Motor Performance, Histologic Damage, and Calcium ... - SAGE Journals

Journal of Cerebral Blood Flow and Metabolism 14:262-268 © 1994 The International Society of Cerebral Blood Flow and Metabolism Published by Raven Press, Ltd., New York

Motor Performance, Histologic Damage, and Calcium Influx in Rats Treated with NBQX after Focal Ischemia

T. J. DeGraba, *P. Ostrow, S. Hanson, and J. C. Grotta Department of Neurology, The University of Texas Medical School at Houston, Houston, Texas, and *Department of Pathology, State University of New York at Buffalo, Buffalo, New York, U.S.A.

Summary: 2,3-Dihydroxy-6-nitro-7-sulfamoylbenzo(F)­ quinoxaline (NBQX), an ex-amino-3-hydroxy-5-methyl-4isoxazole propionic acid (AMPA) receptor antagonist, has been reported to provide neuronal protection after global ischemia. The objectives of this study were to eval­ uate the neuroprotective effects of NBQX initiated after focal cortical ischemia and to validate a method for mea­ suring functional outcome in this model. Male spontane­ ously hypertensive rats (SHRs) were exposed to various durations of transient or permanent tandem middle cere­ bral artery (MCA) occlusion. Studies compared motor performance using balance beam and prehensile-traction tests, calcium-calmodulin (Ca-CaM) binding by immuno­ histochemistry, and infarct volume between NBQX­ treated animals [intravenous (i.v.) 5 mg/kg/h x 6 h or intraperitoneal (i.p.) 30 mg/kg q 30 min x 3 begun postis­ chemia] and controls. All ischemic groups performed less well than sham-operated controls on the motor perfor­ mance tasks in proportion to the severity of ischemia. No

significant improvement in motor performance was noted in the NBQX-treated versus the control animals after 1 h or permanent MCAICCA occlusion. Treatment with NBQX (i.v. or i.p. dosing) did not reduce Ca-CaM bind­ ing after 1 h of occlusion with 1 h of reperfusion or after 2 h of occlusion. Similarly, there was no reduction in infarct size between NBQX-treated and control animals after 24 h of permanent MCAICCA occlusion (74.6 ± 7.1 vs. 80.1 ± 6.0 ml; ns) or after 1 h of occlusion with 23 h of reperfusion (55.1 ± 4.4 vs. 47.4 ± 6.2 ml; ns). We conclude that therapy with NBQX alone has no protec­ tive effect in this model of ischemia, which may be related in part to its inability to prevent a postischemic increase in intracellular Ca in cortical neurons, and that our method of measuring motor performance is a useful indi­ cator of functional outcome in this model. Key Words: Calcium-Calmodulin-Focal ischemia-Motor perfor­ mance-NBQX.

The process of evaluating the effectiveness of a neuronal protective agent after an ischemic insult is multifaceted. In addition to measuring the preser­ vation of neuronal populations by histologic analy­ sis, the ultimate goal of a neuroprotective agent in

acute ischemic injury is the preservation of func­ tional and behavioral activity. Inasmuch as the mid­ dle cerebral artery (MCA)/common carotid artery (CCA) tandem occlusion model mirrors focal infarc­ tion in humans, determination of a reliable method for testing functional motor performance, along with histologic survival, and establishment of the mechanism of action are essential for the preclinical evaluation of a possible neuroprotective agent. The discovery of the competitive non-N-methyl­ D-aspartate (NMDA) glutamate receptor antagonist quinoxalinediones provides the opportunity to ex­ plore the neuroprotective benefit of non-NMDA re­ ceptor blockade after ischemic injury. The potent and selective u-amino-3-hydroxy-5-methyl-4isoxazole propionic acid (AMPA) antagonist, 2,3dihydroxy-6-nitro-7 -sulfamoyl-benzo(F)quinox­ aline (NBQX), has been shown to be effective in reducing neuronal loss in the CAl region of the hip­ pocampus after brief durations of global ischemia

Received April 5, 1993; final revision received August 17, 1993; accepted August 17, 1993. Address correspondence and reprint requests to Dr. J. C. Grotta at Department of Neurology, University of Texas School of Medicine, 6431 Fannin, Suite 7.044 MSB, Houston, TX 77030, U. S. A. The present address of Dr. T. J. DeGraba is Stroke Branch, National Institute of Neurologic Disorder and Stroke, 9000 Rockville Pike, Building 36, Room 40-04, Bethesda, MD 20892, U. S. A. Abbreviations used: AMPA, a-amino-3-hydroxy-5-methyl-4isoxazole propionic acid; ANOVA, analysis of variance; Ca, cal­ cium; CaM, calmodulin; Lp. , intraperitoneal; Lv. , intravenous; MCA, middle cerebral artery; NBQX, 2,3-dihydroxy-6-nitro-7sulfamoylbenzo(F)quinoxaiine; NMDA, N-methyl-D-aspartate; SHR, spontaneously hypertensive rat.

262

NEUROPROTECTION

(Buchan et aI., 1992; Nellgard et aI., 1992; Pulsinelli and Cho, 1992), as well as in transient focal neo­ cortical ischemia (Buchan et aI., 1991). These agents are of particular interest since they have no significant activity at the glutamate, glycine, or phencyclidine binding sites of the NMDA receptor (Sheardown et aI., 1990). Because of the behavioral side effects that may arise from the use of agents that antagonize the NMDA-receptor complex (Wil­ letts et at., 1990; Cosgrove and Newell, 1991), the quinoxalinediones represent a possible alternative for ischemic neuronal protection. In this study, using three separate methods, we endeavored to assess the protective role of NBQX begun after a focal cortical ischemic insult. First, we evaluated its ability to reduce the infarct vol­ ume. Second, we determined whether NBQX atten­ uated the increase in ischemia-induced calcium (Ca) influx by measuring neuronal Ca-calmodulin (CaM) binding using an immunohistochemical assay for free calmodulin. Though early reports concluded that non-NMDA receptor channels are imperme­ able to Ca2+ (Mayer et aI., 1987; Ascher and No­ vak, 1988), more recent studies indicate that some glutamate receptor subgroups are permeable to di­ valent cations (Murphy and Miller, 1989; Hollmann et aI., 1991). Finally, we determined the ability of NBQX to preserve motor performance. Behavioral motor performance paradigms have been used to analyze both global (Volpe et aI., 1984; Combs and D'Alecy, 1987) and focal animal stroke models (Be­ derson et aI., 1986; Young et aI., 1986; Germano et aI., 1987; Yamamoto et aI., 1988; Markgraf et aI., 1992). In the present study, a method for measuring motor performance in this model of focal pure cor­ tical ischemic injury has been developed and em­ ployed for examining functional recovery. METHODS Male spontaneously hypertensive rats (SHRs) weighing 250-300 g were fasted overnight with free access to water and anesthetized with chloral hydrate intraperitoneally (Lp.) prior to surgery. The femoral artery was cannulated with a polyethylene catheter for blood pressure monitor­ ing and serial blood gas sampling. A microprobe was used to monitor skull temperature, maintained with a heating lamp and warming blanket at 37 ± 0.5°C, as was the rectal temperature. The right common carotid artery was iso­ lated through a midline ventral cervical incision and tagged with 3.0 silk. An incision was made perpendicular to a line between the external auditory canal and the lat­ eral canthus of the right eye through the temporalis mus­ cle, which was retracted with 4.0 silk. Under direct visu­ alization with a Zeiss surgical microscope, the right MCA was exposed through a 1 x 3-mm burr hole drilled 2-3 mm rostral to the fusion of the zygomatic arch with the squamosal bone. A continuous 0.9% saline drip was

263

maintained over the drilling site to prevent thermal injury to the underlying brain tissue. Once the MCA was visible through the transparent bone layer, a jeweler's forceps was used to remove the remaining bone. A 23-gauge hy­ podermic needle was used to pierce and open the dura along the entire length of the burr hole. The method used for producing reversible ischemia was a modification of the procedure described previously (Brint et al., 1 988; Kaplan et al., 1 991). A 1 0-0, BV130-5 Ethicon Neurosurgical suture was placed under the MCA rostral to the rhinal fissure, proximal to the major bifur­ cation of the MCA, but distal to the lenticulostriate arter­ ies. The MCA was lifted -1 mm and the suture was ro­ tated clockwise so that one end pointed toward the frontal bone, forming a right angle to the long axis of the burr hold. Verification of the absence of blood flow was made visually with the surgical microscope. Immediately after MCA occlusion, the right common carotid artery was oc­ cluded using two atraumatic Heifetz aneurysm clips. Reperfusion was achieved by turning the suture counter­ clockwise and removing it. Aneurysm clips were re­ moved and flow in both the CCA and the MCA was con­ firmed visually with the operating microscope. In rats exposed to 24 h of focal ischemia, permanent ligation of the right CCA and MCA was performed. Again, absence of blood flow was verified visually. This technique pro­ duces a nearly exclusive cortical infarction, sparing the deep white matter and basal ganglia. Rats were treated with either intravenous (Lv.) N a+ salt NBQX at 5 mg/kg/h in 3 rnl of sterile water initiated immediately after MCA occlusion and infused over 6 h, versus i.v. saline controls, or NBQX, 30 mg/kg/ Lp. every 30 min x 3, begun 30 min after MCA occlusion, versus i.p. saline controls. We were unable to use a higher dose of intravenous NBQX, i.e., 15 mg/kg/h in 3 ml sterile water, because of precipitation of the drug in the vein at the i.v. site, with a concomitant mortality of -50%. To measure alterations in intracellular Ca concentra­ tions, we used an immunohistochemical assay for free CaM described previously (Picone et aI., 1 989). In previ­ ous studies in our lab (DeGraba et al., 1 993), we found that a significant increase in Ca concentration as mea­ sured by Ca-CaM binding occurred by 1 h of focal isch­ emia and was maximal by 4 h, with close correlation with histologic damage. We wanted to use durations of isch­ emia with demonstrable, but submaximal, Ca2+ influx to measure the effect of NBQX. Therefore, animals to be studied for CaM immunostaining were exposed to either 1 h of ischemia with 1 h of reperfusion or 2 h of ischemia. After being anesthetized with ethyl ether (purified grade), rats were perfused through the heart with 4% paraformaldehyde, then killed, and brain slice prepara­ tions (40-f.Lm thickness) were incubated in sheep-induced CaM antibody, which binds specifically to free CaM (i.e., CaM not bound to Ca2+ and target protein) (Picone et al., 1 989). The sections were incubated with rabbit anti-sheep peroxidase-conjugated secondary antibody, which binds and stains only CaM-antibody complex (Picone et al., 1 989). Thus, only normal cells with free CaM will stain, and ischemic cells with a Ca2+ -activated CaM-target pro­ tein complex do not stain (Picone et al., 1 989). Prior stud­ ies in our lab have shown that there are no significant changes in the concentrations of CaM or its target pro­ teins in postischemic rat brains. Therefore, changes in CaM staining are believed to be secondary to increased intracellular Ca2+ and subsequent Ca-CaM binding. Sets

J Cereb Blood Flow Metab, Vol.

14,

No.2,

1994

264

T. J. DeGRABA ET AL.

of normal and postischemic brains were processed simul­ taneously to ensure comparable staining. Staining in the ischemic region of each animal exposed to focal ischemia was graded on a scale of 0 to 15 (0 normal, 15 largest area of severe loss of CaM staining) (DeGraba et aI., 1 993). For measurement of infarct volume, animals were ex­ posed to 1 h of occlusion with 23 h of reperfusion or to 24 h of permanent occlusion. As above, the animals were randomized to two dosing regimens. The low-dose group received NBQX sodium salt, 5 mg/kg/h i.v. x 6 h, initi­ ated immediately after the onset of ischemia, versus a saline control. The other treatment group received NBQX sodium salt, 30 mg/kg i.p. in 1 ml sterile water, at 30, 60, and 90 min after MCA occlusion, versus a saline control. At 24 h, under ether anesthesia, all the rats were perfused with phosphate-buffered 1 0% formalin, and sac­ rificed. The brains were removed and examined to verify the presence or absence of MCA occlusion. After embed­ ding the whole brain in paraffin, 5-j-Lm slices were sec­ tioned at 0.5-mm intervals, stained with luxol fast bluel hematoxylin and eosin, and digitized on an Amiga com­ puter to calculate the infarct volume. CaM stain grading and histologic grading were carried out by a blinded investigator. Motor performance was measured using a combined grading scale consisting of a modification of the balance beam test and the prehensile-traction test (Combs and D'Alecy, 1 987) after either 1 h or permanent MCAICCA occlusion and in sham-operated controls. Prior to surgery, rats underwent motor performance training sessions (described below) for 2 consecutive days. The animals were scored on the morning of surgery (baseline score) and then on the mornings of postopera­ tive days 1, 2, 3, 4, 5, and 7. Two trials of each motor tasks were observed per day, with 5 min of rest between trials to prevent fatigue. The best score of the two trials was recorded. NBQX in the two drug regimens (5 mg/kg/h x 6 h i.v. started immediately after the onset of ischemia or 30 mg/kg i.p. 30, 60, and 90 min after ischemic onset) with the appropriate controls was administered. In the balance beam test, rats were placed at the center of an unfinished wooden beam (2.5-cm diameter), sup­ ported 40 cm above a foam pad (1 0 cm thick). Time to falling off the beam was recorded to a maximum of 30 s (Table 1 ). A zero was scored if the rat was unable to maintain a grip or balance on the beam. The rat received a 1 if it remained balanced up to 10 s, a 2 for 1 1 -20 s, and a 3 for 21 -30 s. The maximal possible score for this task was 3. In the prehensile-traction test, the rat's forepaws were =

=

TABLE 1. Scoring of the motor performance tests Balance beam test

o Unable to maintain balance on the beam 1 Remains on beam for up to 10 s 2 Remains on beam for 11-20 s 3 Remains on beam for 21-30 s Prehensile-traction test o Maintains a grip on the rope for 0--5 s 1 Maintains a grip on the rope for 6-10 s 2 Maintains a grip on the rope for II-IS s An additional 1 point is added for bringing one hind limb to the rope (traction task). The maximal total point score is 6.

J Cereb Blood Flow Metab, Vol.

14,

No.2,

1994

c::::::J ConlrollP Salm"

I!!!I!lNBOX IP30�x3

15 "

13 � 12 � 11 CJ 10 0) .5 c: '(ij U5 .5

9

"8

5 •

::;

E

�

e 7

3 2

1 Hour

Ropoffusion

FIG. 1. Grade of CaM staining in the infarct area (mean ± SO) in groups of rats (n = 6-9 per group) after either 1 h of occlusion with 1 h of reperfusion or 2 h of occlusion. Scores were compared between i.p. saline control and NBQX­ treated (30 mg/kg i.p. q 30 min x 3 postischemia) animals. 0 = normal, 15 = no staining in the entire cortical MCA dis­ tribution. Analysis by Kruskal-Wallis analysis of variance (ANOVA).

placed on a nylon rope positioned horizontally 70 cm above a foam pad (1 0 cm thick). The ability to maintain a grip on the rope for up to 15 s was recorded in the pre­ hensile portion of the test, indicating predominantly mo­ tor strength. The traction portion of the test evaluates the animal's ability to bring one hind limb up to the rope (Table 1 ). The rat received a 0 if the time to falling was 0-5 s, a 1 if it maintained a grip for 6-10 s, and a 2 if it maintained a grip for 1 1 -1 5 s. An additional point was scored if the rat was able to bring one hind leg to the rope within the 1 5-s period, for a possible maximal score of 3 points. The combined total of the best score for each test was recorded by an investigator blinded to the durations of ischemia. Motor performance testing was done at the same time each day to avoid diurnal variation in activity, and treatment and control groups were tested simulta­ neously to minimize potential animal care variation, which could affect performance.

RESULTS

The monitoring of physiologic variables, includ­ ing mean arterial pressure, arterial blood gases, and head temperature, revealed no significant differ­ ences between the NBQX-treated and the control animals. Though blood pressure was noted to de­ crease 20-25% during the period of ischemia, the decline was equal in both the treatment and the con­ trol groups. Mortality was 6% in controls and 4 and 6% with i.v. and i.p. NBQX therapy, respectively. After 1 h of MCAICCA occlusion, followed by 1 h of reperfusion, and 2 h of occlusion, no difference in Ca-CaM binding was found between the control and the i.p. NBQX-treated groups (mean ± SD 8.66 ± 1.53 vs. 11.62 ± 2.33 and 11.11 ± 1.54 vs. 12.13 ± 2.10, respectively) (Fig. 1). The i.v. dosing demonstrated similar results (data not shown). =

NEUROPROTECTION

Treatment with either i.v. or i.p. NBQX caused no significant reduction in infarct size compared to controls (55.1 ± 12.4 vs. 47.2 ± 18.0 vs. 47.4 ± 20.5 mm3, respectively) after 1 h of occlusion with 23 h of reperfusion or between i.v. NBQX-treated rats and controls after 24 h of occlusion (74.5 ± 22.5 vs. 80.1 ± 17.0 mm3) (Fig. 2). Finally, treatment with i.p. or i.v. NBQX did not significantly improve motor performance after 1 h of transient ischemia (Fig. 3A) or permanent MCAI CCA occlusion (Fig. 3B) compared to control ani­ mals (i.v. data not shown). In the permanently oc­ cluded group, ischemic control animals and those treated with NBQX had persistent significantly lower motor performance scores on days 2 through 7 compared to sham-operated rats. The same trend was seen in the transient MCAICCA occlusion groups, though statistical significance was not reached until days 5 and 7. Though not directly measured or quantitated, ataxic behavior and unsteadiness were noted in 83% of the i.p.-treated animals. This behavior, if seen, characteristically was noted by postoperative day 3. Since there appeared to be no correlation with mo­ tor performance (which specifically tested strength), it was felt not to be attributed to weak­ ness. Histological examination of the cerebellum and basal ganglia at the end of the week of motor performance testing failed to reveal any consistent pattern of neuronal injury.

265

pothesis that they would have a neuroprotective ef­ fect after an acute ischemic insult. However, unlike previous reports of a reduction of infarct size in rats treated with NBQX after focal ischemia (Buchan et aI., 1991) as well as a preliminary report (Pulsinelli and Cho, 1992) demonstrating neuronal protection when NBQX was administered (30 mg/kg i.p. x 3) up to 8 h after 10 min of global ischemia, we were unable to demonstrate protection of cortical neuron morphology, Ca2+ homeostasis, or motor perfor­ mance in our model of focal ischemia. NBQX was not effective in reducing infarct size after either a maximal or a submaximal ischemic insult. There are several possibilities for this lack of protection. In the paradigm of submaximal isch­ emia, neuronal injury most likely is the result of excitotoxicity mediated by both NMDA and non­ NMDA receptors. Consequently, monotherapy with an AMPA receptor antagonist may be insuffi­ cient in protecting the neuron. It has been demon­ strated that after severe ischemic injury, a decrease in extracellular pH inhibits NMDA receptor­ mediated Ca influx (Tang et aI., 1990). Thus, exci­ totoxicity mediated by the non-NMDA receptor might play a more dominant role in the severe isch­ emic injury caused by global ischemia, and this may explain the benefit reported even after a delay of 8 h before the initiation of NBQX therapy (Pulsinelli and Cho, 1992). Even with permanent MCAICCA tandem occlusion, regional cerebral blood flow in this model has been reported to persist at 19 to 29 mllOO g-l min-1 (Brint et aI., 1988), thus allowing glutamate activation of multiple receptor types to play a role in neuronal injury. This may explain why NBQX alone was not effective in our study even after permanent occlusion. To understand better the mechanism of action of NBQX, we used an immunohistochemical assay of free CaM to measure increased intracellular Ca ac­ tivity after ischemia (Picone et aI., 1989; DeGraba et aI., 1993). In this study, extensive Ca-CaM bind­ ing, which we have shown correlates with histolog­ ical damage (Picone et aI., 1989; DeGraba et aI., 1992, 1993), was not attenuated by NBQX therapy. Therefore, it would appear that even if blockade of the AMPA receptor does marginally reduce Ca2+ influx, either by antagonizing a non-NMDA recep­ tor subtype that mediates Ca2+ flux or by decreas­ ing intracellular voltage changes, thus attentuating the opening of voltage-dependent (L) Ca2+ chan­ nels (Choi, 1988), activation of other mechanisms must be sufficient to increase the intracellular Ca2+ concentration and initiate Ca2+ -mediated neuronal injury. Other reasons for the lack of neuronal protection �

DISCUSSION

The effect of AMPA antagonists on ionic flux provide a strong theoretical basis to support the hy100

c::::J Con1mI

mmJNBOxrv

90

_NBOXIP

80

�

o >

70 60 50

140

30 20 10

(N-10)

o L....... ._ .. ---J

24 Houno 0a:taI0n

1 Hour 0a:IusI0n 23 Houri �alon

FIG. 2. Volume of cortical infarction (mean ± SO mm3) in groups of rats (n = 9-16 per group) treated with NBQX, 30 mg/kg Lp., given 30, 60, and 90 min after the onset of MeA occlusion, NBQX, 5 mg/kg/h Lp. x 6 h, initiated immediately after the onset of ischemia, or saline after exposure to 1 h of occlusion and 23 h of reperfusion or to 24 h of occlusion. There was no difference in infarct volume between controls and NBQX-treated animals.

J Cereb Blood Flow Metab, Vol.

14,

No.2,

1994

T. J. DeGRABA ET AL.

266 6.5

\)

6 5.5 Q) ... 0

� ...

5 4.5

.sa

4

�

3.5

0

3

**

2.5

*

• SHAM. N.6 • Control Saline IP, N.6 • NBQX 30 mgIkg 31P. NoS x

2 Baseline * **

I)

2

3

4

5

7

Days Post Op

pc.O!i N80X YO. Shorn pc.OO5 Control ... _m

6.5 6 5.5 5

�

8

en

4.5 4

03.5 (5 3 :!: 2.5

** *

*

• SHAM.N.6 • Control Saline IP. N.S • NBQX 30 mgIkg 31P. NoS

2

1.5 Baseline * **

FIG. 3. Motor performance scores (combined score from the balance beam and prehensile· traction tests) in rats treated with NBQX, 30 mg/ kg i.p., given 30, 60, and 90 min after MeA oc­ clusion, were not different from those of saline­ treated control animals after 1 h of MeA occlusion (A) and permanent MeA occlusion (8) (group mean ± SO for each trial). A persistent significant motor score deficit was seen in the NBQX-treated and control ischemic animals compared to shams from days 2 through 7 after permanent occlusion (B). A significant differ­ ence occurred by days 5 and 7 in the ischemic control versus the sham groups (A) after 1 h of occlusion. n = 6 per group. A nalysis by Kruskal-Wallis ANOVA.

x

*

2

3

4

5

pc.1I2 N80X IlL Shorn pc.1I2 Control IlL Shorn

in our study include variability in the tissue distri­ bution of NBQX. After sacrificing our rats receiv­ ing i.p. NBQX, examination of the abdominal cav­ ity revealed crystallization of the drug on the peri­ toneal wall. In addition, with both i.v. and i.p. administration of the drug, there was marked yel­ low discoloration of the rat's tail and coat as well as the visceral organs and urine. We also observed that the sodium salt of NBQX at higher concentra­ tions precipitated at the venous infusion site. Such peripheral tissue and organ deposition would lead to interanimal variability in blood levels. However, the doses, vehicle, and method of administration used in this study were the same as reported in previously published positive experiments (Buchan et al., 1991; Pulsinelli and Cho, 1992). Furthermore, i.p. administration of the same doses of the sodium salt of NBQX to male Wistar rats resulted in pre­ dictable peak plasma concentrations of 5-10 Ilg/ml (data on file, Novo Nordisk Corp.). Finally, treated animals exhibited ataxic behavior indicating some J Cereb Blood Flow Metab. Vol.

7

Days Post Op

14.

No. 2,

1994

central nervous system penetration. Therefore, while unfavorable pharmacokinetics may be an im­ portant contributor to our negative results, they are not the entire explanation. Another possible explanation for the lack of ben­ efit from NBQX is the small "penumbra" seen in this model, which is believed to be the zone of isch­ emic insult in which neuronal protection is most likely to occur with this type of therapy. However, Kaplan et al. (1991) demonstrated little to no in­ farcted tissue at ischemic periods of < 1 h and in­ farct volumes of -to, 100, and 150 mm3 (maximum) after 1, 2, and 3 h of ischemia, respectively, in this model. Since in the present study ischemic dura­ tions were 1, 2, and 24 h, with drug therapy initiated within 30 min of the onset of ischemia, there is a reasonable likelihood that the drug was being given during a submaximal insult. If the drug has neuro­ protective effects, it is not unreasonable to expect it to show some benefit in this setting. Furthermore, we have found a reduction of infarct volume with a

NEUROPROTECTION

competitive NMDA receptor antagonist and with hypothermia in this model (unpublished data). We have found that neuroprotective therapy may improve working memory as well as electrophysio­ logic parameters without detectable histologic pro­ tection (Grotta et aI., 1988, 1990). Behavioral as­ sessment is still poorly defined in this model of focal ischemia. Since the ultimate goal in developing a therapy for acute stroke in humans is to improve overall functional outcome, an analysis of neuro­ logic function must eventually be included in the preclinical evaluation of a therapy's effectiveness. We used a modification of the balance beam and prehensile-traction tests to measure functional out­ come. Though no improvement in motor perfor­ mance was seen with the administration of NBQX versus controls, our method appears to be reliable for future preclinical studies in the assessment of neuroprotective therapy utilizing a relatively small number of rats (six per group). Analysis of the scores demonstrated a persistent significant motor performance deficit in the permanent MCAICCA occlusion rats on postoperative days 2 through 7 compared to sham-operated animals. A significant though less dramatic decrease in motor perfor­ mance was also seen in the I-h occlusion rats versus shams on days 5 and 7. Though performance scores were noted to be somewhat higher in the I-h versus permanent occlusion rats, a larger number of ani­ mals per group (18-23) would be required to guar­ antee recognition of a 25% difference in motor per­ formance between these two degrees of ischemia with a certainty of 0.90. Further analysis of the individual tasks revealed that the balance beam test was less sensitive for detecting mild deficits occurring after 1 h of tran­ sient occlusion. In addition, it was noted that the rats were less likely to perform the motor behavior traction task with successive daily trials, despite their ability to maintain a grip on the rope for greater than 15 s. This "fatigue" in performance of the traction task was seen not only in the ischemic and control groups but also in the sham animals and may represent a loss of motivation to complete the task. Therefore, unless some incentive can be added to ensure that the task is a function of capa­ bility rather than interest, the traction component may not be reliable for sequential testing. In an effort to quantify better the neurologic def­ icit after a MCA infarction, Bederson et al. (1986) described circling behavior after severe proximal MCA occlusion. In our model of exclusive cortical infarction and no subcortical involvement, none of the 12 rats with permanent MCAICCA occlusion exhibited circling behavior. Therefore, even with a

267

maximal-MCA distribution cortical infarction, cir­ cling behavior appears to require additional in­ volvement of the basal ganglia and is not reliable as a measure of functional injury in this model. This highlights the fact that similar models of ischemia may affect different brain regions, resulting in ab­ errations of different behaviors. Our findings accent the need for careful standardization of neurologic functional testing for each ischemia model. In conclusion, NBQX as a single agent showed no significant protective effect on histologic neuro­ nal injury or motor performance in this model of focal ischemia, possibly related to its inability to prevent Ca2+ influx. However, uncertainty about drug absorption and central nervous system bio­ availability preclude a definite conclusion about drug activity. An assessment of motor performance is a valid and useful method for measuring outcome in this model. Acknowledgment: This work was supported by NIH Grant RO-l NS 23979 (J.G.). NBQX was supplied by Novo Nordisk Corp., Denmark.

REFERENCES Ascher P, Nowak L (1988) Quisqualate- and kainate-activated channels in mouse central neurones in culture. J Physiol (London) 399:227-245 Bederson JB, Pitts LH, Tsuji M, Nishimura MC: Davis RL, Bartowski H (1986) Rat middle cerebral artery occlusion: evaluation of the model and development of a neurologic examination. Stroke 17:472-476 Brint S, lacewicz M, Kiessling M, Tanabe J, Pulsinelli W (1988) Focal brain ischemia in the rat: method 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 Buchan AM, Xue D, Huang Z, Smith KH, Lesiuk H (1991) De­ layed AMPA receptor blockade reduces cerebral infarction induced by focal ischemia. Neuro Rep 2:473-476 Buchan AM, Li H, Cho C, et al. (1992) Blockade of the AMPA receptor prevents CAl hippocampal injury following severe but transient forebrain ischemia in the adult rat. Neurosci Lett 132:255-258 Choi DW (1988) Calcium-mediated neurotoxicity: relationship to specific channel types and role in ischemic damage. Trends Neurosci 11:465-469 Combs DJ, D'Alecy LG (1987) Motor performance in rats ex­ posed to severe forebrain ischemia: effect of fasting and 1,3butanediol. Stroke 18:503-511 Cosgrove J, Newell TG (1991) Recovery of neuropsychological functions during reduction in use of phencyclidine. J Clin PsychoI47:159-169 DeGraba TJ, Ostrow PT, Strong RA, Earls RM, Zong ZJ, Grotta JC (1992) Temporal relation of calcium-calmodulin binding and neuronal damage after global ischemia in rats. Stroke 23:876-882 DeGraba TJ, Ostrow PT, Grotta JC (1993) Threshold of calcium disturbances after focal cerebral ischemia in rats: implica­ tions on the window of therapeutic opportunity. Stroke 24: 1212-1216 Germano IM, Bartkowski HM, Cassel ME, et al. (1987) The therapeutic value of nimodipine in experimental focal cere­ bral ischemia. J Neurosurg 67:81-87 Grotta JC, Pettigrew LC, Rosenbaum D, Reid C, Rhoades H,

J Cereb Blood Flow Metab, Vol.

14,

No.2,

1994

T. 1. DeGRABA ET AL.

268

McCandless D (1988) Efficacy and mechanism of action of a calcium channel blocker after global cerebral ischemia in rats. Stroke 19: 447-454 Grotta JC, Picone CM, Ostrow PT, Strong RA, Earls RM, Yao LP, Rhoades HM, Dedman JR (1990) CGS-19755, a compet­ itive NMDA receptor antagonist, reduces ca1cium­ calmodulin binding and improves outcome after global cere­ bral ischemia. Ann Neurol 27: 612-619 Hollmann M, Hartley M, Heinemann S (1991) Ca + + permeabil­ ity of KA-AMPA gated glutamate receptor channels de­ pends on subunit composition. Science 252:851-853 Kaplan B, Brint S, Tanabe J, Jacewicz M, Wang XL, Pulsinelli W (1991) Temporal threshold for neocortical infarction in rats subjected to reversible focal cerebral ischemia. Stroke 22: 1032-1039 Markgraf CG, Green EJ, Hurwitz BE, et al. (1992) Sensorimotor and cognitive consequence of middle cerebral artery occlu­ sion in rats. Brain Res 575:238-246 Mayer ML, MacDermott AB, Westbrook PB, Smith SJ, Barker JL (1987) Agonist- and voltage-gated calcium entry in cul­ tured mouse spinal cord neurons under voltage clamp mea­ sured using arsenazo III. J Neurosci 7: 3230-3244 Murphy SN, Miller RJ (1989) Two distinct quisqualate receptors regulate Ca + + homeostasis in hippocampal neurons in vitro. Mol Pharmacol 35:671-680 Nellgard B, Wieloch T (1992) Postischemic blockade of AMPA but not NMDA receptors mitigates neuronal damage in the rat brain following transient cerebral ischemia. J Cereb Blood Flow Metab 12:2-11

J Cereb Blood Flow Metab. Vol.

14.

No.2.

1994

Picone CM, Grotta JC, Earls R, Strong R, Dedman J (1989) Immunohistochemical determination of calcium-calmodulin binding predicts neuronal damage after global ischemia. J Cereb Blood Flow Metab 9: 805-811 Pulsinelli W, Cho S (1992) Blockade of the AMPA receptor be­ ginning 8 hours after transient forebrain ischemia attenuates CA I hippocampal injury in rats. Neurology 42(4) (Suppl 3):292 Sheardown MJ, Nielsen EO, Hansen AJ, Jacobsen P, Honore T (1990) 2,3-Dihydroxy-6-nitro-7-sulfamoyl-benzo(F)quinoxa­ line: a neuroprotectant for cerebral ischemia. Science 247: 571-574 Tang CM, Dichter M, Morad M (1990) Modulation of the N-methyl-D-aspartate channel by extracellular H +. Proc Natl Acad Sci USA 87: 6445-6449 Volpe BT, Pulsinelli W A, Tribuna J, Davis HP (1984) Behavioral performance of rats following transient forebrain ischemia. Stroke 15: 558-562 Willetts J, Balster RL, Leander JD (1990) The behavioral phar­ macology of NMDA receptor antagonists. Trends Pharma­ cal Sem 11: 423-428 Yamamoto M, Tamura A, Kirino T, et al. (1988) Behavioural changes after focal cerebral ischemia by left middle cerebral artery occlusion in rats. Brain Res 452:323-328 Young RS, Kolonich J, Woods CL, et al. (1986) Behavioral per­ formance of rats following neonatal hypoxia-ischemia. Stroke 17: 1313-1316