The study was conducted in accordance with the Korean Academy of Medical Sciences. Animals were generally anesthetized and monitored following corresponding standard procedures (Laboratory Animal Manual 2000, Korean Academy of Medical Sciences, Seoul). All experimental procedures were approved by the Kyung Hee University Medical Center Institutional Animal Care and Use Committee, Seoul. Thirty seven Sprague-Dawley rats weighting 295 ± 8 g (mean ± standard deviation, 9 weeks old) were used. Animals were purchased from Samtako Bio Korea Co., Ltd. (Gyeonggi-do, Republic of Korea). A 1-week acclimation period to an inverse day and night rhythm (7 p.m.-7 a.m.) was used before animals were used in the experimental protocol. Rats were kept at a constant room temperature (24℃) and humidity (55-70%), and received standard rat chow and tap water ad libitum.

The rats were randomly assigned to four groups: (1) sham group (n = 9) received a sham operation, (2) saline group (n = 12) had middle cerebral artery occlusion/reperfusion (MCAO/R) operation and received 6 mL/kg saline perorally, (3) low fat milk (LFM) group (n = 8) had occlusion/reperfusion (MCAO/R) operation and received 12 mL/kg LFM perorally, (4) and BC group (n = 8) had occlusion/reperfusion (MCAO/R) operation and received 6 mL/kg defatted liquid BC (Percoba liquid, Immuno Dynamics, Ltd., Perry, IA, USA) perorally. Saline, FLM, and BC were administered once daily, perorally for seven days. The rat chow contained 19.5 g/100 g protein, no albumin, 60.6 g/100 g carbohydrate, 10 g/100 g fat, and 4.08 kcal/100 g calories. The BC diet had 6.0 g/dL protein, 2.0 g/dL albumin, 10 g/100 g carbohydrate, < 0.1 g/dL fat, and 64 kcal/dL calories. The LFM formulation had 3.0 g/dL protein, no albumin, 5 g/dL carbohydrate, 1.0 g/dL fat, and 40 kcal/dL calories, which was adjusted to have half the protein and approximately two-thirds the calories of BC.

Rats were treated between 10:00 and 12:00 to avoid non-specific effects due to the circadian rhythm. Behavioral functions were evaluated at 1, 3, and 7 days after operation. Seven days following the operation, rats were anesthetized by 5% isoflurane inhalation and blood samples were withdrawn by cardiocentesis. Rats were then sacrificed by decapitation and brains were removed to measure infarct volume.

Each rat was anesthetized with isoflurane (initiated at 5% and maintained at 2%) in 30% oxygen and 70% nitrous oxide using a vaporizer and face mask. The rat was mounted in the prone position on a dual small animal stereotaxic instrument (David Kopf Instruments, Los Angeles, CA, USA). After cranial midline incision, a 1-mm burr hole was made with microdrill on both sides, 2-mm-posterior and 5-mm-lateral to bregma [15], and the rat was placed in the supine position. A laser Doppler probe was placed on the burr hole under the rat skull, nearly contacting the dura mater and avoiding large vessels, and measured the parietal cortical perfusion in the MCA territory. The laser Doppler probe was connected to BLF21D laser Doppler flowmeter (Transonic systems Inc., Ithaca, NY, USA), and values were obtained beginning 10 minutes before the ischemic event until 15 min after reperfusion [16]. The changes in regional cerebral blood flow (rCBF) were expressed as percentages of baseline values. Signals were digitized for recording, storage and analysis.

After the laser Doppler probe was placed on the skull, left middle cerebral artery occlusion (MCAO) was induced as described by Zea Longa et al. (1989) and revised by Lourbopoulos (2008) [17,18]. The left common carotid artery (CCA) and external carotid artery (ECA) were exposed under an operation microscope followed by electrical coagulation of ECA branches. The internal carotid artery (ICA) was dissected to the level of the pterygopalatine artery. After the distal ECA was ligated permanently, a silk thread (6-0) was placed loosely around the ECA stump. The CCA and ICA were occluded temporarily using microvascular clips. A small incision was made on the ECA, and a 25-mm nylon thread (4-0), with a rounded tip and silicon rubber cylinder (300-340 µm in diameter) was inserted. The silk thread was held around the ECA stump to prevent bleeding, and the microvascular clip on the ICA was removed. The nylon thread in the ECA was gently advanced through the ICA until Laser Doppler Flowmetry showed a sharp decrease in regional blood flow in the MCA to approximately 20% of baseline, as determined by monitoring [19]. The microvascular clip on the CCA was removed and the incision closed. Two hours after inducing ischemia, reperfusion was performed by removing the nylon thread from the ICA under enflurane anesthesia as described above. Restoration of blood flow was identified in the operation field and confirmed by laser Doppler flowmetry. Animals in the sham operation group were subjected to the same operation without inserting the nylon filament. Rectal temperature was monitored continuously and maintained at 37.0 ± 0.5℃ using an electrical blanket and heating lamp throughout the experiment.

Extracted brains from sacrificed rats were chilled in ice-cold saline for 5 min, and sectioned coronally into six 2-mm slices using a rat brain matrix (RBM-4000C, ASI® Instruments Inc., Warren, MI, USA). The slices were then immersed in saline solution containing 2% 2,3,5-triphenyltetrazolium chloride (TTC; Sigma, St. Louis, MO, USA) for 30 min at 37℃ in the dark, followed by 10% buffered formalin fixation [16]. Brain slices were photographed using a digital camera (Camedia C-5060WZ, Olympus Corporation, Tokyo, Japan) and infarct areas were quantified using an image analysis system (Optimas 6.5 Media Cybermetics, Silver Springs, MD, USA). The infarct area in each slice was calculated by subtracting the normal (ipsilateral) area from the contralateral hemisphere to reduce errors due to cerebral edema, and was presented as a percentage of normal [20].

Serum IL-1β, TNF-α, IL-6, and IL-10 concentrations were measured with commercially available quantitative sandwich enzyme-linked immunosorbent assay (ELISA) kits (Quantikine M, R&D Systems Inc. Minneapolis, MN, USA). The change in absorbance in each well at 450 nm was measured with a microplate reader (model ELx800; Bio-Tek Instruments Inc., Winooski, VT, USA) and levels were expressed as picograms per milliliter.

Tests were performed by a blinded investigator at predetermined times to exclude behavioral changes associated with the circadian rhythm.

The modified neurologic severity scores (mNSS) is a composite of motor, sensory (visual, tactile and proprioceptive), and reflex tests. Neurological function was graded on a scale of 0-18 (normal score 0; maximal deficit score 18). One point is awarded for the inability to perform a test or the lack of a tested reflex; thus, a higher score denotes a more severe injury [21].

In the corner test, a rat was placed between two 30 cm×20 cm×1 cm boards. The edges of the boards were attached at a 30° angle with a small opening between the boards to encourage entry into the corner. The rat was placed between the two angled boards facing the corner and halfway to the corner. When the rat reached the corner, both sides of the body were simultaneously stimulated. Then the animal usually reared and turned either right or left. Each animal was tested in 10 trials, and the chosen sides for turning were noted. In the case of ventral turning (i.e., when the animal turned without rearing), the trial was discarded and repeated at the end of the session. The laterality index was calculated according to the formula: Laterality index = (number of left turns-number of right turns) / total number of turns [22].

All data were given as mean ± SE values. Differences between groups were evaluated by analysis of variance (ANOVA) followed by Student-Newman-Keuls multiple comparison test. Statistical analyses were performed with SPSS for Windows 13.0 software (SPSS Inc., Chicago, IL). Differences between two groups were compared by Student's t-test. A P value < 0.05 was considered statistically significant.

Average body weights were increased to approximately 285-330 g during the experimental period. Rat chow intake was decreased significantly in saline and LFM groups compared to the sham group (P < 0.05), but there was no significant difference between sham and BC groups. Protein and calorie intakes were similar among all groups (Table 1).

The regional cerebral blood flow (rCBF) was decreased during MCAO and showed the same basic patterns in all three operation groups. The rCBF reduction during MCAO tended to be slightly less severe in the LFM group compared with the saline or BC groups (LFM, 24.5 ± 4.5% of baseline; Saline, 20.5 ± 2.5% of baseline; BC, 20.3 ± 2.6% of baseline), but there were no significant differences among the MCAO/R groups (P > 0.05). rCBF was also increased following the same basic pattern for after 10 minutes MCAR with no significant differences among groups (P > 0.05) (Fig. 1).

MCAO/R made large infarct in the saline group (40.863 ± 4.907%). LFM (37.073 ± 6.604%) did not reduce the infarct volume significantly compared with the saline group (P = 0.625). BC rats (15.003 ± 3.672%) showed a significant reduction of brain infarct volume compared with saline (P < 0.005) and LFM rats (P < 0.05) (Fig. 2A and B).

The saline and LFM groups showed significantly elevated serum TNF-α levels compared to the sham group (P < 0.005 vs. saline; P < 0.001 vs. LFM). The BC group had significantly lower serum TNF-α than the saline/LFM groups (P < 0.05 vs. saline; P < 0.01 vs. LFM) (Fig. 3A). The saline and LFM groups showed significantly elevated serum IL-1β levels compared to the sham group (P < 0.001). The BC group had significantly lower IL-1β than the saline group (P < 0.001). The BC group had lower IL-1β than the LFM group, but the difference was not significant (P = 0.072) (Fig. 3B). The saline and LFM groups showed significantly elevated serum IL-6 levels compared to the sham group (P < 0.001 vs. Saline; P < 0.005 vs. LFM). The BC group had significantly lower IL-6 than the saline/LFM groups (P < 0.01 vs. saline; P < 0.05 vs. LFM) (Fig. 3C). The saline and LFM groups showed significantly elevated Serum IL-10 levels compared to the sham group (P < 0.01). The BC group had less IL-10 than the saline and LFM groups, but the difference was not significant (P = 0.105 vs. saline; P = 0.088 vs. LFM) (Fig. 3D).

mNSS: One day after MCAO/R injury, all rats in operation groups showed neurologic deficits (saline 10.1 ± 1.02, LFM 10.2 ± 1.33, and BC 10.1 ± 1.60), but there were no significant differences among the groups till 3 days after the operation. Seven days after operation, the BC group (5.4 ± 0.71) scored significantly lower than the saline (9.1 ± 0.97) and LFM (8.4 ± 0.99) groups (P < 0.01 vs. saline group, P < 0.05 vs. LFM group) (Fig. 4A).

Corner test: All rats in operation groups showed higher percentages of right turns (saline 96.5 ± 1.09, LFM 96.4 ± 1.41, and BC 98.1 ± 0.64) at one day after MCAO/R injury, but there were no significant differences among the groups. Three days after the operation, the BC group (82.0 ± 3.32) made significantly fewer right turns compared to the LFM group (92.0 ± 2.61) (P < 0.05). The BC group (76.1 ± 2.72) made significantly fewer right turns than both the saline (88.9 ± 2.54) and LFM (87.6 ± 2.70) groups (P < 0.005 vs. saline group; P < 0.01 vs. LFM group) at 7 days (Fig. 4B).