced using a mouse. The volume of the infarct (mm3) was computed on line as the product of the cross-sectional area and distance between sections. To reconstruct the rostrocaudal extent of the lesion, the cross-sectional area of the stroke was plotted as a func tion of the distance from the interaural line (Mohamed et aI., 1985).
Experimental protocol
FIG. 2. Location of sites at which the middle cerebral artery (MeA) was occluded. For proximal occlusions (site A), the MeA was cauterized from the origin of the lenticulostriate branches to the inferior edge of the olfactory tract and sev ered proximally to the lenticulostriate branches. For distal occlusions (site B), the MeA was cauterized from the inferior cerebral vein to the superior edge of the olfactory tract and severed at the level of the inferior cerebral vein. superior and lateral to the foramen ovale to expose the proximal portion of the MCA. The MCA was elevated and cauterized just proximally to the origin of the lentic ulostriate branches. The occlusion was extended 2-3 mm distally and the MCA was transected. For distal occlu sions (Fig. 2, site B), the MCA was elevated and cauter ized just proximally to the inferior cerebral vein (Fig. 2, site B). The occlusion was extended up to the superior edge of the olfactory tract and the MCA was transected. In most rats, distal or proximal MCA occlusions resulted in sustained reductions in the amplitude of the EEG (6080%). These animals were found to have large infarcts involving mostly the cortex (distal occlusion) or cortex and striatum (proximal occlusion). Rats in which MCA occlusion did not substantially decrease the EEG ampli tude «40%) or in which the EEG recovered spontane ously suffered small cortical infarcts or infarcts restricted to the striatum (Salgado et aI., 1989). These animals were not included in the study.
Determination of infarct volumes Twenty-four hours after MCA occlusion, rats were deeply anesthetized with 5% halothane and mounted on a stereotaxic frame. The coronal plane corresponding to the interauralline was marked by inserting an electrode at the stereotaxic coordinates of the interaural line. This mark was then used as a reference to reconstruct the rostrocaudal extent of the lesion. Rats were then killed by decapitation and their brains were rapidly removed and frozen in isopentane at -30°C. Coronal sections (30 ""m) were cut in a cryostat (-20°C), collected at 300 ""m in tervals, and stained with thionin using conventionalmeth ods. The boundaries of the infarct were usually well de marcated on thionin-stained sections. When the borders of the lesion were not clear, they were verified micro scopically. The infarct volume was determined using an
J Cereb Blood Flow Me/ab, Vol. 12, No.6, 1992
Rats were surgically prepared and the MCA was ex posed as described above. In animals in which the FN was stimulated, an electrode was inserted into the cere bellum at a site 5 mm rostral to the obex and 0.8 mm lateral to the midline. The most active pressor site in the FN was then localized using procedures described in de tail elsewhere (Nakai et aI., 1982). In animals in which the DMRF was stimulated, the electrode was inserted through the cerebellum at a site 1 mm lateral to midline and 3 mm anterior to the obex (Iadecola et aI., 1983a,b). Once the most active site in the FN or DMRF was local ized, procedures for occlusion of the MCA were begun. The stereotaxic frame was rotated to place the animal in a lateral position and the MCA occluded as described above. Occurrence of neocortical ischemia was indicated by a 60-80% reduction in the amplitude of the EEG. Three to 5 min after observing the reduction in EEG am plitude, stimulation of the FN or DMRF was begun. As described in detail elsewhere (Nakai et aI., 1982; Iadecola et aI., 1983a,b), the FN or DMRF were stimulated with intermittent trains of stimuli (I s on/I s off; 50 Hz). During the first 2-4 min, the current intensity was gradually in creased up to five times the threshold current (i.e., the stimulus current producing a 10 mm Hg elevation in the AP, usually 10-20 ""A). At the same time, the evoked elevations in the AP were offset by slow removal of small amounts (1-2 ml) of arterial blood. By this procedure, the AP did not differ between stimulated and unstimulated rats (Table 1) and the maximal AP in stimulated animals was 115 ± 10 mm Hg. The FN or DMRF was stimulated for 1 h. At the end of the stimulation period, catheters were removed and wounds infiltrated with lidocaine and sutured. Animals were allowed to recover and carefully observed during the postoperative period for the occur rence of seizures. Twenty-four hours later, rats were killed for determination of the infarct size. Unstimulated rats and rats that underwent DMRF stimulation tended to have a Peo2 higher than that of awake, behaving rats (e.g., LeDoux et al., 1983) (Table 1). This degree of hy percapnia is commonly observed in spontaneously breathing halothane-anesthetized rats undergoing MCA occlusion (e.g., Brint et a!., 1988; Reis et a!., 1991; Maiese et a!., 1992). The lower Peo2 of rats in which the FN was stimulated (Table 1) can be attributed to the in crease in the ventilatory rate resulting from stimulation of the fastigial pressor area (Onai et a!., 1987). Six groups of rats were studied (Table I). The first group (n 6) consisted of rats in which the MCA was exposed and not occluded. The second group (n 7) consisted of rats in which the MCA was occluded but no brain stimulation was carried out. The third and fourth groups consisted of rats in which the MCA was occluded proximally and the FN (n 8) or the DMRF (n 7) was stimulated. The fifth group (n 6) consisted of unstim ulated rats in which the MCA was occluded distally. The =
=
=
=
=
965
BRAINSTEM STIMULATION AND FOCAL ISCHEMIA
TABLE 1. Body weight, arterial pressure, and blood gases of rats with and without stimulation of FN or DMRF after
MeA occlusion Proximal occlusion
Sham
MAP (mm Hg) Pe o2 (mm Hg) P02 (mm Hg) pH
n
FN stimulation
Unstimulated
Unstimulated
FN stimulation
322 ± 18
323 ± 22
322 ± 15
313 ± 24
317 ± 29
82 ± 9
90 ± 9
93 ± 5
87 ± 6
86 ± 7
95 ± 4
63 ± 3
56 ± 5
41 ± 6a
52 ± 5
56 ± 5
42 ± 7a
321 ± 31
BW
Distal occlusion DMRF stimulation
b
205 ± 53 7.34 ± 0.03 6
247 ± 74 7.49 ± 0.03 7
252 ± 71 7.48 ± 0.03 8
318 ± 50 7.38 ± 0.04a 7
236 ± 58 7.41 ± O.03a 6
250 ± 27 7.49 ± 0.04 6
Values are mean ± SD. BW: body weight; MAP: mean arterial pressure. MAP and blood gases values represent the average of three measurements taken at 5, 30, and 60 min after MCA occlusion. a p < 0.05 from sham and unstimulated groups (analysis of variance and Tukey's test). b p < 0.05 from stimulated groups.
sixth group (n 6) consisted of rats in which the MCA was occluded distally and the FN was stimulated. =
Data analysis Data are expressed as means ± SD. Comparisons among multiple groups were statistically evaluated by analysis of variance and Tukey's test (Systat, Inc.) (Ta bles 1 and 2 and Fig. 6). Two group comparisons were evaluated by the unpaired t test (Fig. 3). For all statistical procedures differences were considered significant for p < 0.05.
RESULTS
et aI., 1992). When the cross-sectional area of the stroke was plotted as a function of the distance from the interaural line (Fig. 4), the largest area of infarc tion occurred approximately 6 mm rostral to the interaural line. Distal occlusions of the MeA (Fig. 2, site B) (n 6) resulted in infarcts involving al most exclusively the cerebral cortex (Table 2). The size of the lesion and the rostrocaudal distribution of its cross-sectional area were comparable to those of neocortical infarcts resulting from proximal MeA occlusion (Fig. 4; Table 2). =
Effect of proximal or distal MeA occlusion on size
Effect of stimulation of the FN or DMRF on the
and regional distribution of the infarct
size of the infarct resulting from MeA occlusion
In rats in which the MeA was exposed and ele vated but not occluded (sham occlusion; n 6), a small infarct restricted to the cortex adjacent to the surgical site was observed (24 ± 14 mm3). Occlu sion of the MeA proximal to the origin of the len ticulostriate arteries (Fig. 2, site A; n 7) resulted in infarcts involving neocortex and striatum (Table 2, Fig. 3). The regional distribution of the infarct was comparable to that previously described in the rat (Fig. 3) (Tamura et aI., 1981; Park et aI., 1988; Ginsberg and Busto, 1989; Reis et aI., 1991; Maiese =
=
With FN stimulation (n 8), the size of the total infarct resulting from proximal MeA occlusion was reduced by 38 ± 8% (p < 0.001 from the unstimu lated group; analysis of variance and Tukey's test) (Table 2). In the neocortex, the volume of the in farct was reduced by 43 ± 9% (p < 0.001) while in the striatum the stroke size was not affected (p > 0.05). The regions of cerebral cortex spared from infarction included the dorsomedial portion of the frontal, parietal, and occipital cortex (Fig. 4). No effect was observed below the rhinal fissure. In rats =
TABLE 2. Effect of FN or DMRF stimulation on the volume of the infarct resulting from occlusion of the middle
cerebral artery Proximal occlusion
Total Cortex Striatum
n
Distal occlusion
Unstimulated
FN stimulation
DMRF stimulation
Unstimulated
FN stimulation
330 ± 41 250 ± 45 57 ± 8 7
205 ± na 144 ± 65a 49 ± 9 8
322 ± 52 241 ± 49 65 ± 12 7
223 ± 28 207 ± 25 10 ± 14 6
73 ± 18* 64 ± 14* 5 ± 6 6
Values are mean ± SD. a p < 0.001 (analysis of variance and Tukey's test).
J Cereb Blood Flow Metab, Vol. /2, No.6. 1992
966
F. ZHANG AND C. IADECOLA
NEOCORTEX 60 C
A N
E E -
Unstimulated FN Stimulation *
40
*
*
as
�
as
-
20
o a-
as
-
c:
O��WW����
15 12 9 6 3 o ·3 Distance from interaural line (mm)
STRIATUM 20 UNSTIMULATED
FN STIM
*
DMRFSTIM
FIG. 3. Distribution of the cross-sectional area of the infarct
resulting from proximal occlusion of the middle cerebral ar tery (MCA) with and without (left column) stimulation of the fastigial nucleus (FN) or dorsal medullary reticular formation (DMRF). Thionin-stained sections were projected on paper using a microprojector, and the area of necrosis was traced. After MCA occlusion, stimulation of the FN (middle column), but not the DMRF (right column), leads to smaller infarcts. The area rescued from infarction involves the dorsomedial aspect of the frontal, parietal, and occipital cortex. FN stim ulation does not affect stroke size in subcortical regions.
in which the MeA was occluded more distally (Fig. 2, site B), FN stimulation (n 6) reduced the total infarct size by 69 ± 8%, a reduction significantly larger than that observed after proximal occlusion (p < 0.001). In contrast to the FN, stimulation of the DMRF did not influence the volume of the in farct resulting from proximal MeA occlusion (Table 2, Fig. 4) (p > 0. 05). The rostrocaudal distribution of the cross-sectional area of the infarct in animals with or without FN stimulation is illustrated in Fig. 4. The greatest reductions in the total cross sectional area occurred between �O and 7 mm ros tral to the interaural line (Fig. 3) (p < 0.05). In the striatum, the area was slightly reduced only at the rostrocaudal level, where the stroke was largest (Fig. 3, p < 0. 05). The effect of FN stimulation on the cross-sectional area of infarcts resulting from distal MeA occlusion was similar to that observed after proximal occlusion. These observations indi cate that, in agreement with the study of Reis et al. =
J
Cereb Blood Flow Metab, Vol. 12, No.6, 1992
N
E E as Q)
r
15
10
a-
as
-
0 a-
as c:
-
5
o
I � I I ���__�������...�.�I-L�
-3
o
6
3
9
12
15
Distance from interaural line (mm) FIG. 4. Cross-sectional area of the infarct resulting from proximal MCA occlusion plotted as a function of the distance from the interaural line. Note that the area of stroke in the cortex (upper panel) is smaller in FN-stimulated animals, par ticularly in the caudal regions of the infarct. In striatum (lower panel), a small reduction was observed only at the site where the infarct was largest. Asterisks indicate p values of less than 0.01 (t test).
(1991), the largest amount of tissue rescued from infarction involves the dorsomedial and caudal por tions of the infarct. Effect of FN and DMRF stimulation on the reduction in EEG amplitude after MeA occlusion
In unstimulated rats (n 7), proximal MeA oc clusion reduced the EEG amplitude by 73 ± 8% (p =
BRAINSTEM STIMULATION AND FOCAL ISCHEMIA
967
AFTER OCCLUSION
BEFORE OCCLUSION
Smin
60 min
�
UNSTIM. FIG. 5. Effect of stimulation of the fastigial nucleus (FN) or dorsal medullary reticular formation (DMRF) on the reduction in EEG amplitude resulting from MeA occlusion. Five minutes after MeA occlusion, the EEG amplitude is reduced equally in all groups (middle panel). FN or DMRF stimulation was initiated at this time. Note that FN stimulation (right panel, middle trace) enhances the recovery of the EEG amplitude after 60 min of ischemia.
FN STIM.
DMRF STIM.
�
� 1
-- L..____
FN or DMRF stim
�0.5mv 2 sec < 0.001; Fig. 5). Sixty minutes later, the EEG am plitude recovered by 13 ± 12% (Figs. 5 and 6). With FN stimulation (n = 8), the initial EEG depression was not different from that of unstimulated controls (72 ± 5%; p > 0.05 from the unstimulated group; Fig. 5). However, 60 min after occlusion, the EEG recovery was significantly greater than that of the unstimulated group (Fig. 6; p < 0.003). Stimulation of the DMRF did not influence the changes in am plitude of the EEG resulting from MCA occlusion (Figs. 5 and 6; p > 0.05 from the unstimulated group). Thus, stimulation of the FN, but not the DMRF, enhances the recovery of the EEG after neocortical ischemia.
the functional impairment and tissue damage result ing from cerebral cortical ischemia. Therefore, the present study confirms and extends the results of Reis et al. (1991), who first described that FN stim ulation reduces the size of the infarct resulting from MCA occlusion. The effect of FN stimulation on the size of the infarct is unlikely to be a consequence of the lower peo2 of the rats in which the FN was stimulated since differences in peo2 comparable to those seen between stimulated and unstimulated rats do not appear to influence the tissue damage resulting from focal ischemia (Maiese et al., 1992), nor can the
100
DISCUSSION
This study sought to determine whether activa tion of the FN or the DMRF, regions associated with intracerebral pathways that have profound ef fects on CBF and/or CGU, could influence the func tional impairment and tissue damage resulting from focal cerebral ischemia. We found that electrical stimulation of the FN is able to revert the depres sion in EEG amplitude and to reduce the size of the hemispheric infarct resulting from proximal or dis tal MCA occlusions. Analysis of the regional distri bution of the effect revealed that the brain tissue rescued from infarction is restricted to the neocor tex, as in the striatum the size of the infarct is not affected. In contrast to the FN, stimulation of the DMRF did not influence the EEG changes and the size of the infarct resulting from MCA occlusion. These findings indicate that excitation of the fasti gial pressor area, but not the DMRF, ameliorates
�
Mean±SD !!lI Unstimulated III DMRF Stimulation • FN stimulation * p
-
�o
o Q) a: " w w
50 25 0 '--
__-"-'
FIG. 6. Effect of fastigial nucleus (FN) or dorsal medullary
reticular formation (DMRF) stimulation on the recovery of the EEG amplitude after MeA occlusion. EEG recovery is the difference between the EEG amplitude at 5 and 60 min, nor malized with respect to the preocclusion amplitude and ex pressed as a percentage of the EEG amplitude at 5 min. Stim ulation of the FN but not DMRF enhances the EEG recovery at 60 min after MeA occlusion.
J
Cereb Blood Flow Metab, Vol. /2, No.6, 1992
968
F. ZHANG AND C. IADECOLA
protective effect exerted by FN stimulation be a consequence of the controlled hemorrhage per formed to prevent hypertension. Removal of 1-2 ml of blood does not affect the resting CBF and pro duces only minimal hemodilution (ladecola et aI., 1983a). These changes cannot account for the sub stantial reduction in infarct size elicited by FN stim ulation. Furthermore, similar amounts of blood were removed during DMRF stimulation, a treat ment that does not affect the magnitude of ischemic damage. It is also unlikely that the effect of FN stimulation on stroke size is a nonspecific conse quence of brain stimulation as electrical stimulation of the DMRF (present study) or of the dentate nu cleus (Reis et aI., 1991) does not influence the size of the infarct. Therefore, the reduction in stroke size produced by FN stimulation appears to be a direct consequence of activation of neural pathways originating or passing through the region of the ros tral FN. The reduction in stroke size elicited by stimula tion of the FN is regionally selective as it occurs in the neocortex and not in the striatum. This finding is of interest in view of the fact that the mechanisms mediating the cerebrovascular actions of FN stim ulation in the neocortex and striatum are substan tially different. While in the neocortex the increases in CBF are independent of increased metabolic ac tivity, in the striatum the vasodilation is associated with, and presumably secondary to, increased me tabolism (Nakai et aI., 1983). Therefore, in the neo cortex, FN stimulation may increase substrate de livery without enhancing the energy requirements of the tissue. An excess of substrates may result in a greater resistance to energy failure during isch emia. In support of this hypothesis is our observa tion that stimulation of the DMRF, a treatment that increases CBF by increasing the metabolic de mands of the tissue (Iadecola et aI., 1983b), does not ameliorate ischemic damage. The regional selectivity of the effect of FN stim ulation could also result from differences in the pat tern of vascular supply in the neocortex and stria tum. The blood supply of the striatum originates largely from terminal arteries (Yamori et aI., 1976). Thus, the potential for development of a compen satory collateral circulation in the striatum is less than in the neocortex. The striatum is therefore more vulnerable to ischemic damage and could be less sensitive to the protective effects of FN stim ulation. These observations may indicate that FN stimulation improves the outcome of focal ischemia by enhancing the development of a compensatory collateral circulation. This hypothesis is supported by our finding that the magnitude of the tissue J
Cereb Blood Flow Metab, Vol. 12, No.6, 1992
rescued from infarction is greater with distal than with proximal MCA occlusions. Occlusion of the MCA proximal to the lenticulostriate branches cuts off blood supply in branches arising proximal to the olfactory tract and supplying the lateral aspect of the frontal lobe (Yamori et aI., 1976; Reike et aI., 1981). More distal MCA occlusions spare these branches, thereby increasing the potential for the development of a collateral circulation in the neo cortex. Thus, FN stimulation, unlike hypercapnia, may be effective in increasing CBF to the ischemic cortex, thereby preventing the occurrence of a "steal phenomenon" (Jones et aI., 1989). However, further studies are required to determine whether the protective effect exerted by FN stimulation de pends upon enhanced collateral flow to the isch emic territory. On the other hand, the possibility that the protec tive effect of FN stimulation on ischemic damage is independent of CBF and is mediated by local re lease of endogenous "cytoprotective" agents, neu rotransmitters (Reis et aI., 1991), cannot be ex cluded. In this regard, the observation that the cere brovasodilation elicited by FN stimulation is cholinergically mediated (ladecola et aI., 1986) sug gests a possible neuroprotective role for acetylcho line. Indeed, cholinergic agonists have been pro posed to ameliorate cerebral ischemia. However, these agents seem to mediate their effect through an increase in CBF to the ischemic territory (Scremin and Scremin, 1986). In addition, the recent finding that acetylcholine mediates the cerebrovasodilation elicited by FN stimulation through nitric oxide (NO) (ladecola, 1992b) raises the possibility that NO is involved in protecting the brain tissue from ischemic cell death. While there have been reports suggesting that NO contributes to ischemic neuro nal death (Dawson et aI., 1991; Nowicki et aI., 1991), preliminary studies suggest that arginine, the substrate for NO synthesis, ameliorates brain dam age in focal ischemia (M. Moskowitz, personal communication, 1992). The latter finding would in dicate that NO protects the brain from ischemia. Thus, the precise role of NO in ischemic damage remains to be elucidated. Whatever the mechanisms of the salvage, FN stimulation, like nimodipine, glutamate receptor an tagonists, or imidazole receptor agonists (Mohamed et aI., 1985; Park et aI., 1988; Maiese et aI., 1992), is able to rescue from infarction only the tissue lo cated at the border of the ischemic territory. Thus, the protection from ischemic damage exerted by FN stimulation may be effective only on the so called "ischemic penumbra," the transitional zone between ischemic core and normal tissue in which
BRAINSTEM STIMULATION AND FOCAL ISCHEMIA
neurons are electrically silent but still viable (see Hakim, 1987 for review). Consistent with this hy pothesis is our observation that FN stimulation en hances the recovery of the EEG amplitude after MCA occlusion. This finding may reflect regaining of spontaneous electrical activity in viable neurons rendered electrically silent by the ischemic insult. The origin of the neural pathways mediating the effect of FN stimulation on ischemic damage has not been elucidated. We have previously demon strated that the increases in CBF and AP elicited by electrical stimulation of the FN are mediated by excitation of fibers passing through or terminating in the area (Chida et al., 1989). Indeed, activation of FN neurons by local microinjection of excitatory amino acid produces decreases in the AP, CBF, and CGU (Chida et al., 1986,1989). Thus, if the mecha nisms by which FN stimulation limits the extent of ischemic damage depend upon enhancement of CBF to the ischemic area, the effect is more likely to be initiated by excitation of fibers of passage than from local FN neurons. The source of these neural processes has not yet been determined. Miura and Takayama (1988) have suggested that these fibers originate in the laterodorsal portion of the dorsal parabrachial nucleus of the pons. This hypothesis is supported by recent studies from our laboratory in which the distribution of neural activity evoked by FN stimulation was mapped using c-fos expression as a marker of active cells (Xu et aI., 1992). We found that one of the largest increases in the num ber of Fos-positive cells occurred bilaterally in the dorsal parabrachial nucleus, a region that does not receive monosynaptic projections from the FN (An drezik et aI., 1984). This finding is consistent with the hypothesis that electrical stimulation of the re gion of the FN activates reciprocal projections be tween the parabrachial nuclei as they cross the mid line via the subfastigial bundle (Miura and Takayama, 1988). In conclusion, this study demonstrates that exci tation of neural pathways associated with the FN, but not the DMRF, ameliorates functional impair ment and tissue damage resulting from MCA occlu sion. Our findings suggest that FN stimulation may ameliorate ischemic damage by enhancing blood flow to the ischemic brain without increasing the local metabolic demands. Thus, the present study supports the hypothesis that there are selected neu ral networks that can modulate the expression of the functional impairment and structural damage re sulting from focal cerebral ischemia. Acknowledgment: This work was supported by Grants in-Aid from the American Heart Association (Minne-
969
sota). F.Z. is a Fellow of the American Heart Association (Minnesota).
REFERENCES Andrezik JA, Dormer KJ, Foreman RD, Person RJ (1984) Fas tigial nucleus projections to the brain stem in beagles: Path ways for autonomic regulation. Neuroscience 11:497-507 Brint S, Jacewicz M, Kiessling M, Tanabe J, Pulsinelli W (1988) Focal brain ischemia in the rat: Methods for reproducible neocortical infarction using tandem occlusion of the distal middle cerebral and ipsilateral common carotid arteries. J Cereb Blood Flow Metab 8:474--485 Chida K, Iadecola C, Underwood M, Reis DJ (1986) A novel vasodepressor response elicited from the rat cerebellar fas tigial nucleus: The fastigial depressor response. Brain Res 370:378-382 Chida K, Iadecola C, Reis DJ (1989) Global reduction in cerebral blood flow and metabolism elicited from intrinsic neurons of fastigial nucleus. Brain Res 500: 177-192 Dawson VL, Dawson TM, London ED, Bredt DS, Snyder SH (1991) Nitric oxide mediates glutamate neurotoxicity in pri mary cortical cultures. Proc Natl Acad Sci USA 88:63686371 Ginsberg MD, Busto R (1989) Rodent models of cerebral isch emia. Stroke 20: 1627-1642 Hakim AM (1987) The cerebral ischemic penumbra. Can J Neu ral Sci 14:557-559 Iadecola C (l992a) Intrinsic and extrinsic neural regulation of the cerebral circulation. In: Stimulated Cerebral Blood Flow (Schmiedek P, Einhiiupl K, Kirsch CoM, eds), Heidelberg, Springer-Verlag, pp 19-36 Iadecola C (1992b) Nitric oxide participates in the cerebrova sodilation elicited from cerebellar fastigial nucleus. Am J Physiol (in press) Iadecola C, Nakai M, Arbit E, Reis DJ (l983a) Global cerebral vasodilation elicited by focal electrical stimulation within the dorsal medullary reticular formation in anesthetized rat. J Cereb Blood Flow Metab 3:270--279 Iadecola C, Nakai M, Mraovitch S, Ruggiero DA, Tucker LW, Reis DJ (l983b) Global increase in cerebral metabolism and blood flow produced by focal electrical stimulation of dorsal medullary reticular formation in rat. Brain Res 272;101-114 Iadecola C, Mraovitch S, Meeley M, Reis D (l983c) Lesions of the basal forebrain in rat selectively impair the cortical va sodilation elicited from cerebellar fastigial nucleus. Brain Res 279:41-52 Iadecola C, Underwood MD, Reis DJ (1986) Muscarinic cholin ergic receptors mediate the cerebrovasodilation elicited by stimulation of the cerebellar fastigial nucleus in rat. Brain Res 368:375-379 Jones SC, Bose B, Furlan AJ, Friel HT, Easley KA, Meredith MP, Little JR (1989) co2 reactivity and heterogeneity of ce rebral blood flow in ischemic, border zone, and normal cor tex. Am J Physiol 26: H473-H482 LeDoux JE, Thompson ME, Iadecola C, Tucker LW, Reis DJ (1983) Local cerebral blood flow increases during auditory and emotional processing in the conscious rat. Science 221: 576-578 Maiese K, Pek L, Berger SB, Reis DJ (1992) Reduction in focal cerebral ischemia by agents acting at imidazole receptors. J Cereb Blood Flow Metab 12:53--63 Miura M, Takayama K (1988) The site of the origin of the so called fastigial pressor response. Brain Res 473:352-358 Mohamed AA, Gotoh 0, Graham Dl, Osborne KA, McCulloch J, Mendelow AD, Teasdale GM, Harper AM (1985) Effect of pretreatment with the calcium antagonist nimodipine on lo cal cerebral blood flow and histopathology after middle ce rebral artery occlusion. Ann Neural 18:705-711 Nakai M, Iadecola C, Reis DJ (1982) Global cerebral vasodila-
J
Cereb Blood Flow Metab, Vol. 12, No.6, 1992
970
F. ZHANG AND C. IADECOLA
tion by stimulation of rat fastigial cerebellar nucleus. Am J Physiol 243: H226-H235 Nakai M, Iadecola C, Ruggiero DA, Tucker LW, Reis DJ (1983) Electrical stimulation of cerebellar fastigial nucleus in creases cerebrocortical blood flow without change in local metabolism: Evidence for an intrinsic system in brain for primary vasodilation. Brain Res 260:35-49 Nowicki JP, Duval D, Poignet H, Scatton B (1991) Nitric oxide mediates neuronal death after focal cerebral ischemia in the mouse. Eur J Pharmacol 204:339-340 Onai T, Takayama K, Miura M (1987) Projections to areas of the nucleus tractus solitarii related to circulatory and respiratory responses in cats. J Auton Nerv Syst 18: 163-175 Park CK, Nehls DG, Graham DI, Teasdale GM, McCulloch J (1988) The glutamate antagonist MK-801 reduces focal isch emic brain damage in the rat. Ann Neurol 24:543-551 Reike GK, Bowers DE, Penn P (1981) Vascular supply pattern to rat caudatoputamen and globus pallidus: Scanning electron microscopic study of vascular endocasts of stroke-prone vessels. Stroke 12:840-847 Reis DJ, Berger SB, Underwood MD, Khayata M (1991) Elec trical stimulation of cerebellar fastigial nucleus reduces isch-
J
Cereb Blood Flow Metab, Vol. ]2, No.6, 1992
emic infarction elicited by middle cerebral artery occlusion in rat. J Cereb Blood Flow Metab 11:810-818 Salgado AV, Jones SC, Furlan AJ, Korfali E, Marshall SA, Lit tle JR (1989) Bimodal treatment with nimodipine and low molecular-weight dextran for focal cerebral ischemia in the rat. Ann Neurol 26:621-627 Scremin OU, Scremin AME (1986) Physostigmine induced re versal of ischemia following acute middle cerebral artery occlusion in the rat. Stroke 17:1004-1009 Tamura A, Graham DI, McCulloch J, Teasdale GM (1981) Focal cerebral ischaemia in the rat: 1. Description of technique and early neuropathological consequences following middle ce rebral artery occlusion. J Cereb Blood Flow Metab 1:53-69 Xu X, Zhang F, Iadecola C (1992) Activation of the fastigial pressor area increases Fos-like immunoreactivity in brain stem autonomic nuclei. Soc Neurosci Abstr 18: 1177 Yamori Y, Horie R, Handa H, Sato M, Fukase M (1976) Patho genetic similarity of strokes in stroke-prone spontaneously hypertensive rats and humans. Stroke 7: 46-53 Zhang F, Iadecola C (1992) Fastigial nucleus stimulation reduces ischemic damage in neocortex but not striatum. [Abstract] Fed Proc 6:A1262
