Journal of Cerebral Blood Flow & Metabolism (2015) 35, 11–19 © 2015 ISCBFM All rights reserved 0271-678X/15 $32.00 www.jcbfm.com
RAPID COMMUNICATION
Longitudinal changes in resting-state brain activity in a capsular infarct model Donghyeon Kim1,7, Ra Gyung Kim2,7, Hyung-Sun Kim2, Jin-Myung Kim3, Sung Chan Jun1, Boreom Lee 2, Hang Joon Jo4, Pedro R Neto5, Min-Cheol Lee3 and Hyoung-Ihl Kim2,6 Strokes attributable to subcortical infarcts have been increasing recently in elderly patients. To gain insight how this lesion influences the motor outcome and responds to rehabilitative training, we used circumscribed photothrombotic capsular infarct models on 36 Sprague-Dawley rats (24 experimental and 12 sham-operated). We used 2-deoxy-2-[18F]-fluoro-D-glucose-micro positron emission tomography (FDG-microPET) to assess longitudinal changes in resting-state brain activity (rs-BA) and daily singlepellet reaching task (SPRT) trainings to evaluate motor recovery. Longitudinal FDG-microPET results showed that capsular infarct resulted in a persistent decrease in rs-BA in bilateral sensory and auditory cortices, and ipsilesional motor cortex, thalamus, and inferior colliculus (P o0.0025, false discovery rate (FDR) qo 0.05). The decreased rs-BA is compatible with diaschisis and contributes to manifest the malfunctions of lesion-specific functional connectivity. In contrast, capsular infarct resulted in increase of rs-BA in the ipsilesional internal capsule, and contralesional red nucleus and ventral hippocampus in recovery group (P o0.0025, FDR qo0.05), implying that remaining subcortical structures have an important role in conducting the recovery process in capsular infarct. The SPRT training facilitated motor recovery only in rats with an incomplete destruction of the posterior limb of the internal capsule (PLIC) (Pearson’s correlation, P o 0.05). Alternative therapeutic interventions are required to enhance the potential for recovery in capsular infarct with complete destruction of PLIC. Journal of Cerebral Blood Flow & Metabolism (2015) 35, 11–19; doi:10.1038/jcbfm.2014.178; published online 29 October 2014 Keywords: cerebrovascular disease; lacunar infarcts; positron emission tomography; rehabilitation; white-matter disease
INTRODUCTION Subcortical white-matter infarcts represent 15% to 25% of all strokes, mostly found in the end-arterial regions of the white matter.1,2 Recent increase of subcortical infarct in elderly people demands an intensive research in this field.3,4 Nonetheless, a few studies have investigated longitudinal changes in brain activity (BA) after subcortical capsular infarct or the neural mechanisms of subsequent motor recovery, because there is a lack of pertinent animal models of subcortical capsular infarct.2,5,6 Internal capsule is known as a key anatomic structure that contains corticospinal tract (CST) carrying the motor signals from several motor cortices to lower motor neurons in the spinal cord. Also, the density of corticospinal fibers is greatest in the posterior part of the posterior limb of internal capsule (PLIC). However, lesioning of PLIC for subcortical white-matter stroke has been limited for several reasons. For example, a rat has substantially less white matter compared with human or primates. Also, the internal capsule of a rat has irregular and narrow shape, making it difficult not only to access it stereotactically but also to control the extent of infarct. 2,7
Capsular infarct leads to different clinical features and recovery patterns depending on the size and location of infarct within the internal capsule. It was assumed that motor deficit after infarct of internal capsule has good prognosis.8 However, other studies argued that capsular infarct can have poor prognosis if the lesion is extended into the PLIC or thalamus since these areas contain many efferent fibers of the CST.9 The detailed studies on relationship of location and extent of infarct within internal capsule are lacking. In addition, pathophysiology of cell death and repair in capsular infarct resulting from the occlusion of the end arteries in subcortical stroke is likely to be different from large artery gray-matter cortical stroke,2 leading to different mechanisms of poststroke recovery. Recently, functional reorganization of surviving neurons to compensate for the stroke-induced functional deficit has been extensively studied. Functional imaging studies, such as functional magnetic resonance imaging and positron emission tomography (PET), have been used to assess the multifarious aspect of cortical reorganization during stroke recovery. Converging evidence from
1 School of Information and Communication, Gwangju Institute of Science and Technology, Gwangju, Republic of Korea; 2Department of Medical System Engineering and School of Mechatronics, Gwangju Institute of Science and Technology, Gwangju, Republic of Korea; 3Department of Pathology, Chonnam National University Medical School, Gwangju, Republic of Korea; 4Section on Functional Imaging Methods, Laboratory of Brain and Cognition, National Institute of Mental Health, National Institutes of Health, Bethesda, Maryland, USA; 5Translational Neuroimaging Laboratory, McGill Center for Studies in Aging, Douglas Mental Health University Institute, Montreal, Quebec, Canada and 6 Department of Neurosurgery, Presbyterian Medical Center, Jeonju, Republic of Korea. Correspondence: Dr H-I Kim, Department of Medical System Engineering, Gwangju Institute of Science and Technology, 261 Cheomdan-gwagiro, Gwangju 500-712, South Korea. E-mail:
[email protected] 7 These authors contributed equally to this work. This work was supported by a grant from the Institute of Medical System Engineering (iMSE) & GIST-Caltech Collaborative Fund (K03912) from GIST and by the Basic Science Research Program through NRF of Korea funded by the Ministry of Science, ICT and future Planning (NRF-2013R1A2A2A01067890). Hang Joon Jo was supported by the National Institute of Mental Health, National Institutes of Health, Division of Intramural Research. Received 4 August 2014; revised 10 September 2014; accepted 11 September 2014; published online 29 October 2014
Brain activity in capsular infarct D Kim et al
12 imaging studies on the striatocapsular stroke showed that there are increased activations of bilateral motor pathways and recruitment of sensory and secondary motor structures.8,10 In addition, several imaging studies suggest that integrity of CST is important to predict the behavioral outcome as well as to assess the recovery potential in capsular infarct.11,12 However, longitudinal studies using neuroimaging tools in the model of circumscribed pure capsular infarct in PLIC have never been performed before. Recently, it has been suggested that functional connectivity related to the lesion is more relevant in manifesting the clinical symptoms of the focal stroke lesion as well as in reorganizing the functional and structural network toward stroke recovery.13,14 It was recognized that stroke and its recovery are closely associated with lesion-specific alterations in functional connectivity.14,15 In line with this, 2-deoxy-2-[18F]-fluoro-D-glucose-microPET (FDG-microPET) is an unique imaging method to measure resting-state regional glucose metabolism, which can provide an index of local synaptic activity as well as biochemical maintenance process in the pathologic brain.16,17 Exploration of the metabolic brain network may lead to more insight into the relationship between behavioral deficit and resting-state BA (rs-BA).17,18 Moreover, if this relationship is further studied in the longitudinal framework during behavioral recovery, then one can study dynamic activity of the neural structures involved in poststroke recovery. In our previous study, we established a circumscribed capsular infarct model with minimal encroachment of neighboring tissues and persistent motor deficit.7 However, it had not been shown what may be confounding factors in motor recovery and how rehabilitative training may affect motor recovery in this specific model. In this study, we longitudinally analyzed behavioral changes coupled with changes of rs-BA in capsular infarct model. We hypothesized that our ‘pure capsular infarct model’ may show capsular infarct-specific brain activity and longitudinal follow-up of rs-BA with micro-PET may reveal the mechanism of recovery across the brain. We also used the single-pellet reaching task (SPRT) training to enhance the motor recovery. The SPRT training was known to be the most sensitive test to measure the poststroke behavioral recovery or learning-dependent neural plasticity.19,20 Also, repeated training of SPRT may influence functional reorganization not only in reforming a neural network but also in interacting together for reacquisition of the skills.20,21 Thus, we aimed to elucidate the specific role of SPRT training in capsular stroke recovery by analyzing the relationship between behavioral patterns and changes of rs-BA. MATERIALS AND METHODS Experimental Animals Animal experiments were performed according to the institutional guidelines of the GIST (Gwangju Institute of Science and Technology) and all procedures were approved by Institutional Animal Care and Use Committee at GIST. Animal ARRIVE guidelines were followed in the preparation of the manuscript. Rats were housed two per cage in a controlled animal facility at 21 ± 1° C with water ad libitum. The animal care unit was maintained on a 12-hour light-dark cycle with lights on at 07:00 h. Thirty-six Sprague-Dawley rats (9 weeks old) were used in this experiment. Twenty-four rats underwent photothrombotic stroke lesioning in the PLIC and twelve rats served as the sham-operated group (SOG). All animals performed daily SPRT training for 3 weeks. Performance was measured immediately after stroke and 1 week after stroke. On the basis of their performances in the SPRT, animals were subdivided into a ‘moderate recovery group’ (MRG) and a ‘poor recovery group’ (PRG). Moderate recovery was defined as an increase in the performance score by 50% Journal of Cerebral Blood Flow & Metabolism (2015), 11 – 19
Figure 1. Experimental set-up: grouping of experimental animals (top right), time line of longitudinal microPET scan (middle bar) and behavioral training (low bar) in the capsular stroke model. For the time line of longitudinal microPET scan, each rat was scanned five times: Baseline scanning before infarct lesioning, postlesion days 4, 7, 14, and 21 (PL 4, PL 7, PL 14, and PL 21) to longitudinally follow-up on changes in regional glucose metabolism from stroke onset to behavioral recovery. For the behavior training, initial single-pellet reaching task (SPRT) was performed 7 to 10 days before the surgery. Additional training was resumed 2 days after operation and continued until 21 days after the surgery. PET, positron emission tomography.
compared with scores before lesioning. Poor recovery was defined as recovery of o 50%. Circumscribed Capsular Infarct Lesioning using a Photothrombotic Technique A diagram of the experimental design is shown in Figure 1. The photothrombotic capsular infarct model was created as previously described.7 In brief, experimental animals were anesthetized with a mixture of ketamine hydrochloride (100 mg/kg) and xylazine (7 mg/kg). Rat heads were fixed using a small animal stereotactic frame. A thermocouple blanket was placed under the experimental rat, and rectal temperature was maintained at 37 ± 0.5°C. Rat heads were shaved and prepared aseptically. An approximately 2-cm skin incision was made in the midline in the frontoparietal area, and the pericranium was stripped out. After confirming that the bregma and the lambda were at the same level, a small hole (2 mm posterior to bregma and 3.1 mm lateral to midline) was made, and durotomy was performed. An unjacked optical fiber (core diameter: 62.5 μm and outer diameter: 125 μm) was slowly lowered to reach the PLIC (7.2 mm ventral to dura). Rose Bengal dye (20 mg/kg) was injected through the tail vein, followed by 1.5 minutes of light irradiation with a green laser (intensity of approximately 4.0 mW at the tip of the optical fiber). After light irradiation, the optical fiber was slowly raised from the target. The scalp wound was secured, and the rat was released from the stereotactic frame and transferred to a recovery chamber. For the SOG, the same procedure for photothrombotic infarct lesioning was performed, except that this group received an injection of saline (0.2 mL/100 g) instead of Rose Bengal dye. Behavioral Task and Assessment of Recovery Behavioral training and assessments were performed according to Gharbawie et al..22 Briefly, rats were food restricted to 90% of the control rats throughout the experiments to motivate food intake. The reaching task box was made of Plexiglas (40 cm × 45 cm × 13 cm wide) with 1 cm wide slit in the middle of the frontal wall. The food shelf attached to the slot outside the box, and a sucrose pellet (Bio-Serve, Frenchtown, NJ, USA) was obliquely placed on the shelf contralateral to the preferred limb. Rats were trained to reach through the slit and retrieve the pellet from the shelf in a schedule of 20 pellets or 20 minutes in one session per day. Successful reaches were counted as the number of trials when a rat grabbed the pellet and brought it into the cage and to its mouth without dropping. The percentage of successful reaching © 2015 ISCBFM
Brain activity in capsular infarct D Kim et al
13
Figure 2. Histologic findings of the capsular infarct lesion using H&E (A), glial fibrillary acidic protein (GFAP) (B), and Luxol fast blue–PAS staining (C). Note the destruction of the internal capsule with minimal encroachment of the neighboring thalamus (arrow).
was calculated by the following formula: Number of successf ul reaches ´ 100 20 All animals were trained for 2 weeks before surgery. Average performance during the last 3 days of preoperative training was set as the preinfarct baseline. Postoperative training was given for 3 weeks using the same procedure. Image Acquisition Each rat was scanned five times: Baseline scanning before infarct lesioning, postlesion days 4, 7, 14, and 21 (PL 4, PL 7, PL 14, and PL 21) to longitudinally follow-up on changes in regional glucose metabolism from stroke onset to behavioral recovery. All the scans were performed using a microPET/CT scanner (Inveon, Siemens Medical Solution, Knoxville, TN, USA). Every individual PET scan was reconstructed in a volume space having a transaxial resolution of 1.4 mm full width at half maximum (field of view = 12.7 mm). Each rat was deprived of food for 12 hours and injected with FDG (0.1 mCi/100 g) through a tail vein under brief isoflurane anesthesia. After an uptake period of 30 minutes in awake state, animals were anesthetized by inhalational isoflurane (2% in 100% oxygen) and positioned on the scanning bed with their head immobilized in a customized head holder (Hyosung Inc., Gwangju, Korea). Vital signs, including respiration (50 ± 5 per minute), heart rate (280 ± 20 beats/min), and body temperature (37.0 ± 1°C), were monitored throughout scanning (BioVet, m2m Imaging Corp., Cleveland, OH, USA). A 25-minute static acquisition and 5-minute attenuationcorrection computed tomography were performed. Image Processing After scanning, images were corrected for attenuation and reconstructed using the iterative OSEM3D/MAP algorithm. Imaging analysis was performed using an MINC tool kit (McConnell Brain Imaging Centre, Montreal Neurological Institute, Montreal, Canada) and Analysis of Functional NeuroImages (AFNI) packages.23 Images for every individual animal were normalized using an intensity scaling approach, spatially registered to the first scan of a subject, and manually corrected for slight misalignments with visual inspection. The registered images were then manually aligned to a standard histologic template and Schweinhardt atlas in Paxinos coordinates24 and spatially smoothed using an isotropic Gaussian kernel with 1.2 mm full width at half maximum. We chose one coordinate to represent the anatomic location of increased or decreased activation among four follow-up scans. © 2015 ISCBFM
Figure 3. Motor recovery patterns in single-pellet reaching task (SPRT) performances for the three different groups after capsular infarct over time. Statistical significance was determined using Student’s t-test (Po0.05). †SOG versus MRG; ‡SOG versus PRG; *MRG versus PRG. SOG, sham-operated group; MRG, moderately recovered group; PRG, poorly recovered group.
Statistical Analysis A group-level linear mixed-effect model was conducted with the 3dLME program (in the AFNI package) to assess group differences between the baseline and PL 4, PL 7, PL 14, or PL 21 images for each group (PRG, MRG, and SOG).25 Statistical maps of the results were corrected and thresholded at the significance level (P o 0.0025, false discovery rate (FDR) qo 0.05), and then overlaid on the histologic template to show significant changes in BA. We used AlphaSim program in AFNI package with the significance threshold for each comparison set at α = 0.05, resulting in a cluster threshold, k, of 17 voxels for our PET data analysis. Artifacts presumed to be generated by the ear channel were masked out. To determine the recovery patterns after capsular infarct, SOG was chosen as a reference to be compared with MRG and PRG. The MRG and the PRG were directly compared to elucidate the difference of rs-BA between two groups. For multiple comparisons, we used AlphaSim program in AFNI to reduce the missing of significant data (family-wise error correction) and set a threshold level (P o 0.05, α o 0.05) in the group-level comparisons. We manually defined region of interest (ROI) masks over anatomic brain regions that were activated or deactivated to quantify changes in cerebral metabolic rate of glucose after capsular Journal of Cerebral Blood Flow & Metabolism (2015), 11 – 19
Brain activity in capsular infarct D Kim et al
14
Figure 4. Longitudinal change in resting-state brain activity (rs-BA) in the capsular infarct model over time at PL4, PL7, PL14, and PL21 in MRG, PRG, and SOG. Brain metabolic differences between baseline (PL0) and longitudinal scans were used to determine decrease or increase in rsBA. Images were thresholded at a significant level of Po0.0025 after false discovery rate correction (qo 0.05). The cortical network and ipsilesional inferior colliculus are impaired (colored in cyan to blue), and motor-related subcortical structures (internal capsule and red nucleus) and contralesional ventral hippocampus show increased activity (colored in yellow to red). The y coordinates indicate the distance from anterior commissure. MRG, moderately recovered group; PRG, poorly recovered group; SOG, sham-operated group; M, motor cortex; S, sensory cortex; Aud, auditory cortext; IC, internal capsule; RN, red nucleus; HT, hypothalamus; VH, ventral hippocampus; CC, corpus callosum; I. Coll, inferior colliculus; S.Coll, superior colliculus; RSG, retrosplenial cortex; LSN, lateral septal nucleus; PL, postlesion.
infarct. Activity in each ROI mask was separately averaged for each scan and normalized to the activity in the whole brain. To investigate the correlation between changes of metabolic activities and SPRT performances, we selected three ROIs including ipsilesional motor, sensory, and auditory cortex, which showed the marked change of brain activities after capsular infarct lesioning and additional three ROIs including ipsilesional internal capsule, and contralesional red nucleus and ventral hippocampus, which are Journal of Cerebral Blood Flow & Metabolism (2015), 11 – 19
likely to be related with motor recovery. We compared the metabolic activities in these six ROIs with SPRT performances across the session using Pearson’s correlation (Po0.04, FDR qo0.05). Neurohistologic Examination After completing the behavioral and microPET studies, the rats were euthanized after deep anesthesia by CO2 inhalation and © 2015 ISCBFM
Brain activity in capsular infarct D Kim et al
15 perfused with 0.9% saline solution followed by a 4% paraformaldehyde solution. Brains were carefully removed from the skull cavity, postfixed overnight in 4% paraformaldehyde and cryoprotected in 30% sucrose in phosphate-buffered saline. Brains were coronally sectioned with 40-μm thickness at 200-μm intervals and stained with hematoxylin and eosin (n = 14). Luxol fast blueperiodic acid-Schiff and immunohistochemical staining, including glial fibrillary acidic protein, were performed on brain slices. Infarct volumes were measured using the ImageJ software (NIH, Bethesda, MD, USA). The completeness of lesioning in the PLIC was compared by three of the authors who were blinded to the groups. RESULTS Neurohistologic Finding of Capsular Infarct Lesion A daily behavioral test (SPRT) and longitudinal microPET scans were performed (Figure 1) using the previously established model of capsular infarct to examine the unique features of capsular infarct and its subsequent impact on other functional networks.7 Neurohistologic analysis showed losses in axons and myelin in the central cavity (Luxol fast blue–periodic acid-Schiff), which were surrounded by reactive gliosis (as measured using the glial marker, glial fibrillary acidic protein) and infiltrating macrophages (as measured using an hematoxylin and eosin stain) (Figure 2). The infarct volume was 0.7 ± 0.4 mm3 in the MRG and 0.8 ± 0.5 mm3 in the PRG. Despite the similar lesion sizes between MRG and PRG, the lesion extent of the PRG covered the whole breadth of PLIC whereas that of the MRG did not cover the PLIC completely (Supplementary Figure 1). Behavioral Recovery Single-pellet reaching task was used as a measure of motor recovery for 3 weeks after the lesion. Rats were divided into three groups as follows: PRG (N = 12), MRG (N = 12), and SOG (N = 12). The MRG exhibited improved motor recovery 1 week after lesion compared with the SOG, and this group continued through the 21st postoperative day. However, the PRG did not show improvements in this task throughout the observation period (Figure 3). Longitudinal Change of Resting-State Brain Activity We acquired serial images of rs-BA in awake animals using FDGmicroPET to assess longitudinal changes in rs-BA in a network of regions.16 We measured the cerebral metabolic rate of glucose at 0 day (PL0; base), 4 days (PL 4), 7 days (PL 7), 14 days (PL 14), and 21 days (PL 21) (Figures 4 and 5). Brain metabolic differences between baseline and longitudinal scans were used to determine decrease or increase in rs-BA. Decreases in rs-BA were observed largely in cortical structures, including bilateral sensory and auditory cortices and the ipsilesional motor cortex (Figure 4). In contrast, two subcortical structures, including the thalamus in the MRG and the PRG and the inferior colliculus in the PRG (Figure 4), showed prominent decreases in rs-BA on the ipsilateral side. These decreases persisted in sensory and auditory cortices during the observation period, but they disappeared in subcortical structures until PL 14 (Figure 4). These same areas showed a significant decrease in activation after FDR correction (qo 0.05). We extracted brain areas with increased rs-BA from trained animals in the MRG and the PRG to identify neural substrates that might underlie functional reorganization in the brain after SPRT training. The ipsilesional internal capsule, bilateral red nucleus and ventral hippocampi, and contralesional superior colliculus in the MRG exhibited increased rs-BA. In contrast, increased activity was showed in the peri-lesional internal capsule (hereafter, internal capsule refers to peri-lesional PLIC), contralesional ventral hippocampus, superior colliculus and striatum, corpus callosum, and bilateral retrosplenial gyrus (RSG) in the PRG (Figures 4 and 5). © 2015 ISCBFM
Figure 5. Schematic drawing showing the changes of resting-state brain activities (corrected, *P o0.05) during poststroke recovery period. Asterisk (*) indicates the significant brain activities compared with the basal brain activities (baseline scan). I, ipsilesional; C, contralesional; Sup. Coll., superior colliculus; Inf. coll, inferior colliculus; RSG, retrosplenial gyrus; IC, internal capsule; Red Nu., red nucleus; Vent. Hippo., ventral hippocampus; Lat. Sep. Nu., lateral septal nucleus; MRG, moderately recovered group; PRG, poorly recovered group; SOG, sham-operated group.
Notably, increased activity was prominent in the ipsilesional internal capsule and the contralesional red nucleus during the early phase of recovery (PL 7) in both groups, whereas increased activity was continuously maintained in the contralesional ventral hippocampus and superior colliculus (Figure 4). These same areas showed a significant increase in activation after FDR correction (qo 0.05) (Figure 4; Table 1). The internal capsule showed the areas of activation beyond the size of infarct (2.64 mm3 in MRG and 2.04 mm3 in PRG). The MRG and the PRG were again compared with SOG to identify the difference of rs-BA during the recovery period. The MRG showed the increased activation in ipsilesional internal capsule, and contralesional red nucleus and hippocampus in early Journal of Cerebral Blood Flow & Metabolism (2015), 11 – 19
Brain activity in capsular infarct D Kim et al
16 Table 1. Group
MRG
Results of resting-state brain activity (rs-BA) in experimental groups of capsular stroke Activation
Decreased rs-BA
Increased rs-BA
PRG
Decreased rs-BA
Increased rs-BA
SOG
Decreased rs-BA Increased rs-BA
Anatomic area
Motor (ipsi) Sensory (ipsi) Sensory (cont) Thalamus (ipsi) Auditory (ipsi) Auditory (cont) IC (ipsi) Red.Nu (ipsi) Red.Nu (cont) V.Hippo (ipsi) V.Hippo (cont) S.Col (cont) Hypothalamus Motor (ipsi) Sensory (ipsi) Striatum (ipsi) Thalamus (ipsi) I.Col (ipsi) Auditory (ipsi) Auditory (cont) IC (ipsi) V.Hippo (cont) RSG (ipsi) RSG (cont) S.Col (cont) Striatum (cont) CC Auditory (ipsi) Auditory (cont) Lat. Sep. Nu.
Representative coordinates (mm)
Cluster (t-value)
X
Y
Z
2.9 5.3 − 5.6 2.4 6.2 − 6.4 3 0 − 0.8 5.6 − 4.1 − 0.8 0 3.2 5.4 3.8 2.6 2.8 6.8 − 6.4 3.3 − 4.6 0.8 − 0.9 − 0.1 − 3.7 − 0.3 6.7 −7 0
− 3.5 0.8 0.9 3.1 4.6 4.6 2.9 6.9 6.3 5.8 6.1 4.3 2.8 7.3 0.7 − 0.1 2.6 7.6 4.7 4.9 2.8 6.5 5.5 5.2 5.1 0.6 3 5 4.2 2.8
3.9 3.3 4.1 1.4 3.5 3.5 − 0.9 0.4 − 0.1 − 1.2 2.7 2.6 − 0.5 5.6 1.7 − 0.8 1.3 3.2 1.9 3 − 0.8 1.3 6.2 6.1 2.1 0.8 4.6 3 3 − 0.5
Day_4 107 90 56 51 172 103
(−3.8 ± 0.3) (−3.8 ± 0.5) (−3.6 ± 0.3) (−3.7 ± 0.4) (−4.4 ± 0.9) (−3.9 ± 0.5)
Day_7
Day_14
Day_21
142 (−3.7 ± 0.3) 116 (−3.7 ± 0.4)
38 (−3.4 ± 0.2) 53 (−3.5 ± 0.3)
26 (−3.5 ± 0.2)
183 (−4.2 ± 0.6) 54 (−3.6 ± 0.3) 22 (3.6 ± 0.3)
121 (−4.3 ± 0.8) 30 (−3.6 ± 0.4)
87 (−4.0 ± 0.6) 26 (−3.5 ± 0.2)
21 (3.5 ± 0.2) 53 (3.5 ± 0.3) 25 (3.7 ± 0.3)
17 (3.4 ± 0.1) 28 (3.5 ± 0.3) 29 (3.7 ± 0.4)
123 (3.75 ± 0.4)
78 (3.64 ± 0.3)
98 (−3.7 ± 0.4) 119 (−3.9 ± 0.6) 76 (−3.6 ± 0.3)
76 (−3.7 ± 0.4)
163 (−4.3 ± 0.8) 68 (3.8 ± 0.4) 42 (3.5 ± 0.3) 76 (3.7 ± 0.3)
18 (3.5 ± 0.3)
34 265 42 17
(−3.7 ± 0.4) (−4.5 ± 1.0) (−4.0 ± 0.5) (3.9 ± 0.5)
53 (3.5 ± 0.2) 80 (−3.6 ± 0.9) 21 (−3.6 ± 0.2)
57 (3.7 ± 0.4) 26 36 42 37 24
(3.4 ± 0.1) (3.7 ± 0.4) (3.61 ± 0.4) (−3.5 ± 0.2) (−3.3 ± 0.1)
38 (4.0 ± 0.5)
106 (−4.2 ± 0.8)
46 (−3.7 ± 0.4)
90 (3.6 ± 0.3) 30 (3.8 ± 0.4) 21 (3.4 ± 0.2)
32 (3.3 ± 0.1) 69 (−3.4 ± 0.7) 41 (−3.5 ± 0.8)
CC, corpus callosum; contra, contralesional; IC, internal capsule; I.Col, inferior colliculus; ipsi, ipsilesional; Lat. Sep. Nu., lateral septal nucleus; MRG, moderate recovery group; Nu, nucleus; PRG, poor recovery group; Red Nu, red nucleus; rs-BA, resting-state brain activity; RSG, retrosplenial gyrus; S.Col, superior colliculus; SOG, sham-operated group; V.Hippo, ventral hippocampus. The stereotactic coordinates according to Paxinos atlas represent the overlapped areas of peak metabolic changes in longitudinal scans and coordinates indicate the distance from anterior commissure. Results at Po0.05 (corrected).
phase of recovery (PL4 and PL7) compared with SOG (P o 0.05, αo 0.05). In addition, the MRG showed the decrease of rs-BA in ipsilesional motor cortex (PL 4) and striatum (PL 14), and contralesional auditory cortex (PL 21, P o0.05, α o 0.05). The PRG showed early increased activation in the internal capsule at PL 7 and specifically lasted longer in contralesional striatum, superior colliculus, and ventral hippocampus (PL 21) (Figures 4 and 5; Supplementary Figure 2). Also, the PRG showed relatively prolonged persistent decrease of rs-BA in bilateral motor cortices (PL 14, P o 0.05, α o 0.05). Direct comparison between the MRG and the PRG showed the decreased activity in bilateral striatum and contralesional auditory cortex, and the increased activity in contralesional hippocampus. Comparison of FDG-microPET Findings with Behavioral Recovery The SOG showed no significant change of motor performances after capsular infarct. Although the PRG showed cortical and subcortical imaging changes, it did not show any improvement in motor deficits. The MRG showed motor recovery after PL 4. A positive correlation was found in the MRG in the ipsilesional motor and sensory cortex, with SPRT performances indicating that motor recovery was associated with a restoration of decreased rs-BA in these regions (ipsilesional motor, r = 0.35, P = 0.02; ipsilesional sensory, r = 0.41, P = 0.01 (Figure 6A). A negative correlation was observed in the MRG in the contralesional red nucleus and ventral hippocampus, which indicated that these regions showed timeJournal of Cerebral Blood Flow & Metabolism (2015), 11 – 19
dependent increases in rs-BA and returned to baseline (contralesional red nucleus, r = − 0.36, P = 0.02; contralesional ventral hippocampus, r = − 0.31, P = 0.04) (Figure 6B). However, ipsilesional auditory cortex and internal capsule did not show statistically significant correlations. DISCUSSION In this study, we investigated the longitudinal change of rs-BA after a circumscribed capsular infarct. Capsular infarct decreased the cortical activity in motor, sensory, auditory cortex, and subcortical structures including thalamus and inferior colliculus, implying that PLIC is functionally connected with these structures and interacting together to manifest the poststroke symptoms. Capsular infarct induced the activation mostly in peri-infarct internal capsule and red nucleus regardless of severity of infarct lesion, which seemed to have a role in stroke recovery. Daily reaching training did not contribute much to motor recovery if the degree of capsular infarct was severe; therefore, more aggressive therapeutic intervention is required. Targeted destruction of the PLIC with subsequent motor impairment due to capsular infarct is difficult to establish an animal capsular infarct model.2,7 Consequently, the lack of a pertinent model has been a barrier to the comprehensive study of capsular infarct. Previously, Frost et al6 established an endothelin-1 induced capsular infarct model that showed subsequent behavioral changes. Nevertheless, the degree of deficit was not as © 2015 ISCBFM
Brain activity in capsular infarct D Kim et al
17
Figure 6. Correlation analyses of activity changes after single-pellet reaching task (SPRT) performances in the moderately recovered group (N = 11). (A) Example plots of positive correlation between metabolic activity and SPRT in ipsilesional motor and sensory cortices (B) Example plots of negative correlation in the contralesional red nucleus and ventral hippocampus. Pearson’s correlation analysis (P o0.04, false discovery rate (FDR) q o0.05). I, ipsilesional; C, contralesional; Nu., nucleus; Vent. Hippo., ventral hippocampus.
severe as human sensorimotor cortex lesions. There were inconsistent data on the clinical recovery presumably depending on the extent and location of infarct lesion within internal capsule. We previously generated a circumscribed pure capsular infarct model with comparable sustained motor deficits. The model was created using low energy light and different scattering effects in gray and white matter. The light intensity (o 814 mW/mm2) was low enough to limit the extent to which photothrombosis lesions affected the target area (i.e., PLIC). Accordingly, this technique was helpful in inducing infarct lesions accurately in the PLIC with minimal encroachment on the striatum. This is the first study to delineate the relation between circumscribed capsular infarct in PLIC and longitudinal change of BA. We used FDG-microPET to image rs-BA in awake rats. MicroPET is a useful tool for the imaging of regional distributions of glucose metabolism in the resting state because [18F]-FDG is trapped in neurons after injection at a rate that is proportional to glucose metabolism in the brain. ‘Resting state’ functional magnetic resonance imaging or diffusion tensor imaging have been developed to determine brain connectivity in normal and pathologic states,26,27 but PET imaging is unique because it defines a physiologic baseline state of the brain. Specifically, PET provides absolute and qualitative measurements of regional glucose metabolism, which index continuous synaptic communications among neurons and astrocytes.17,28 Our findings illustrate the changes that occur in intrinsic BA coupled with dynamic signaling within neurovascular units and the temporal changes in regional interactions and metabolic plasticity during recovery after stroke.29,30 © 2015 ISCBFM
Our results showed that capsular infarct induced decreases in rs-BA in various cortical areas despite the absence of cortical stroke lesions. Different from previous task-based functional imaging studies in which activation of sensorimotor cortex is the main feature, decreased rs-BA across brain regions in ipsilateral or bilateral hemispheres may provide a useful allegory that is consistent with most recent views of the brain as an ensemble of functional networks.13,31 In accordance with this idea, the PLIC is likely to have relevant subcortico-cortical connections within the affected hemisphere and between the two hemispheres. A decrease in rs-BA is compatible with the concept of ‘diaschisis,’ which affects remote areas of the brain that are nonetheless connected to the primary site of pathology.15,32 Diaschisis exhibits a prolonged time course (e.g., lasting several weeks), which is similar to the persistent cortical decrease in rs-BA in our capsular model. For example, Seitz et al33 noted that the diaschisis area seems to subserve active relearning in stroke recovery. Therefore, one could imagine that any maneuvers that could increase metabolic or neuronal interactions in the decreased rs-BA could enhance motor recovery in capsular stroke. Our results also showed an increase in rs-BA across different regions after capsular infarct. Notably, motor-related subcortical areas showed increased rs-BA in a time-dependent manner. For example, peak activity was showed in the internal capsule and red nucleus at PL 7 and PL 14. This finding suggests that the remaining fibers in the internal capsule may be actively recruited in the reorganization of functional connections after capsular stroke at a maximum rate approximately 1 week after capsular stroke. Although our histologic examination showed that destroyed internal capsule is surrounded by reactive gliosis and Journal of Cerebral Blood Flow & Metabolism (2015), 11 – 19
Brain activity in capsular infarct D Kim et al
18 macrophage infiltration, we observed that the area of activation was much wider than the area of infarct lesion, indicating that axonal outgrowth and myelination may occur during stroke recovery around the infarct area of the internal capsule.34 In addition, activation of red nucleus and contralesional ventral hippocampus was predominantly observed only in MRG but not in PRG. It is likely that corticorubral projection to contralateral red nucleus responsive to reaching trainings could provide the drive to the rubrospinal tract serving the recovery of the affected limb when task-specific rehabilitative interventions are used. Hippocampus is known to contribute to acquire new motor skills through the repeated performances of motor sequence.35 However, it remains to be determined how contralesional ventral hippocampus contributes to the motor recovery in capsular infarct. Taken together, the MRG is hypothesized to show the recovery by coordinating action of contralesional red nucleus and ventral hippocampus in addition to the increased rs-BA in internal capsule. Skilled reaching training improves poststroke motor deficits of the forelimb by increasing dendrite arborization,36 forelimb movement representation area and behavioral LTP induction in the perilesional area.37 Similarly, recovery in capsular and medullary CST injuries is critically mediated by fiber sprouting in the remaining pyramidal fibers.38,39 Although increased rs-BA in subcortical structures, including the internal capsule and red nucleus, was showed in both groups, SPRT training failed to improve motor deficit in the PRG, in whom the PLIC was completely destroyed. The degree to which the CST is intact is critical to recovery in our capsular stroke model; therefore, more potent methods alternative to SPRT are required to boost motor recovery. In conclusion, the present study shows that capsular stroke induces the decreased rs-BA in both hemispheres due to its extensive cortico-subcortical connections. The decreased rs-BA is compatible with diaschisis and contributes to manifest the malfunctions of lesion-specific functional connectivity. The increased rs-BA was evident in the internal capsule and red nucleus and may indicate the relevant role of remaining corticospinal and rubrospinal fibers in capsular stroke recovery. This study shows the limitation of conventional reaching training in improving the persistent motor deficit in the PRG due to capsular infarct of PLIC despite its small size lesion. Further alternative therapeutic interventions are thus required to enhance the potential for recovery in capsular infarct. DISCLOSURE/CONFLICT OF INTEREST The authors declare no conflict of interest.
REFERENCES 1 Bamford J, Sandercock P, Dennis M, Burn J, Warlow C. Classification and natural history of clinically identifiable subtypes of cerebral infarction. Lancet 1991; 337: 1521–1526. 2 Sozmen EG, Kolekar A, Havton LA, Carmichael ST. A white matter stroke model in the mouse: axonal damage, progenitor responses and MRI correlates. J Neurosci Methods 2009; 180: 261–272. 3 Debette S, Markus HS. The clinical importance of white matter hyperintensities on brain magnetic resonance imaging: systematic review and meta-analysis. Br Med J 2010; 341: c3666 4 Roger VL, Go AS, Lloyd-Jones DM, Benjamin EJ, Berry JD, Borden WB et al. Heart disease and stroke statistics--2012 update: a report from the American Heart Association. Circulation 2012; 125: e2–e220. 5 Lecrux C, McCabe C, Weir CJ, Gallagher L, Mullin J, Touzani O et al. Effects of magnesium treatment in a model of internal capsule lesion in spontaneously hypertensive rats. Stroke 2008; 39: 448–454. 6 Frost SB, Barbay S, Mumert ML, Stowe AM, Nudo RJ. An animal model of capsular infarct: endothelin-1 injections in the rat. Behav Brain Res 2006; 169: 206–211. 7 Kim HS, Kim D, Kim RG, Kim JM, Chung E, Neto PR et al. A rat model of photothrombotic capsular infarct with a marked motor deficit: a behavioral, histologic, and microPET study. J Cereb Blood Flow Metab 2014; 34: 683–689.
Journal of Cerebral Blood Flow & Metabolism (2015), 11 – 19
8 Calautti C, Leroy F, Guincestre JY, Baron JC. Dynamics of motor network overactivation after striatocapsular stroke: a longitudinal PET study using a fixedperformance paradigm. Stroke 2001; 32: 2534–2542. 9 Fries W, Danek A, Scheidtmann K, Hamburger C. Motor recovery following capsular stroke. Role of descending pathways from multiple motor areas. Brain 1993; 116: 369–382. 10 Gerloff C, Bushara K, Sailer A, Wassermann EM, Chen R, Matsuoka T et al. Multimodal imaging of brain reorganization in motor areas of the contralesional hemisphere of well recovered patients after capsular stroke. Brain 2006; 129: 791–808. 11 Lindenberg R, Renga V, Zhu LL, Betzler F, Alsop D, Schlaug G. Structural integrity of corticospinal motor fibers predicts motor impairment in chronic stroke. Neurology 2010; 74: 280–287. 12 Ward NS, Newton JM, Swayne OB, Lee L, Thompson AJ, Greenwood RJ et al. Motor system activation after subcortical stroke depends on corticospinal system integrity. Brain 2006; 129: 809–819. 13 Carter AR, Shulman GL, Corbetta M. Why use a connectivity-based approach to study stroke and recovery of function?Neuroimage 2012; 62: 2271–2280. 14 van Meer MP, van der Marel K, Wang K, Otte WM, El Bouazati S, Roeling TA et al. Recovery of sensorimotor function after experimental stroke correlates with restoration of resting-state interhemispheric functional connectivity. J Neurosci 2010; 30: 3964–3972. 15 Witte OW, Bidmon HJ, Schiene K, Redecker C, Hagemann G. Functional differentiation of multiple perilesional zones after focal cerebral ischemia. J Cereb Blood Flow Metab 2000; 20: 1149–1165. 16 Rilling JK, Barks SK, Parr LA, Preuss TM, Faber TL, Pagnoni G et al. A comparison of resting-state brain activity in humans and chimpanzees. Proc Natl Acad Sci USA 2007; 104: 17146–17151. 17 Raichle M, Mintun M. Brain work and brain imaging. Annu Rev Neurosci 2006; 29: 449–476. 18 Eidelberg D. Metabolic brain networks in neurodegenerative disorders: a functional imaging approach. Trends Neurosci 2009; 32: 548–557. 19 Whishaw IQ, Whishaw P, Gorny B. The structure of skilled forelimb reaching in the rat: a movement rating scale. J Vis Exp 2008; 18: doi:3791/816. 20 Kleim JA, Barbay S, Cooper NR, Hogg TM, Reidel CN, Remple MS et al. Motor learning-dependent synaptogenesis is localized to functionally reorganized motor cortex. Neurobiol Learn Mem 2002; 77: 63–77. 21 Metz GA, Antonow-Schlorke I, Witte OW. Motor improvements after focal cortical ischemia in adult rats are mediated by compensatory mechanisms. Behav Brain Res 2005; 162: 71–82. 22 Gharbawie OA, Gonzalez CL, Williams PT, Kleim JA, Whishaw IQ. Middle cerebral artery (MCA) stroke produces dysfunction in adjacent motor cortex as detected by intracortical microstimulation in rats. Neuroscience 2005; 130: 601–610. 23 Cox RW. AFNI: software for analysis and visualization of functional magnetic resonance neuroimages. Comput Biomed Res 1996; 29: 162–173. 24 Rubins DJ, Melega WP, Lacan G, Way B, Plenevaux A, Luxen A et al. Development and evaluation of an automated atlas-based image analysis method for microPET studies of the rat brain. Neuroimage 2003; 20: 2100–2118. 25 Chen G, Saad ZS, Britton JC, Pine DS, Cox RW. Linear mixed-effects modeling approach to FMRI group analysis. Neuroimage 2013; 73: 176–190. 26 Takenobu Y, Hayashi T, Moriwaki H, Nagatsuka K, Naritomi H, Fukuyama H. Motor recovery and microstructural change in rubro-spinal tract in subcortical stroke. Neuroimage Clin 2013; 4: 201–208. 27 Xu H, Qin W, Chen H, Jiang L, Li K, Yu C. Contribution of the resting-state functional connectivity of the contralesional primary sensorimotor cortex to motor recovery after subcortical stroke. PLoS ONE 2014; 9: e84729 28 Raichle ME, Snyder AZ. A default mode of brain function: a brief history of an evolving idea. Neuroimage 2007; 37: 1083–1090, discussion 1097-9 29 Tatsch K, Koch W, Linke R, Poepperl G, Peters N, Holtmannspoetter M et al. Cortical hypometabolism and crossed cerebellar diaschisis suggest subcortically induced disconnection in CADASIL: an 18F-FDG PET study. J Nucl Med 2003; 44: 862–869. 30 Magistretti PJ. Neuron-glia metabolic coupling and plasticity. J Exp Biol 2006; 209: 2304–2311. 31 Dijkhuizen RM, Singhal AB, Mandeville JB, Wu O, Halpern EF, Finklestein SP et al. Correlation between brain reorganization, ischemic damage, and neurologic status after transient focal cerebral ischemia in rats: a functional magnetic resonance imaging study. J Neurosci 2003; 23: 510–517. 32 Andrews RJ. Transhemispheric diaschisis. A review and comment. Stroke 1991; 22: 943–949. 33 Seitz RJ, Azari NP, Knorr U, Binkofski F, Herzog H, Freund HJ. The role of diaschisis in stroke recovery. Stroke 1999; 30: 1844–1850. 34 Ueno Y, Chopp M, Zhang L, Buller B, Liu Z, Lehman NL et al. Axonal outgrowth and dendritic plasticity in the cortical peri-infarct area after experimental stroke. Stroke 2012; 43: 2221–2228.
© 2015 ISCBFM
Brain activity in capsular infarct D Kim et al
19 35 Gheysen F, Van Opstal F, Roggeman C, Van Waelvelde H, Fias W. Hippocampal contribution to early and later stages of implicit motor sequence learning. Exp Brain Res 2010; 202: 795–807. 36 Luke LM, Allred RP, Jones TA. Unilateral ischemic sensorimotor cortical damage induces contralesional synaptogenesis and enhances skilled reaching with the ipsilateral forelimb in adult male rats. Synapse 2004; 54: 187–199.
37 Rioult-Pedotti MS, Friedman D, Donoghue JP. Learning-induced LTP in neocortex. Science 2000; 290: 533–536. 38 Terashima T. Anatomy, development and lesion-induced plasticity of rodent corticospinal tract. Neurosci Res 1995; 22: 139–161. 39 Z'Graggen WJ, Fouad K, Raineteau O, Metz GA, Schwab ME, Kartje GL. Compensatory sprouting and impulse rerouting after unilateral pyramidal tract lesion in neonatal rats. J Neurosci 2000; 20: 6561–6569.
Supplementary Information accompanies the paper on the Journal of Cerebral Blood Flow & Metabolism website (http://www.nature. com/jcbfm)
© 2015 ISCBFM
Journal of Cerebral Blood Flow & Metabolism (2015), 11 – 19
