|Home | About | Journals | Submit | Contact Us | Français|
Abnormal β-amyloid (Aβ) deposits in the thalamus have been reported after cerebral cortical infarction. In this study, we investigated the association of Aβ deposits, with the secondary thalamic damage after focal cortical infarction in rats. Thirty-six stroke-prone renovascular hypertensive rats were subjected to distal middle cerebral artery occlusion (MCAO) and then randomly divided into MCAO, vehicle, and N-[N-(3,5-difluorophenacetyl)--alanyl]-S-phenylglycine t-butyl ester (DAPT) groups and 12 sham-operated rats as control. The DAPT was administered orally at 72hours after MCAO. Seven days after MCAO, sensory function, neuron loss, and glial activation and proliferation were evaluated using adhesive removal test, Nissl staining, and immunostaining, respectively. Thalamic Aβ accumulation was evaluated using immunostaining and enzyme-linked immunosorbent assay (ELISA). Compared with vehicle group, the ipsilateral thalamic Aβ, neuronal loss, glial activation and proliferation, and the mean time to remove the stimulus from right forepaw significantly decreased in DAPT group. The mean time to remove the stimulus from the right forepaw and thalamic Aβ burden were both negatively correlated with the number of thalamic neurons. These findings suggest that Aβ deposits are associated with the secondary thalamic damage. Reduction of thalamic Aβ by γ-secretase inhibitor may attenuate the secondary damage and improve sensory function after cerebral cortical infarction.
Focal cerebral infarction not only causes local neural tissue death and neurologic deficits, but also leads to secondary neurodegeneration in the nonischemic areas remote from yet connected synaptically with the site of primary ischemic lesion, thereby hindering the functional recovery (Freret et al, 2006; Iizuka et al, 1990; Liang et al, 2007). Accumulating experimental evidences have revealed neuronal loss and glial proliferation in the ipsilateral ventroposterior nucleus (VPN) of the thalamus after focal cerebral infarction (Dihne et al, 2002; Iizuka et al, 1990; Ling et al, 2009; Watanabe et al, 1998). Several possible explanations have been proposed for the secondary neuronal damage in the VPN such as retrograde degeneration of thalamocortical projections (Dihne et al, 2002; Ross and Ebner, 1990), inflammation (Block et al, 2005), apoptosis (Soriano et al, 1996), inhibition of axonal regeneration by Nogo-A (Wang et al, 2007), and oxidative DNA damage (He et al, 2007). However, the precise mechanism is still unclear.
β-Amyloid (Aβ) peptide, a major component of senile plaque in the Alzheimer's disease brain is derived from proteolytic cleavage of a large transmembrane amyloid precursor protein through the amyloidogenic pathway mediated by the sequential action of β- and γ-secretase (Selkoe, 2001). The blockade of either β- or γ-secretase would ultimately prevent production of Aβ (Wolfe, 2002), which is known to be neurotoxic (Lambert et al, 1998; Yankner et al, 1990; Yankner and Lu, 2009). Recently, Aβ has been shown to accumulate abnormally and aggregate to dense plaque-like deposits in the ipsilateral thalamus for up to 9 months after transient middle cerebral artery occlusion (MCAO) in rats (Hiltunen et al, 2009; Makinen et al, 2008; van Groen et al, 2005). However, the role of Aβ deposits in the secondary thalamic damage after MCAO is unknown.
This study was conducted to investigate whether the Aβ deposits were associated with the secondary thalamic damage after permanent distal MCAO in hypertensive rats and the reduction of Aβ deposits by N-[N-(3,5-difluorophenacetyl)--alanyl]-S-phenylglycine t-butyl ester (DAPT), a functional γ-secretase inhibitor, could attenuate the secondary damage and improve sensory function.
The experimental protocol was approved by the local ethics committee for animal research and was conducted under the guidelines for animal experimentation. All efforts were made to minimize the number of animals used and their suffering. A total of 60 male Sprague–Dawley rats (5 weeks old, body weight 70 to 90g) were established as stroke-prone renovascular hypertensive rats, with the method described previously (Zeng et al, 1998b). Briefly, the rats were anesthetized with 10% chloral hydrate (3mL/kg, intraperitoneally) and underwent an operation of renal artery constriction with two-kidney two clips. Resting blood pressure was measured by an indirect tail-cuff sphygmomanometer (ML866 Powerlab 4/30, ADInstruments Pty Ltd, Sydney, Australia) once weekly. Food and water were available ad libitum except for the night before DAPT treatment.
After 12 weeks, 52 of 60 rats weighing 350 to 450g, whose systolic blood pressure were steadily higher than 180mmHg without stroke symptoms, were selected for further study, in which 40 rats were subjected to permanent distal MCAO and 12 rats as sham-operated control randomly. Permanent distal MCAO by electrocoagulation was performed as previously described (Bederson et al, 1986). Briefly, the rats were anesthetized as above. Under an operating microscope, the left MCA was exposed through a burr hole and occluded by bipolar electrocoagulation distal to the origin of the striatal branches, which caused permanent focal infarction in the left dorsolateral cerebral cortex. Sham-operated animals underwent the same surgical procedures except for electrocoagulation of the MCA. Body temperature was maintained at 37°C±0.5°C during the surgical procedures and recovery periods using a heating pad. Thirty-six rats with successful MCAO were divided randomly into three groups: MCAO, vehicle, and DAPT groups (12 rats per each group), and the remaining four rats with failed MCAO were excluded from the experiment.
We first optimized the intervention time and DAPT dose in a pilot study. In our pilot study, we found that Aβ did not appear in the ipsilateral thalamus at day 2, started to increase at day 3, and robustly increased at day 7 after MCAO. On the basis of our previous studies, day 3 after MCAO was selected as the time point for DAPT intervention. To select an appropriate dose for the intervention, two doses of DAPT (25 and 50mg/kg; Sigma-Aldrich, St Louis, MO, USA) were applied to ischemic rats at day 3 after MCAO according to the previous report (El Mouedden et al, 2006). We found that a single administration of 25mg/kg dose of DAPT given at 72hours by oral gavage after MCAO had no significant effect on brain Aβ levels in stroke-prone renovascular hypertensive rats, whereas 50mg/kg dose significantly reduced the thalamic Aβ. Therefore, DAPT at 50mg/kg was used in this study. At 72hours after distal MCAO and an overnight fast, vehicle and DAPT group rats were treated with vehicle (5% ethanol in corn oil) and a solution of DAPT in vehicle at the final volume of 10mL/kg given by oral gavage, respectively. The MCAO and sham-operated group rats received an overnight fast but no oral gavage.
Adhesive removal test was performed blindly in all experimental rats to assess the somatosensory deficit as previously described (Schallert et al, 1983) before MCAO and at 7 days after MCAO. Briefly, after the rat was familiarized with the testing environment, two small pieces of adhesive paper dots (of equal size, 113.1mm2) were accurately attached to the distal-radial region on the wrist of each forelimb with an equal pressure. The time to remove each stimulus from the forelimbs was recorded in five trials per day for each forepaw. Individual trials were separated by at least 5minutes. After 3 consecutive days training, all the rats were able to remove the dots within 10seconds and then subjected to MCAO.
At 7 days after distal MCAO, after a deep anesthesia with 10% chloral hydrate intraperitoneally, eight rats selected randomly from each group were perfused transcardially with 0.9% saline at 4°C, followed by 4% paraformaldehyde in phosphate buffer (0.1mol/L, pH 7.4). The brains were removed, postfixed in the same fixative for 6hours at 4°C, and immersed sequentially in 20% and 30% sucrose until sunk. Coronal sections (10-μm thick) were cut with a cryostat (CM1900, Leica, Heidelberger, Germany) and stored at −80°C. A serial tissue section from Bregma 4.7 to −5.2mm was used for Nissl staining, and serial sections from Bregma −2.4 to −4.4mm for immunohistochemistry.
Nissl staining with 0.1% cresyl violet (Sigma) was performed after a standard histochemical procedure to determine the infarct volume in cerebral cortex and neural morphology in the thalamus. The relative infarct volume was presented as a volume percentage of the lesion compared with the contralateral hemisphere and quantitatively analyzed from Bregma 4.7 to −5.2mm as described previously (Ling et al, 2009; Swanson et al, 1990).
Immunohistochemistry was performed using Envision immunohistochemical technique as described previously (He et al, 2007). Briefly, sections were pretreated for 10minutes with hot (85°C) 0.01mol/L citrate buffer (pH 6.0), rinsed in phosphate-buffered saline three times for 5minutes each, then treated with 3% hydrogen peroxide for 10minutes followed by rinsing in phosphate-buffered saline three times. After blocking with 5% normal goat serum for 1hour at room temperature, the sections were incubated with the primary antibody for rabbit polyclonal antirodent Aβ3-16 (1:1000, #SIG-39151, Covance, Emeryville, CA, USA) overnight at 4°C. The sections were then rinsed in phosphate-buffered saline three times and incubated with peroxidase-marked rabbit/mouse secondary antibody (ready-to-use, #K5007, Dako, Glostrup, Denmark) for 1hour at room temperature. The signal was visualized with 3,3-diaminobenzidinetetrahydrochloride (Dako).
Immunostaining for antineuronal nuclei (NeuN), GFAP, OX-42, and Ki67, the specific markers of neurons, astrocytes, microglias, and proliferating cells, respectively, were performed using mouse NeuN (1:1000, #MAB377, Millipore, Bedford, MA, USA), mouse anti-GFAP (1:1000, #610565, BD Biosciences, San Jose, CA, USA), mouse anti-rat CD11b (OX-42, 1:50, #CBL1512, Millipore), and mouse anti-ki67 (1:200, #MAB4190, Millipore), respectively, following the same protocol as described above. Negative-control sections were incubated with phosphate-buffered saline instead of the primary antibody, and showed no positive signal.
At 7 days after distal MCAO, the remaining four rats in each group were killed to measure the thalamic Aβ by enzyme-linked immunosorbent assay (ELISA). Under deep anesthesia, the rats were perfused intracardially with 50mL ice-cold 0.9% saline. The left thalamus was immediately isolated on ice, frozen and stored at −80°C, then homogenized separately at 5mL/g in ice-cold homogenization buffer (containing 50mmol/L Tris, pH 7.4, 150mmol/L NaCl, 1% Triton X-100, 0.6% sodium dodecyl sulfate), with complete protease inhibitor cocktail tablets (Roche, Mannheim, Germany). The resulting lysate was centrifuged at 250,000g for 30minutes at 4°C. The supernatant was collected and stored at −80°C.
The total Aβx-42 level was quantified by an established sandwich ELISA, using the colorimetric BetaMark Aβx-42 ELISA kit (#SIG-38956, Covance) following the manufacturer's instruction. The Aβx-42 ELISA kit is specific for the x-42 isoforms of Aβ. Levels of Aβx-42 were calculated by comparison with a standard curve generated by the Aβx-42 standard. The standard curve was linear over the range 0 to 250pg/mL with R2=0.98 to 0.99. Each Aβ standard was run in duplicate and each experimental sample in triplicate. Serial dilutions of samples were used for measurements in the linear range. The Aβx-42 levels of left thalamus were expressed as pg/mg of wet tissue weight.
Histological sections were examined by one author who was not aware of the experimental protocol. Images were captured under a microscope (BX51; Olympus, Tokyo, Japan) equipped with a DP70 digital camera and analyzed by image analysis software (Image pro plus 4.5, Media Cybernetics, Silver Spring, MD, USA). For Nissl staining or immunostaining, serial sections between Bregma −2.4 and −4.4mm were selected at intervals of every sixth sections from each rat for quantification. For quantification of Nissl-stained sections, the number of intact neurons with intact membranes and nuclei (Dihne et al, 2002) within the ipsilateral VPN of thalamus was counted using Image pro plus image analysis software in that area within three nonoverlapping fields (425 × 320μm2) under × 400 magnification (Wang et al, 2007), and was expressed as the average number of intact neurons per field on each section. Meanwhile, for quantification of immunostained sections, the number of NeuN+, GFAP+, OX-42+, or Ki67+ cells within the ipsilateral VPN was counted in the same way, and was expressed as the average number of NeuN+, GFAP+, OX-42+, or Ki67+ cells per field on each section. At this time point, the ipsilateral thalamic atrophy after MCAO was not obvious, and the cell number of each field may not be affected by the shrinkage of thalamus. For quantification of thalamus Aβ burden, the area of Aβ and the ipsilateral thalamus region were recorded using Image pro plus image analysis software at × 40 magnification, and the percentage of thalamus region covered by Aβ immunoreactivity was used to measure Aβ burden (Dodart et al, 2002).
All data were presented as mean±s.e.m. and analyzed using one-way analysis of variance followed by Bonferroni test for multiple comparisons, except that thalamic Aβ burden detected by immunohistochemistry was expressed as median (quartile range) and compared using Kruskal–Wallis test followed by Mann–Whitney U-test. The correlation coefficients between the number of neurons within the ipsilateral VPN and the right forepaw sensory performance or the thalamic Aβ burden were calculated by Pearson's correlation coefficients. All data were analyzed by SPSS 13.0 for windows (Abacus concepts Inc., Chicago, IL, USA). Statistical significance was defined as P<0.05.
At 7 days after distal MCAO, the mean time to remove the stimulus from the right forepaw increased significantly in MCAO, vehicle, and DAPT groups (30.3±3.9, 27.9±4.4, 22.9±4.8seconds, respectively) compared with the sham-operated group (6.4±1.6seconds, all P<0.01). However, the mean time to remove the stimulus in the DAPT group decreased significantly when compared with the vehicle group (P<0.05).
Cerebral cortex infarct was evident in all MCAO rats. At 7 days after distal MCAO, the relative infarct volume was 15.64%±2.15% in MCAO, 14.69%±2.08% in vehicle, and 13.08%±3.20% in DAPT group. Although the lesions size was 19.1% bigger in MCAO versus DAPT-treated animals, the lesion size was only 12% bigger in vehicle versus DAPT-treated animals. There were no significant differences among the three groups (all P>0.05).
Strong Aβ3-16 immunoreactivity was revealed in the ipsilateral thalamus at 7 days after distal MCAO (Figures 1B and 1F), but not in the sham-operated group (Figures 1A and 1E). The Aβ3-16 expression was localized as diffuse small dots in the VPN (Figures 1F–1H). The DAPT treatment significantly decreased thalamic Aβ (Figures 1D and 1H). Quantitative immunohistochemistry revealed that thalamic Aβ3-16 burden was reduced by 50.7% in the DAPT group when compared with the vehicle group (Figure 1I; P<0.001). The ELISA showed that DAPT treatment led to 41.5% reduction of Aβx-42 in the left thalamus when compared with the vehicle group (Figure 1J; P<0.001).
Nissl staining showed that many abnormal neurons with shrunken cytoplasm, pyknotic nuclei, and reduction of Nissl substance are within the ipsilateral VPN at 7 days after distal MCAO (Figure 2B1), but not in sham-operated group (Figure 2A1). Intact neurons were decreased significantly within the ipsilateral VPN in MCAO, vehicle, and DAPT groups compared with the sham-operated group (Figures 2B1–2E1; all P<0.01). However, the number of intact neurons within the ipsilateral VPN was significantly greater in the DAPT group than that in the vehicle group (Figure 2E1; P<0.05).
In agreement with the findings of Nissl staining, NeuN+ cells within the ipsilateral VPN were reduced significantly in MCAO, vehicle, and DAPT groups at 7 days after distal MCAO compared with the sham-operated group (Figures 2A2–2E2; all P<0.01). However, the number of NeuN+ cells within the ipsilateral VPN was significantly greater in the DAPT group than that in the vehicle group (Figure 2E2; P<0.05).
In contrast, GFAP+ astrocytes, OX-42+ microglias, and Ki67+ cells within the ipsilateral VPN were increased significantly in MCAO, vehicle, and DAPT groups at 7 days after distal MCAO compared with the sham-operated group, respectively (Figures 2A3–2E5; all P<0.01). Nevertheless, there were less GFAP+ astrocytes, OX-42+ microglias, and Ki67+ cells within the ipsilateral VPN in the DAPT group compared with the vehicle group (Figures 2E3–2E5; P<0.05).
The mean time to remove the stimulus from the right forepaw at 7 days after MCAO was negatively correlated with the number of neurons in the ipsilateral VPN detected by both Nissl staining and immunostaining for NeuN (Table 1). Meanwhile, the number of neurons in the ipsilateral VPN was negatively correlated with the thalamic Aβ burden (Table 1).
In this study, we found that abnormal accumulation of endogenous Aβ coexisted with secondary damage, as evidenced by neuronal loss, glial activation and proliferation, within the ipsilateral VPN of the thalamus at 7 days after permanent distal MCAO. Moreover, treatment with γ-secretase inhibitor was associated with reduction of the thalamic Aβ accumulation, attenuation of the thalamus damage, and the sensory functional improvement. These findings demonstrate that abnormal accumulation of Aβ may be associated with the secondary thalamic damage and sensory impairments after cerebral cortical infarction in hypertensive rats.
Hypertension has long been recognized as the most important risk factor for stroke (Goldstein et al, 2001; Sacco et al, 2006). Therefore, hypertensive rats are better to mimic clinical conditions when compared with normotensive rats. In this study, distal MCAO in stroke-prone renovascular hypertensive rats, which are hypertensive animals without hereditary deficit (Zeng et al, 1998b), is an appropriate animal model to investigate the secondary thalamic damage. In the model, the infarction is highly reproducible in size and localization in the ipsilateral frontoparietal cortex because the hypertensive rats have poor collateral circulation compared with normotensive rats (He et al, 2007; Ling et al, 2009; Wang et al, 2007; Zeng et al, 1998a), but the thalamus is not directly affected because of its blood supply through the posterior cerebral artery. Therefore, any thalamic damage is secondary to the primary cortical infarction rather than a direct injury owing to the distal MCAO (Ling et al, 2009). We observed that delayed secondary damage was prominent in the VPN but mild in the substantia nigra (data not shown).
Secondary damage (Dihne et al, 2002; Iizuka et al, 1990; Kataoka et al, 1989; Ling et al, 2009; Watanabe et al, 1998) and deposition of Aβ in the thalamus (Clarke et al, 2007; Hiltunen et al, 2009; Makinen et al, 2008; van Groen et al, 2005) has been reported after transient MCAO in rats. In this study, we found that both of them coexisted in the ipsilateral thalamus after permanent distal MCAO in stroke-prone renovascular hypertensive rats, indicating a positive association between Aβ deposits and neuronal loss in the thalamus. The Aβ is well known to be neurotoxic (Lambert et al, 1998; Yankner et al, 1990; Yankner and Lu, 2009), and it is crucial for the pathogenesis of Alzheimer's disease. To examine whether the Aβ deposits were harmful to the thalamus, DAPT was used as the intervention to reduce thalamic Aβ. The DAPT is a functional γ-secretase inhibitor and can effectively decrease the levels of Aβ (Comery et al, 2005; Dovey et al, 2001; El Mouedden et al, 2006; Lanz et al, 2003). In this study, we found that a single-acute administration of DAPT (50mg/kg) reduced the thalamic Aβ up to 50%, which is consistent with these previous findings observed in the cortex, cerebrospinal fluid, and plasma (Dovey et al, 2001; El Mouedden et al, 2006; Lanz et al, 2003). One of the novel findings of our study was that neuronal loss within the ipsilateral VPN was significantly reduced in parallel to the reduction of Aβ deposits in DAPT-treated rats. In addition, the number of neurons in the ipsilateral VPN was negatively correlated with the thalamic Aβ burden. These results strongly indicate that Aβ deposits in the thalamus were indeed associated with the secondary thalamic damage.
The mechanisms underlying secondary thalamic neurodegeneration after focal cerebral infarction are quite complex, including retrograde and anterograde fibers degeneration (Dihne et al, 2002; Ross and Ebner, 1990; Tamura et al, 1991), inflammation (Block et al, 2005; Rupalla et al, 1998), apoptosis (Soriano et al, 1996), inhibition of axonal regeneration by Nogo-A (Wang et al, 2007), and oxidative DNA damage (He et al, 2007). In this study, we demonstrated that Aβ deposits in the thalamus are associated with the secondary thalamic damage. However, a question arises whether attenuation of thalamic lesion is specific to reduction of Aβ deposits by DAPT. Previously, we have reported that several neuroprotective agents, such as ebselen (a glutathione peroxidase mimic) and NEP1-40 (a Nogo-66 receptor antagonist peptide), can reduce the secondary neuronal damage in the ipsilateral VPN after distal MCAO in rats without reduction of infarct volume (He et al, 2007; Wang et al, 2007). The Aβ is a neurotoxin and has a pivotal role in the pathogenesis of Alzheimer's disease. The Aβ can induce neuronal death by inducing apoptosis (Estus et al, 1997), mediating inflammatory injury (Schlachetzki and Hull, 2009; Wyss-Coray, 2006), generating oxidative stress (Lustbader et al, 2004), and disturbing the homeostasis of Ca2+. Interestingly, we found that glia activation and proliferation within the ipsilateral VPN were significantly reduced in parallel to the reduction of Aβ deposits in DAPT-treated rats, suggesting that inflammatory reaction may be a downstream event of Aβ deposits, and Aβ deposits may mediate the secondary thalamic neurodegeneration at least in part through inflammatory mechanisms. Inflammatory reaction has an important role in the pathophysiology of secondary damage. Previously, neuroprotective agents such as ebselen and NEP1-40 have been shown to reduce secondary damage through inhibiting inflammatory reaction. Therefore, Aβ deposits may be one of critical factors responsible for the secondary thalamic neurodegeneration after MCAO. To confirm the exact role of Aβ in the secondary thalamic damage after MCAO, conditional knockout of the amyloid precursor protein gene or gene silencing strategies might be needed in future studies.
It has been reported that thalamic damage is not correlated with early neurologic impairment, but with late sensory deficit after MCAO (Freret et al, 2006). In our study, the infarct was restricted to the ipsilateral somatosensory cortex, and the thalamus was not directly affected after MCAO. Obviously, the sensory deficit appeared at early stage is related to the ischemia of somatosensory cortex. However, DAPT treatment significantly attenuated the sensory impairments without reduction of the cortical infarct volume, suggesting that cortical infarct may not be responsible for DAPT-mediated improvement in sensory impairments. Interestingly, the sensory deficit had a strong correlation with the secondary thalamic damage, and reduction of the secondary thalamic damage by DAPT significantly attenuated sensory impairments. Therefore, we speculate that the secondary thalamic damage may be also involved in the early sensory deficit after cortical infarction. However, it should be noted that the functions of the thalamus are highly complex, including sensation, language, and cognition (Schmahmann, 2003). In this study, we only investigated the association between the sensory impairments and secondary thalamic damage. Further study is needed to address the secondary thalamic damage in other neurologic function deficits, such as cognitive deficits and to evaluate the potential significance of thalamic Aβ deposits in stroke patients. Furthermore, it is necessary to assess whether reduction of Aβ by DAPT, or other γ-secretase inhibitors, can improve long-term functional and histological outcomes. Moreover, it is highly imperative to elucidate the mechanisms responsible for the Aβ deposition.
The authors declare no conflict of interest.
This study was supported by grants from the National Natural Science Foundation of China (Nos. 39940012, 30271485, and 30770764), China Medical Board of New York Inc. (No. CMB00-730), and the fund on collaboration study for First Affiliated Hospital and Life Science Institute in Sun Yet-Sen University (2006).