Buprenex

Previous nonhuman primate stroke models have employed temporary occlusion of arteries, had limited behavioral testing and imaging, and focused on the short-term outcome. Our goals were 1. to develop a stable model of chronic stroke in the nonhuman primate, 2. to study in vivo the long-term biochemical changes in the area adjacent to the infarct, using proton magnetic resonance spectroscopy (^sup 1^ H MRS), and 3. evaluate these changes in relation to the histopathological effects of stroke. Four adult cynomologous monkeys had an occlusion of the M1 segment of the right MCA. Behavioral tests included a clinical rating scale, motor planning task, fine motor task, and activity monitoring. Eight months afterwards, MRI and ^sup 1^H MRS were performed. Following the imaging studies the monkeys were perfused transcardially, their brains extracted and processed. Nissl staining and immunohistochemistry for neuronal markers (NeuN) were performed and used to measure the lesion volume and neuronal optical density (OD). All animals developed a left hemiparesis and were unable to perform a fine motor task with the left hand. There was a significant (31%) decline in the motor planning ability with the nonparetic extremity. Monkeys displayed a stooped posture, episodes of rotation to the side of the lesion, partial left hemianopsia, and transient changes in activity. The clinical signs improved over the first 6-8 weeks but the deficits remained stable for the remaining six months of follow up. MRI demonstrated a subcortical and cortical infarction in the right MCA distribution. ^sup 1^H MRS data detected a significant decrease in the N-acetyl-aspartate (NAA) /creatine (Cr) ratio in the area adjacent to the infarction (VOI-St) compared to a mirror area in the contralateral hemisphere (VOI-Co). Histopathological measurements revealed a significant decline in neuronal cross-sectional area and neuronal optical density in the region of the VOI-St. We established a stable and reproducible model of chronic stroke in the MCA distribution, in the macaque monkey. Our data indicate that NAA detected by ^sup 1^H MRS can be used to measure neuronal loss in vivo and help target this area for intervention. Our model may be particularly suitable for studies testing the effects of therapeutic strategies involving neural or stem cell transplantation, trophic factors or gene therapy. [Neurol Res 2003; 25: 68-78]

Keywords: Stroke; middle cerebral artery; behavior; nonhuman primates; MR spectroscopy

INTRODUCTION

Stroke is the third leading cause of death and the leading cause of disability in the USA1,2. Early thrombolysis and attempts to maximize perfusion of the ischemic penumbra are the current therapeutic options. After the acute period, a stable neurological deficit remains. Although gradual improvement overtime, especially with physical and occupational therapy, is common, there are no specific treatments that can help reverse the deficits.

Several animal models of ischemic stroke have been developed. While rodent models of stroke are frequently used3,4, primate studies are fewer in number. Primate stroke models employed to date have been geared toward the study of the effects of acute stroke and the results of reperfusion or other interventions5-9. Many authors used temporary occlusion of the MCA10-12. The temporary occlusion models serve primarily to study the dynamics of the development of stroke. They were used to determine the optimal time for revascularization and to try various strategies aimed at preserving neuronal function. The primary outcome measure was infarct size7,11,12. Frazee et al.11 also used a neurological examination score.

With the advent of new therapeutic strategies for chronic stroke, a long term and more sensitive model becomes necessary because the new approaches may have subtle but clinically significant effects. Therefore, we applied a behavioral battery that addresses specific aspects of fine motor function, a cognitive test (motor planning ability), activity level, and a standardized neurological examination, to monkeys who had permanent surgical occlusion of the middle cerebral artery (MCA). We also looked for methods that may allow us to identify neuronal loss in the area adjacent to the necrotic zone of the cerebral infarction, in vivo. This zone, which generally corresponds to the ischemic penumbra in acute stroke, may be a target for transplantation or regeneration efforts. A method, which identifies neuronal activity in vivo may also be used for follow up and evaluation of neural transplantation or regeneration efforts. Proton magnetic resonance spectroscopy (^sup 1^H MRS) is a method that provides noninvasive in vivo measurements of brain metabolites. In ^sup 1^H MRS studies, a volume of interest (VOI) within the brain is defined and the spectrum from this volume is collected. Among the metabolites that can be detected are the neuronal marker N-acetyl aspartate (NAA)12, myo-inositol (MI)14 which has been shown in vitro to be present in glial, but not in neuronal, cell cultures, choline (Cho)15 a constituent of the phospholipid metabolism of cell membranes, lactate (La)15 an indicator of carbohydrate catabolism and creatinine (Cr)15 which has been related to metabolic states. Because the Cr peak remains fairly stable even during disease, it is routinely used as an internal control value. ^sup 1^H MRS has previously been used to assess stroke areas during the initial days following an ischemic episode. Changes in the concentration of brain metabolites in the area of stroke have been described in rodents16-20, monkeys21,22 and humans14,23-25. These studies reported decreases in NAA/Cr associated with increases in La/Cr, Ml/Cr and Cho/Cr as early signs of ischemic injury. In contrast, the chronic effects of ischemia in the area adjacent to the necrotic center of the stroke have not been studied with ^sup 1^H MRS. To the best of our knowledge we are the first to use ^sup 1^H MRS to analyze in vivo the chronic effects of stroke in the penumbra.

We chose to occlude the MCA because stroke in its distribution is a common and devastating condition this artery is the most commonly affected by embolic events, and infarction in its territory is common after occlusion of the internal carotid artery1. Different methods to induce a complete or temporary occlusion of the MCA have been used26. The endovascular methods are effective and minimally invasive. However, open surgery allows a direct identification of the arterial segment to be occluded and may improve reproducibility of the infarction. For example, good reproducibility was seen in open surgical occlusion of the MCA in marmoset monkeys27.

Our goal was to model chronic stroke, which would include cortical and subcortical structures, but will be nonfatal, follow the animals for eight months behaviorally and perform correlative imaging (MRI and ^sup 1^H MRS) and anatomical studies. Such a model may allow the study of the behavioral and anatomical effects of neural repair and introduce the use of ^sup 1^H MRS to select and monitor the best region for reparative treatments in chronic stroke.

MATERIALS AND METHODS

Subjects

Nine adult cynomologous macaques (10-13 years old, 6-8 kg) were singly housed in quarters with a 12-h light/ dark cycle. Four received unilateral strokes and five were age-matched controls. All animals received water ad libitum and were fed once a day (after the behavioral testing) with monkey chow, supplemented with fruit and seeds. The study was performed in accordance with federal guidelines of proper animal care and with the approval of IACUC.

MCA occlusion

The animals were fasted the evening prior to surgery. Monkeys were intubated and placed on isoflurane (1%-2%) inhalation anesthesia. Cardiac rhythm, arterial O2 saturation, respiration and rectal temperature were monitored. The monkeys were placed on the surgical table with their heads turned to the left side. Under sterile conditions, a right pterional craniectomy was performed. The dura was opened and the brain gently retracted with a cottonoid strip to expose the sylvian fissure. The right MCA was identified, coagulated in the middle of the M1 segment, and cut with microscissors. The operative field was then washed with saline, and the incision was closed. Post-operatively, the animals were placed in the recovery room on a heating pad, where they were closely monitored for the following 48 h. Intravenous fluids (0.9% NaCl with 5% glucose to avoid hypoglycemia) were given, and analgesics (Buprenex) administered. Cefazolin (25mgkg^sup -1^ i.m.) before induction of anesthesia and then twice a day for 24 h was also given. Post-operative assessment included recorded observations of the surgical site, feeding, feces, urine, drooling, vomiting, activity, epileptic seizures, pupil response, motor coordination, balance, paresis and locomotion.

Clinical rating

A stroke clinical rating scale (SCR scale, Table 1) was used once per week before and after MCA occlusion to quantify the results of a clinical neurological examination. The rating scale was inspired by a measure previously used in a primate model of Parkinsonism28. All the ratings were performed by the same trained observer. Blinding of the observer to operative status was not possible because the deficits were obvious. Bradykinesia and circling are self explanatory. Defense reaction is a complex response which included sharp movement away from a threat, lifting the head and opening of the mouth enough to show the canines. Gait was evaluated in the animals' home cage. The stimulus for the visual field evaluation was a piece of fruit presented from each side, with the examiner directly facing the animal in its home cage. Normal response to this stimulus is turning the gaze and the head toward it, and usually reaching for the food. Blinking reflex is elicited by rapid approach of a stimulus from the periphery toward the eye. All the assessments were done in the home cage. From a total of 41 points, 0 corresponded to normal behavior and 41 to severe bilateral neurological impairment.

Activity monitoring

Each monkey was sedated with ketamine (15 mg kg^sup -1^, i.m.) and fitted with a primate vest that contained a PAM2 activity monitor (IM Systems, Baltimore, MD, USA28) in the inside back pocket. These monitors measure acceleration. Every time the monitor senses acceleration that exceeds a threshold of 0.1 G, an electrical pulse is generated and recorded. The number of pulses is expressed for a pre-selected time period (1 min). After one week, the animals were sedated again, the jacket was removed, the activity monitor interfaced with a Macintosh computer and the data downloaded. The data are expressed as a mean of each 12-h light/dark cycle.

Fine motor test

Each monkey was tested for fine motor performance of the upper limbs using a modification of a previously described food pick-up task28,29. Testing occurred in a modified home cage. The animals were presented with a 3x3 matrix of recessed food wells embedded in a Plexiglas board. During trials, six cubes of apple (0.5 cm) were placed within the same six food wells and the time to retrieve all six pieces was recorded. The test board was configured with a transparent Plexiglas wall to limit access only to the arm being evaluated. Monkeys received 10 trials per arm in each session. The arm being tested alternated for each trial. The test provided information about overall strength of the extremity, fine motor function and hemineglect or visual field cut. The same investigator tested each animal at the same time of the day three days a week throughout the course of the study.

ORDT

This test evaluates cognitive deficits in tasks involving procedural strategies and the programming of reaching movement30. Briefly, a Plexiglas box (8x8x9 cm) was fixed on a tray attached to a modified home cage. The box has one open side, either facing the animal or turned right or left by 90 degrees. The reward was a piece of fruit (slice of apple or banana). The animal could not see or reach the box until an opaque barrier was raised at the beginning of each trial. The difficulty level was adjusted by changing the orientation of the open part of the box and its position relative to the animal, and the location of the reward in the box. The positions of the rewards were designed to be reached with the right hand, so that left hemiparesis would not preclude performance on this task. Each test session consisted of 15 trials with the direction of the open side presented randomly to the animals. Each test series consisted of four sessions. Responses were recorded on videotape. The results were scored as 'success' if the animal retrieved the reward on the first attempt and 'correct' if it required additional attempts but retrieved the fruit within 40 sec. Barrier hits were scored if the animal tried to reach the reward through the transparent wall of the box. A 'motor problem' was scored if the monkey reached the correct (open) side of the box but failed to retrieve the reward. A series of four sessions was performed monthly after MCA occlusion.

MRI scanning

MRI scans were performed The animals were anesthetized with Telazol (10-15 mg kg^sup -1^, i.m.), supplemented with atropine 0.04 mg kg^sup -1^ i.m. for transportation and scanning. Vital signs were monitored throughout the procedure. We used an MRI compatible stereotactic frame. MRI studies were performed after MCA occlusion to delineate the extent of the lesions.

MRI and ^sup 1^H MRS procedures

Scans were performed eight months after the MCA occlusions. The night before the procedures, each monkey's food was withheld as a precaution for anesthesia. The monkeys were anesthetized with Telazol (2-6mgkg^sup -1^, i.m.), supplemented with atropine (0.04 mg kg^sup -1^, i.m.) for transportation and scanning. Vital signs were monitored throughout the procedure. Anesthesia was adjusted to maintain immobility during the scan. The scanning was performed in a 1.5 Tesla Signa Unit GE (General Electric Medical Systems Inc., Milwaukee, WI, USA). The animals were placed in a ^sup 1^H MRS-compatible stereotaxic frame. The coronal zero was identified by the location of ear bars filled with vegetable oil. T1- and T2-weighted images were obtained. The extent of the lesion and the volume of interest (VOI) for the ^sup 1^H MRS were determined from high-resolution fast spin echo T2-weighted images. The images were obtained in axial and coronal planes. The relevant acquisition parameters were: TR, 3,800 msec; field of view, 12 cm; slice thickness, 3 mm, eight averages, total acquisition time 9 min, 45 sec. The VOI was placed in the area medial to the stroke in the right side of the brain (VOI-St) and in a mirror position on the left hemisphere (VOI-Co; Figure 2). A double spin echo sequence (PRESS) was employed to obtain the brain metabolite spectra from within the VOI. Water suppression was performed by means of two chemical shift-selective radio-frequency pulses, each followed by a spoiling gradient. Relevant spectroscopic acquisition parameters were TE, 35 msec; repetition time, 1,600 msec; VOI, 1.0x1.0x1.0 cm; number of averages, 384. The duration of each of the spectroscopic scans was 9 min. The entire MRI scan, including anatomical scans for the location of the VOI, had a duration of approximately 1 h. All resonances were integrated using software provided by GE. The metabolites were identified by their chemical shift observed in ^sup 1^H MRS; ppm value: Cho, 3.2; Cr, 3.03; Ml, 3.56; NAA, 2.0. Their peak values were determined in arbitrary units. Values of the metabolites were expressed as ratios to an internal standard so that intra-individual comparisons could be made. We employed Cr as the reference value, as it is routinely done in ^sup 1^H MRS studies. The metabolite ratios examined were NAA/Cr, Cho/Cr, and Ml/Cr.

Necropsy, preparation of tissue

Following the imaging studies (eight months post-infarct),, monkeys were anesthetized with pentobarbital (25 mg kg^sup -1^, i.v.) and perfused transcardially with normal saline (200 ml) followed by 4% Zamboni's fixative (400 ml). The brains were slabbed in the coronal plane using a calibrated Lucite brain slice apparatus. The slabs were then immersed in a 4% Zamboni's fixative for 48 h then cryoprotected by immersion in a graded (10%-40%) sucrose/0.1 M phosphate buffered saline (PBS, pH 7.2) solutions. The tissue slabs were cut frozen (40 [mu]m) on a sliding knife microtome. All the sections were stored in a cryoprotectant solution before processing.

Immunohistochemistry

Sections were processed for immunohistochemical staining of NeuN according to our previously published procedures28. Endogenous peroxidase activity was removed with a 20-min incubation in 0.1 M sodium periodate. After 3x10 min washes in PBS plus 0.05% Triton-X (dilution media), background staining was blocked with a 1 h incubation in a Tris buffered saline solution containing 3% normal horse serum, 2% bovine serum albumin, and 0.05% Triton X-100. The sections were then incubated with a monoclonal NeuN (1-1000) (Chemicon Inc., Temecula, CA, USA) primary antibody for 48 h at room temperature. Sections were then incubated for 1 h in horse antimouse biotinylated secondary antibodies (1:100; Vector Laboratories, Burlingame, CA, USA). After 12x10 min washes in dilution media, the sections were placed in the avidin biotin (ABC, 'Elite' kit, Vector Laboratories) substrate (1 : 1,000) for 75 min. Sections were then washed in a 0.1 M imidazole/1.0 M acetate buffer, pH 7.4, and then reacted in a chromagen solution containing 0.05% 3,3'-diaminobenzidine, and 0.05% H^sub 2^O^sub 2^. Controls consisted of processing tissue in an identical manner except for by using the primary antibody solvent or an irrelevant immunoglobulin G (IgG) in lieu of the primary antibody. Sections were mounted on gelatin-coated slides, dehydrated, and coverslipped with Permount. Additional sections were stained for Nissl using cresyl violet acetate.

Image analysis

Lesion volume was measured in every fifth coronal section between the start of the striatum to just past the substantia nigra. Sections were viewed using a COHU Model 2122 solid state black and white microscope video camera mounted on an Olympus BHS microscope with 2X lens. NIH Scion Image version 1.6 was used to measure the area of the parenchyma of the normal hemisphere and the stroked hemisphere of the same section. The difference between the two areas represents the area of infarcted surface for that section. Infarct volume was calculated by integration of area over the number of slices that encompassed the portion of the lesion being evaluated.

Quantitative optical density (OD) measurements were performed on individual NeuN positive cells by using a modification of a previously described procedure31,32. For each case, five equally spaced sections through the right and left thalamus were sampled and evaluated. Sections were matched for level across cases. The thalamus was chosen because the VOI included most of its volume and its borders were easy to identify. The thalamus was measured in coronal sections from the rostral pole, behind the plane of the foramen magnum and the genu of the internal capsule to the sections just caudal to the putamen. Measurements were performed at a total of 20x magnification using an Olympus microscope coupled to a computer-assisted morphometry system (Image 1200 NIH). A visible nucleus was used as a criterion for cell body. The OD measurements were performed by using a scale, such that 0 represented a white image and 256 represented a black image. NeuN positive neurons were identified, and the perikarya were manually outlined. The OD was then automatically measured by using the NIH image software. To account for differences in background staining intensity, five background OD measurements in each section were taken from each microscopic field that lacked NeuN profiles and the mean of these five measurements constituted the background OD. The background OD was then subtracted from the OD of each individual neuron to provide a final value. Once all the NeuN positive cells were evaluated in a field, the stage was manually moved to a new field by using fiduciary landmarks to ensure a completely nonredundant evaluation.

Data analysis

The fine motor task and the clinical rating scores were analyzed using a Wilcoxon Signed Ranks test. Results are expressed as mean + or - SE. p

RESULTS

Clinical rating

Prior to MCA occlusion all the animals displayed normal behavior appropriate to their age, corresponding to score of 0 on the rating scale. No variance was seen.

The clinical examination score was 22 + or - 3.27 at the first post-operative evaluation. Salient features exhibited by the monkeys receiving strokes initially included spastic left hemiparesis, left facial weakness, left visual field deficit, impaired gait, impaired balance, transient decrease in defense reaction and decreased responsiveness to their environment. Bradykinesia was seen in two animals, and circling behavior was transiently observed in two of the animals. The average score declined significantly over the first two months (to 15.75 + or - 1.89, p = 0.012). Gait, balance and facial weakness improved, while the rest of the signs remained stable (Figure 1).

Following the initial improvement the score remained unchanged for the duration of the experiment, indicating a stable long term pattern of neurological deficits after an initial partial neurological recovery. In contrast, control animals scored O on this rating scale for the duration of the experiment.

Activity

A marked change in the individual activity level was observed in all the experimental animals. Two of the animals had an overall decrease in activity. One animal developed marked hyperactivity that peaked five months after MCA occlusion. One monkey had an initial decrease in activity followed by an increase to levels above baseline after five months. Since the individual variability was high, and the changes were in different directions, there was no significant difference between the average activity scores of the animals as a group before and after the stroke (p = 0.739). Although circling behavior was observed in two animals and was seen as continuous movement, we did not find a direct correlation between circling behavior and overall activity levels. Control animals did not exhibit abnormal behavior for the duration of the experiment, and there was no significant difference between the average activity of controls and that of monkeys after MCA occlusion (p = 0.564) (Figure 2A,B).

ORDT

The scores for four monkeys with stroke were compared with five age-matched controls. The success score of the animals with stroke was 47.77 + or - 12.88 (mean + or - SD) and that of the normal controls was 68.67 + or - 9.62 (p = 0.032, Mann-Whitney). The decline was entirely due to an increase in barrier hits, and not motor problems (52.23% versus 31.33% barrier hits in animals with stroke versus normal controls (Figure 3).

Fine motor task

Severe left hemiparesis resulted in a complete inability to perform the fine motor task with the affected hand. Post-operatively, the animals still completed the task using the normal limb without errors of repetition or omission of food wells. However, performance on the right (intact) side tended to be slower after the procedure, but the difference did not reach statistical significance. Experimental animals took 14.24 + or - 1.08 (SE) sec to complete the task, with the right hand, post-operatively versus 11.7 + or - 1.29 pre-operatively, (p = 0.068, Wilcoxon signed ranks test). Control animals took 10.25 + or - 1.26 sec to complete the task with the right hand (p = 0.063, Mann-Whitney test versus post-operatively) (Figure 4).

MRI and ^sup 1^H MRS findings

Figure 5A shows a T2-weighted MRI with the VOI areas where the ^sup 1^H MRS data were obtained. The VOI areas were placed adjacent to, but exclusive of, the areas of hyperintensity on T2-weighted images. The lesion comprised part of the cortex and underlying white matter on the lateral side of the hemisphere, the main part of the striatum, the globus pallidus and the internal capsule. Neighboring structures had normal signal intensity, although some atrophy could be seen in the ipsilateral thalamus and cortex. This area corresponded to previous descriptions of territories affected by MCA-M1 occlusions8. Figure 6 shows a typical spectrum obtained from the intact (VOI-Co) and stroke hemisphere (VOI-St) of the same animal. Eight months after the complete occlusion of the right MCA, ^sup 1^H MRS analysis revealed a 46% significant decrease (paired t-test, p

Histopathology

Eight months post-MCA occlusion, the animals presented cavitation in the area corresponding to the ischemie stroke that comprised cortex and regions of the basal ganglia, as described by other authors12. Nissl staining revealed loss of tissue and debris surrounding the lost areas, as well as the presence of small, shrunken round cells. In general, the affected hemisphere exhibited three zones: 1. a central area of necrosis with liquefaction and cavitation; 2. around the zone of necrosis, an area which had, by and large, preserved architecture but on microscopic examination of NeuN stained sections displayed a marked paucity of neurons; and 3. a normal appearing area (out of the MCA irrigation zone). The damaged areas closely corresponded to regions of hyperintensity and loss of tissue in the T2-weighted MRI scans (Figure 5). The mean volume of the lesion as seen on the anatomical examination was relatively uniform at 55.2% + or - 10.8% of the zone between the tip of the striatum and the end of the substantia nigra caudally.

The area selected for ^sup 1^H MRS on the ischemic hemisphere (VOI-St) that mainly corresponded to the thalamus showed a decline in NeuN positive staining (Figure 7). The remaining NeuN positive cells in that region presented an atrophic morphology with small rounded bodies and stunted or missing neurites (Figure 7C). In comparison, the neurons corresponding to the VOI-Co of the intact hemisphere appeared healthy with triangular perykarya and multipolar varicose neurites emanating from the cell body (Figure 7B).

Analysis of the thalamus area revealed a significantly smaller size on the stroke side compared to the intact side (p

DISCUSSION

We aimed to establish a stable chronic stroke model in nonhuman primates, suited for the study of neural transplantation and neuroprotection strategies that will be relevant to the patients most likely to be the candidates for such treatments. Therefore we set out to model the human condition of an established stroke when early interventions have failed and there is a persistent deficit.

The animals exhibited a pattern of acute impairment followed by gradual improvement of neurological signs over two months. Chronic deficits remained, exhibited as stable clinical examination scores for the remainder of the eight months of observation. The natural long-term history of stroke in monkeys appears to mimic the course of the disease described in human patients1,2.

This model may allow us to follow the late effects of relatively early interventions and also to study the outcome of treatments targeted specifically at improving the condition of patients with established stable deficits.

One of our goals was the creation of a large infarction that has subcortical and cortical components. This type of lesion carries a high risk of mortality in the acute phase. Human patients frequently survive the acute phase because they are treated in an intensive care setting, but are left with marked deficits. We reproduced an intensive care setting with close monitoring and treatment of the animals.

Another goal was to increase the sensitivity and objectivity of the outcome measures. Therefore, our model has three components:

1. Behavioral: In addition to the clinical examination, we included several quantifiable neurobehavioral measures activity measurement with an electronic accelerometer, a fine motor task and a cognitive (motor planning) test

2. Imaging: MRI was combined with ^sup 1^H MRS evaluation of neuronal population of the area adjacent to the liquefied necrotic stroke.

3. We correlated our behavioral and imaging findings with qualitative and semi quantitative histopathological examination.

Behavioral tests

The ability to perform a motor planning task was significantly impaired in the experimental animals compared with intact age-matched controls. The specific deficit is considered to be a lack of ability to control the impulse to reach directly to the reward29,30. The impairment is even more marked than the numbers (31% decline) suggest - more than one third of the positions in the ORDT test involve no detour, and are easily performed by all monkeys. Therefore the lowest score a monkey with no 'impulse control' or no ability to perform a detour should be close to 40%. Thus, a score of 47% success, as seen in the experimental animals, would indicate almost no ability to perform a detour of a transparent barrier.

Several factors could have caused this decline. Nonspecific effects of the stress of surgery may have contributed to the low scores. However, the animals were tested three months after the operation, and behaved in a normal and calm fashion otherwise. They did not appear distressed by the behavioral testing procedure, a familiar environment for these animals (they did not exhibit any of the responses usually associated with fear or aggression). Moreover, they appeared to have no problem with the easy reaches, but only a specific problem with the detour action, resulting in barrier hits instead of a successful detour. It is possible that unilateral injury to the striatum, as seen in our animals, is enough to result in a marked impairment of motor planning, despite an apparently intact striatum on the side contralateral to the normal extremity.

The fine motor test demonstrated complete loss of fine motor function of the contralateral hand. With the ipsilateral hand we observed a tendency toward slower performance. This finding did not reach statistical significance, and even if the tendency is real, it may not be specific. Abnormal posture, slower motor planning, and bradykinesia in some of the animals may have contributed to a decreased speed of performance.

Other deficits that occur with MCA distribution stroke in human patients are visual field deficits and visual tracking problems. Similar deficits may also have contributed to slower performance on this task in our animals, although visual field deficits are difficult to prove in monkeys.

The behavioral testing in our study has some limitations. Macaque monkeys are a limited resource, therefore we did not perform sham surgical control. Therefore a 'surgical confound' may have been introduced. We believe, however, that a long term follow up minimized the acute effect of surgery on the behavior. The results of our behavioral tests demonstrate several severe deficits, well known to be caused by stroke. MRI and pathological studies confirmed the presence of a large MCA distribution stroke in all of the experimental animals.

Imaging and histopathology

In this project we introduced two innovations related to imaging of stroke in nonhuman primates: 1. The use of ^sup 1^H MRS to identify and monitor the area of partial neuronal loss (presumably corresponding to the ischemic penumbra) and 2. The study of the chronic effects of stroke in this area. We found that eight months following a MCA occlusion there was a significant decline in the neuronal marker NAA in the area adjacent to an infarct compared with the intact contralateral side. This effect was specific for NAA as myo-inositol levels were unchanged. These results are supported by the finding that NeuN positive neurons were lost and remaining neurons were atrophic in the affected region.

The high resolution T2-weighted images eight months following MCA occlusion revealed hyperintensity in areas that closely resembled the areas of necrosis and cavitation observed in the histopathology. The events leading to infarct and tissue loss have been reported to occur early after the ischemie episode. Monsein and colleagues21 have described in baboons that received a complete MCA occlusion, increased intensity in T2-weighted images as early as 3.1 h after the MCA occlusion. Furthermore, post-mortem analysis of the tissue proved that the hyperintensity was related to early tissue edema and progressed to infarction and necrosis in the brain.

Anatomical analysis revealed a relatively large area bordering the necrotic lesion which appeared normal on T1- and T2-weighted images on MRI, yet had fewer NeuN stained cells. The expression of NeuN is observed in most neuronal cell types throughout the nervous system32. Immunohistochemically detectable NeuN protein first appears at developmental stages that correspond with the withdrawal of the neuron from the cell cycle and the initiation of terminal differentiation of the neuron, and is well known neuronal marker32. Glial and vascular tissue were seen and no necrosis was apparent. This area with partial cell loss may correspond to zones of the ischemic penumbra at the time of occlusion of the MCA. Neurons were lost in that area in the acute phase of the stroke. However, the long term conditions in this zone may allow to sustain cells and in that fashion, be an appropriate target for neural transplantation as well as for neuroprotective strategies. Identification of this area will be key in order to perform these therapies. Although T1- and T2-weighted images lack sensitivity to distinguish between normal and neuronal depleted areas of the brain, ^sup 1^H MRS may identify areas of neuronal loss in the brain.

Decreased concentration of NAA has been described in rodents16-20, monkeys21,22 and humans14,23-25. These previous analyses had targeted the most abnormal appearing area on MRI, which undergoes neuronal death and tissue loss. In this study we found a significant decrease in NAA in the area adjacent to the stroke, a region that maintained overall neuroanatomical integrity in T1- and T2-weighted images. It can be speculated that the decreased NAA value could be related to a 'washed out' signal induced by increased water in the cerebral tissue, as observed during post-ischemic edema. However, in our study the measurements were obtained eight months after the ischemic episode and the reactive edema had already resolved.

Changes in Cho as well as Ml have also been described few hours after stroke in the ischemic area21,22,25. In addition to the different target area, the long period of time after stroke could stabilize those metabolites, such as Cho25 that is affected by phospholipid membrane turnover or La (lactate)23 that increases when cells shift to carbohydrate catabolism. Local changes, such as astrocytosis (suggested to be measured by Ml levels) were limited to the periphery of the VOI area and could pass undetected, averaged by VOI recording.

Previous stroke studies performed in human and nonhuman primates have used ^sup 1^H MRS VOIs ranging between 4 and 8 cm^sup 3^. In comparison, in our measurements a volume of 1 cm^sup 3^ was chosen to adjust to the specific area of interest in the monkey brain and prevent 'volume averaging'. In spite of a smaller VOI, which provides less sensitive signal-to-noise ratio, we were able to detect changes in NAA concentration. Translation of this technology to human patients will benefit from the larger volume of the human brain compared to the Cynomologous macaque brain. In consequence, a larger VOI will improve the signal-to-noise ratio of the spectroscopies and the sensitivity of the method. Furthermore, advanced ^sup 1^H MRS techniques are being developed with capabilities of mapping the metabolite concentration over a smaller volume of interest and in multiple voxels within the VOI, therefore contributing to more sensitive delineation of the changes over the infarcted and adjacent areas.

Diverse therapeutic approaches can benefit from this scanning technology. Based on the pre-clinical and clinical success of cell transplantation for the treatment of Parkinson's and Huntington's disease, cell replacement therapy is being investigated for cerebral ischemia. Different authors3,33 have proposed the use of fetal tissue or cloned cells to induce functional restoration. However, the selection of the target area for transplantation will determine the survival and any functional benefit of the implanted cells. The combination of MRI scans and ^sup 1^H MRS may provide a noninvasive tool for identification of a region adjacent to the stroke that may benefit from the intervention. Furthermore, trophic and other neuroprotective therapies34 may use NAA/Cr levels measured by ^sup 1^H MRS as an in vivo index of neuronal survival to monitor the affected area.

CONCLUSION

New strategies to treat completed stroke are being developed. Neural grafting to treat cerebral infarction was tried in rodent models35-37. A pilot study of neural transplantation in humans was started recently33. We believe that an appropriate nonhuman primate model is important at this time in order to study the effects of treatment for completed, chronic stroke, as well as the effects of acute neuroprotective intervention. However, even recent models are geared toward the acute phase of the disease7,11,12.

To the best of our knowledge, our model is the first to address chronic MCA distribution stroke. It is also the first to combine behavioral testing, including a cognitive function test, long term behavioral and imaging follow up, the use of MR spectroscopy and neuropathological correlation. This combination of quantifiable parameters increases the sensitivity of this model and makes it particularly suitable for studies of the effects of both neuroprotective and regenerative interventions using cell transplantation.

ACKNOWLEDGEMENTS

We would like to thank Atilla Halil Elhan from the Biostatistics Department at the Ankara University Faculty of Medicine for expert assistance with statistical analysis. We thank Dr Sylvia E. Perez for expert photographic assistance and Dr Mahmoud Mafee for critical comments. This work was funded by seed grants from the departments of Radiology and Neurosurgery of the University of Illinois at Chicago, and the department of Neurological Sciences at Rush University.

REFERENCES

1 Goldstein LB, Adams R, Becker K, Furberg C, Gorelick PB, Hademenos G, Hill M, Howard G, Howard VJ, Jacobs B, Levine SR, Mosca L, Sacco RL, Sherman DG, Wolf PA, del Zoppo GJ. Primary prevention of ischemic stroke. A statement for Healthcare professionals from the stroke council of the American Heart Association. Stroke 2001; 32: 280-299

2 Gorelick PB, Sacco RL, Smith DB, Alberts M, Mustone-Alexander L, Rader D, Ross JL, Raps E, Ozer MN, Brass LM, Malone ME, Goldberg S, Booss J, Hanley DF, Toole JF, Greengold NL, Rhew DC. Prevention of a first stroke. A review of guidelines and multidisciplinary consensus statement from the National Stroke Association. JAMA 1999; 281: 1112-1120

3 Borlongan CV, Tajima Y, Trojanowski JQ, Lee VM, Sanberg PR. Transplantation of cryopreserved human embryonal carcinoma-derived neurons (NT2N cells) promotes functional recovery in ischemic rats. Exp Neural 1998; 149: 310-321

4 Lawrence MS, McLaughlin JR, Sun GH, ho DY, McIntosh L, Kunis DM, Sapolsky RM, Steinberg GK. Herpes simplex viral vectors expressing Bcl-2 are neuroprotective when delivered after a stroke. J Cereb Blood Flow Metab 1997; 17: 740-744

5 Crowell RM, Olsson Y, Klatzo I, Ommaya A. Temporary occlusion of the middle cerebral artery in the monkey: Clinical and pathological observations. Stroke 1970; 1: 439-448

6 Yonas H, Gur D, Claassen D, Wolfson SK Jr, Moossy J. Stable xenon enhanced computed tomography in the study of clinical and pathologic correlates of focal ischemia in baboons. Stroke 1988; 19: 228-238

7 Henry PT, Chandy MJ. Effect of ascorbic acid on infarct size in experimental focal cerebral ischemia and reperfusion in a primate model. Acta Neurochir (Wien) 1998; 140: 977-980

8 Takamatsu H, Tsukada H, Kakiuchi T, Nishiyama S, Noda A, Umemura K. Detection of reperfusion injury using PET in a monkey model of cerebral ischemia. J Nucl Med 2000; 41 : 1409-1416

9 Kito G, Nishimura A, Susumu T, Nagata R, Kuge Y, Yokota C, Minematsu K. Experimental thromboembolic stroke in cynomolgus monkey. J Neurosci Meth 2001; 105: 45-53

10 Auer RN, Coupland SG, Jason GW, Archer DP, Payne J, Belzberg AJ, Ohtaki M, Tranmer Bl. Postischemic therapy with MK-801 (dizocilpine) in a primate model of transient focal brain ischemia. Mol Chem Neuropathol 1996; 29: 193-210

11 Frazee JG, Luo X, Luan G, Hinton DS, Hovda DA, Shiroishi MS, Barcliff LT. Retrograde transvenous neuroperfusion: A back door treatment for stroke. Stroke 1998; 29: 1912-1916

12 Young AR, Touzani O, Derlon JM, Sette G, MacKenzie ET, Baron JC. Early reperfusion in the anesthetized baboon reduces brain damage following middle cerebral artery occlusion: A quantitative analysis of infarction volume. Stroke 1997; 28: 632-638

13 Castillo M, Kwock L, Mukherji SK. Clinical applications of proton MR spectroscopy. AJNR 1996; 17: 1-15

14 Federico F, Simone II, Lucivero V, Giannini P, Laddomada G, Mezzapesa DM, Tortorella C. Prognostic value of proton magnetic resonance spectroscopy in ischemic stroke. Arch Neurol 1998; 55: 489-494

15 Brand A, Richter-Landsherg C, Leibfritz D. Multinuclear NMR studies on the energy metabolism of glial and neuronal cells. Dev Neurosci 1993; 15: 289-298

16 Dreher W, Kuhn B, Gyngell ML, Busch E, NiendorfT, Hossmann KA, Leibfritz D. Temporal and regional changes during focal ischemia in rat brain studied by proton spectroscopic imaging and quantitative diffusion NMR imaging. Magn Reson Med 1998; 39: 8778-8888

17 Malisza KL, Kozlowski P, Peeling J. A review of in vivo ^sup 1^H magnetic resonance spectroscopy of cerebral ischemia in rats. J Biochem Cell Biol 1998; 76: 487-496

18 Norris DG, Hoen-Berlage M, Dreher W, Kohno K, Busch E, Schmitz B. Characterization of middle cerebral artery occlusion infarct development in the rat using fast nuclear magnetic resonance proton spectroscopic imaging and diffusion-weighted imaging. J Cereb Blood Flow Metab 1998; 18: 749-757

19 Sager TN, Laursen H, Fink-Jensen A, Topp S, Stensgaard A, Hedehus M, Rosenbaum S, Valsborg JS, Hansen AJ. N-Acetyl-aspartate distribution in rat brain striatum during acute brain ischemia. J Cereb Blood Flow Metab 1999; 19: 164-172

20 Wardlaw JM, Marshall I, Wild J, Dennis MS, Cannon J, Lewis SC. Studies of acute ischemic stroke with proton magnetic resonance spectroscopy: Relation between time from onset, neurological deficit, metabolite abnormalities in the infarct, blood flow, and clinical outcome. Stroke 1998; 29: 1618-1624

21 Monsein LH, Mathew VP, Barker PB Pardo CA, Blackband SJ, Whitlow WD, Wong DF, Bryan RN. Irreversible regional cerebral ischemia: Serial MR imaging and proton MR spectroscopy in a nonhuman primate model. AJNR 1993; 14: 963-970

22 Handa Y, Kaneko M, Matuda T, Kobayashi H, Kubota T. In vivo proton magnetic resonance spectroscopy for metabolic changes in brain during chronic cerebral vasospasm in primates. Neurosurgery 1997; 40: 773-790

23 Graham GD, Blamire AM, Rothman DL, Brass LM, Fayad PB, Petroff OA, Prichard JW. Early temporal variation of cerebral metabolites after human stroke. A proton magnetic resonance spectroscopy study. Stroke 1993; 24: 1891-1896

24 Gideon P, Sperling B, Arlien-Soborg P, Olsen TS, Henriksen O. Long-term follow-up of cerebral infarction patients with proton magnetic resonance spectroscopy. Stroke 1994; 25: 967-973

25 Houkin K, Kamada K, Kamiyama H, Iwasaki Y, Abe H, Kashiwaba T. Longitudinal changes in proton magnetic resonance spectroscopy in cerebral infarction. Stroke 1993; 24: 1316-1321

26 Laurent JP, Molinari GF, Moseley JI. Clinicopathological validation of a primate stroke model. Surg Neurol 1975; 4: 449-455

27 Marshall JW, Cross AJ, Jackson DM, Green AR, Baker HF, Ridley RM. Clomethiazole protects against hemineglect in a primate model of stroke. Brain Res Bull 2000; 5291: 21-29

28 Emborg ME, Ma SY, Mufson EJ, Levey Al, Taylor MD, Brown WD, Holden JE, Kordower JH. Age-related declines in nigral neuronal function correlate with motor impairment in rhesus monkeys. J Comp Neurol 1998; 401: 253-265

29 Roitberg BZ, Emborg ME, Sramek JG, Palfi S, Kordower JH. Behavioral and morphological comparison of two nonhuman primate models of Huntington's disease. Neurosurgery 2002; 50: 137-147

30 Palfi S, Conde F, Riche D, Brouillet E, Dautry C, Mittoux V, Chibois A, Peschanski M, Hantraye P. Fetal striatal allografts reverse cognitive deficits in a primate model of Huntington's disease. Nature Medicine 1998; 4: 963-966

31 Chu Y, Cochran EJ, Bennett DA, Mufson EJ, Kordower JH. Downregulation of trkA mRNA within nucleus basalis neurons in individuals with mild cognitive impairment and Alzheimer's disease. J Comp Neurol 2001; 437: 296-307

32 Mullen RJ, Buck CR, Smith AM. NeuN, a neuronal specific nuclear protein in vertebrates. Development 1992; 116: 201-211

33 Kondziolka D, Wechsler L, Goldstein S, Meltzer C, Thulborn KR, Gebel J, Jannetta P, DeCesare S, Elder EM, McGrogan M, Reitman MA, Bynum I. Transplantation of cultured human neuronal cells for patients with stroke. Neurology 2000; 55: 565-569

34 Emborg ME, Kordower JH. Delivery of therapeutic molecules into the CNS. Prog Brain Res 2000; 128: 323-332

35 Grabowski M, Brundin P, Johansson BB. Functional integration of cortical grafts placed in brain infarcts of rats. Ann Neurol 1993; 34: 362-368

36 Mattsson B, Sorensen JC, Zimmer J, Johansson BB. Neural grafting to experimental neocortical infarcts improves behavioral outcome and reduces thalamic atrophy in rats housed in enriched but not in standard environments. Stroke 1997; 28: 1225-1231

37 Nishino H, Czurko A, Onizuka K, Fukuda A, Hida H, Ungsuparkorn C, Kunimatsu M, Sasaki M, Karadi Z, Lenard L. Neuronal damage following transient cerebral ischemia and its restoration by neural transplant. Neurobiology (Bp) 1994; 2: 223-234

Ben Roitberg*, Naimath Khan*, Eray Tuccar*, Katie Kompoliti[dagger], Yaping Chu[double dagger], Noam Alperin[sec], Jeffrey H. Kordower[double dagger] and Marina E. Emborg[double dagger]

* Department of Neurosurgery, [sec] Department of Radiology, University of Illinois at Chicago, Chicago, IL [dagger] Department of Neurological Sciences, [double dagger] Research Center for Brain Repair and Department of Neurological Sciences, Rush Presbyterian-St. Luke's Medical Center, Chicago, IL, USA

Correspondence and reprint requests to: B. Roitberg, MD, Department of Neurosurgery M/C 799, University of Illinois at Chicago, 912 S. Wood St., Chicago, IL 60612, USA. [roitberg@uic.edu] Accepted for publication September 2002.

Copyright Forefront Publishing Group Jan 2003
Provided by ProQuest Information and Learning Company. All rights Reserved