Accelerated infarct development, cytogenesis and apoptosis following transient cerebral ischemia in aged rats
- First Online:
- Cite this article as:
- Popa-Wagner, A., Badan, I., Walker, L. et al. Acta Neuropathol (2007) 113: 277. doi:10.1007/s00401-006-0164-7
Old age is associated with a deficient recovery from stroke, but the cellular mechanisms underlying such phenomena are poorly understood. To address this issue, focal cerebral ischemia was produced by reversible occlusion of the right middle cerebral artery in 3- and 20-month-old male Sprague–Dawley rats. Aged rats showed a delayed and suboptimal functional recovery in the post-stroke period. Using BrdU-labeling, quantitative immunohistochemistry and 3-D reconstruction of confocal images, we found that aged rats are predisposed to rapidly develop an infarct within the first few days after ischemia. The emergence of the necrotic zone is associated with a high rate of cellular degeneration, premature accumulation of proliferating BrdU-positive cells that appear to emanate from capillaries in the infarcted area, and a large number of apoptotic cells. With double labeling techniques, we were able to identify, for the first time, over 60% of BrdU-positive cells either as reactive microglia (45%), oligodendrocyte progenitors (17%), astrocytes (23%), CD8+ lymphocytes (4%), or apoptotic cells (<1%). Paradoxically, despite a robust reactive phenotype of microglia and astrocytes in aged rats, at 1-week post-stroke, the number of proliferating microglia and astrocytes was lower in aged rats than in young rats. Our data indicate that aging is associated with rapid infarct development and a poor prognosis for full recovery from stroke that is correlated with premature cellular proliferation and increased cellular degeneration and apoptosis in the infarcted area.
KeywordsStrokeAgingRecoveryIschemiaRatBrdUGFAPOligodendrocytesMicrogliaCD8+ lymphocytesCytogenesisVascular tree
Studies of stroke in experimental animals have demonstrated the neuroprotective efficacy of a variety of interventions, but most of the strategies that have been clinically tested failed to show benefit in humans. One possible explanation for this discrepancy between laboratory and clinical investigations is the role that age plays in the recovery of the brain from insult. Although it is well known that aging is a risk factor for stroke , the majority of experimental studies of stroke have been performed on young animals, and therefore may not fully replicate the effects of ischemia on neural tissue in aged subjects [14, 44, 45, 60, 64]. In this light, the aged post-acute animal model is clinically most relevant to stroke rehabilitation and cellular studies [6, 11, 35, 37].
Following ischemic stroke, microglia and astrocytes increase in number and reactivity in the infarcted region [36, 58]. In addition, oligodendrocyte progenitors proliferate in the outer border of the infarct, and contribute a number of components to the glial scar, particularly chondroitin sulfate proteoglycans (such as NG2, phosphacan, brevican, versican and neurocan) that have been hypothesized to impede the regeneration of tissue ([2, 13; for review see ). However, we do not know in detail the molecular mechanisms underlying the decrease in brain plasticity with increasing age. Senescent rats generally have low expression levels of plasticity-associated genes such as microtubule-associated protein 1B (MAP1B), c-fos, tissue plasminogen activator, and neurofilament-68 [42, 46, 47, 50, 54]. Following insult to the brain, old rats are capable of upregulating gene expression, but the response is often blunted and temporally uncoordinated [1, 3, 4, 26, 42, 54, 56, 59, 66]. Moreover, there appear to be substantial age-dependent differences in the transcriptional profile of distinct brain regions that may account, in part, for the increased vulnerability of the aged brain to traumatic brain injury .
In a rat model of transient cerebral ischemia, we found that the interplay between plasticity-promoting factors such as MAP1B and MAP2, and neurotoxic factors such as C-terminal β-amyloid precursor protein (βAPP), may influence the time-course of functional recovery after stroke [7, 44, 45]. These age-associated alterations in the molecular response to stroke correlate with the premature formation of the inhibitory glial scar, which can act to limit neurorestoration and behavioral recovery in older rats . In the present study, we show that the increased susceptibility to stroke in aged rats is associated both with premature cellular proliferation and with an increased number of degenerating and apoptotic cells in the infarcted area.
Materials and methods
Young (3–4 months, N = 12 per age group and time point) and aged (18–20 months, N = 18) male Sprague–Dawley rats, bred in-house, were used. Body weights ranged from 290 to 360 g for the young rats and from 420 to 500 g for the old rats. The group sizes for the aged rats were larger (N = 18) to compensate for the higher post-ischemic mortality rate. The control groups (sham-operated rats) consisted of age-matched animals (N = 10 per age group and time point). To study the phenotype of newly generated cells after stroke, additional rats (N = 3 per age group and time point) were included so that the total number of rats used in this study was of 224.
The experiments reported in this study were conducted in accordance with the statement regarding the care and use of animals and were approved by a federal animal care committee.
To evaluate changes in neurological function associated with ischemia, the rats were tested behaviorally before and after surgery. All testing was performed between 9 and 11 a.m., as described in detail in a prior publication . Results obtained before surgery were used to define 100% functionality for each animal on each test, and functional recovery was expressed as percent recovery relative to this value over a 28-day period. In short, the rats were tested for (1) their ability to traverse a rotating (4 rpm) horizontal rod (beam-walking; Rotarod). The rats were tested daily in the first week, and then every other day through day 28 post-surgery. The time taken for the rat to traverse the rotating cylinder and join a group of rats visible at the finish line was measured; (2) limb-placement while approaching the edge of a table. To assess the symmetry of limb movement, the rat was gently held in the air by the tail and limb movements scored as follows: 3—all four limbs extended symmetrically; 2—limbs on left side extended more or less slowly than those on the right; 1—limbs on left side showed minimal movement; and 0—forelimb on left side did not move at all . The rats were tested daily up to day 14 post-surgery; and (3) the ability of the animal to maintain its position at a given angle on an inclined plane . The relative angle at which the rat could no longer maintain its position was taken as a measure of functional impairment. This test was conducted once before surgery and daily thereafter.
Blood flow through the middle cerebral artery (MCA) was transiently interrupted as previously described [6, 44]. Throughout surgery, anesthesia was maintained by spontaneous inhalation of 1–1.5% halothane in a mixture of 75% nitrous oxide and 25% oxygen. Body temperature was controlled at 37°C by a Homeothermic Blanket System (Harvard Apparatus). The anesthetized animals were immobilized in a supine position and the tail artery catheterized to enable the continuous measurement of blood pressure and the withdrawal of blood samples for determination of pH and blood gasses (Blutgassystem IL 1620, Instrumentation Laboratory, Munich), as well as arterial glucose levels (Omnican7 Balance, B. Braun, Melsungen). The right MCA was slowly lifted with a tungsten hook attached to a micromanipulator (Maerzhaeuser Precision Micro-manipulator Systems, Fine Science Tools) until blood flow through the vessel was completely interrupted. To prevent continued perfusion of the target region by arterial collaterals, both common carotid arteries were then occluded by tightening pre-positioned thread loops. Local changes in blood flow were monitored using a laser Doppler device (Perimed, Stockholm, Sweden). Sham operations were conducted by removing the MCA hook within 60 s.
After 70 min, the middle cerebral artery and the common carotid arteries were re-opened, allowing full reperfusion of the brain. Subsequent to survival times of 1, 3, 7, 14, or 28 days, the rats were deeply anesthetized with 2.5% halothane in 75% nitrous oxide and 25% oxygen, and perfused with neutral buffered saline followed by buffered, 4% freshly depolymerized paraformaldehyde. The brain was removed, post-fixed in 4% buffered paraformaldehyde for 24 h, cryoprotected in 20% sucrose prepared in 10 mmol/l phosphate buffered saline, flash-frozen in isopentane and stored at -70°C until sectioning.
Determination of infarct volume
To assess the size of the infarct induced by transient focal ischemia, every tenth section was stained with Fluoro-Jade B, a marker of cellular degeneration, or every 20th section was stained with NeuN, a marker of neuronal nuclei. In previous studies, we have found that the disappearance of NeuN is a reliable indicator that neurons have been lost [6, 7]. Images of the stained sections were taken to cover the entire infarcted area, which was then calculated as the sum of partial areas using Scion image analysis software. Integration of the resulting partial volumes gave the total volume of the ipsilateral hemisphere along with the total volume of the cortical infarct; lesion volume was then expressed as percent of the hemispheric volume. The two staining methods yielded essentially the same results.
To label newly generated cells, rats were given two injections of bromodeoxyuridine (BrdU; 50 mg/kg body weight, i.p.; Sigma), 24 and 8 h before sacrifice. The rats were perfused transcardially 8 h after the last injection, so that for each time point, BrdU-labeled cells range in age from 8 to 24 h. To study the phenotype of newly generated cells that have had time to go through a maturation process, additional rats (N = 3 per age group and time point) were injected daily in the first week post-stroke, and every other day in the second week post-stroke, for a total period of 2 weeks.
Detection of degenerating neurons with Fluoro-Jade B
Cryostat sections were immersed as floating sections in 0.06% potassium permanganate for 10 min to block background staining. After an additional rinse in water, the sections were stained for 20 min in 0.004% Fluoro-Jade B, 0.0002% DAPI (4′,6–diamidino-2-phenylindole), and 0.1% acetic acid. The slides were rinsed in water three times for 1 min, dried, soaked in xylene, and coverslipped with DPX . In order to make certain that the Fluoro-Jade B staining is restricted to the infarct core, sections (n = 3 per age group and time point) that were stained for neuronal viability with NeuN also were stained for degenerating neurons with Fluoro-Jade B.
Terminal deoxynucleotidyl transferase-mediated UTP nick-end labeling (TUNEL)
To detect cells undergoing apoptosis, we adapted a recently described procedure  employing TUNEL on free-floating sections and the Apoptag in situ cell death detection kit (Intergene, Purchase, NY). Sections were incubated with equilibration buffer for 30 min at room temperature, followed by TdT-reaction solution diluted 1:1 with TUNEL dilution buffer (Roche Diagnostics, Mannheim, Germany). After 2 h at 37°C, the reaction was stopped by the addition of Stop Buffer. Alkaline phosphatase-conjugated sheep anti-digoxigenin-Fab fragment, diluted 1:1,000 in blocking buffer, was then applied. After 15 h of incubation at 4°C, sections were washed and stained at 37°C in the dark for 30 min in a chromogen solution consisting of nitro-blue tetrazolium (NBT), 5-bromo-4-chloro-3-indolyl-phosphate (BCIP), and 250 μg/ml levamisole. Finally, the sections were mounted on slides, air-dried, and coverslipped. To determine non-specific labeling, some sections were incubated without the Klenow fragment enzyme.
A summary of the primary- and secondary antibodies used for immunostaining
Circulating blood cells;
Hemotopoietic progenitor cells
1:500; Santa Cruze
Apoptag kit; Intergenei
Cellular phenotyping by immunofluorescence
Mitotically active microglia were phenotyped by using a combination of mouse anti-rat antibody recognizing a cytoplasmic determinant of activated brain macrophages ED1 (clone MCA341), conjugated with FITC in conjunction with the rat anti-BrdU antibodies (Table 1).
Circulating blood cells and hematopoietic progenitor cells
The following antibodies were used to identify circulating blood cells with BrdU-positive nuclei: mouse anti-rat CD4, a marker for T helper cells; mouse anti-rat CD8, a marker for cytotoxic T cells; mouse anti-rat CD11b, which reacts with the CR3 complement receptor (C3bi) expressed on rat monocytes, granulocytes, macrophages, dendritic cells, NK cells, and a subset of lymphocytes; goat anti-rat CD34, a marker for hematopoietic progenitor cells and vascular endothelial cells; mouse anti-CD45, a marker for rat leukocyte common antigen; rabbit anti-myeloperoxidase, a marker for immature myeloid cells; and rabbit anti-rat polymorphonuclear leukocytes (PMN).
Oligodendrocyte progenitors and astrocytes
Oligodendroglia were double-immunolabeled with rabbit anti-NG2-specific antibodies (1:3,000, Chemicon, Temecula, CA) and rat anti-BrdU-specific antibodies. Astrocytes were double-immunolabeled with rabbit anti-GFAP-specific antibodies and rat anti-BrdU-specific antibodies. The antigen–antibody complexes were visualized with donkey anti-rabbit rhodamine-conjugated antibodies (1:4,000) and donkey anti-rat FITC-conjugated antibodies (1:2,000), respectively (Table 1).
Basal lamina and BrdU-positive cells
To determine the relationship between proliferating cells and the vascular basal lamina, sections were double-immunolabeled with rabbit anti-laminin-specific antibodies and rat anti-BrdU-specific antibodies. The antigen–antibody complexes were visualized with a mixture of donkey anti-rabbit Cy5-conjugated antibodies (1:2,000) and donkey anti-rat rhodamine-conjugated antibodies (1:5,000).
Brain capillaries and BrdU-positive cells
To determine the relationship between proliferating cells and the vascular wall, sections were double-immunolabeled with mouse anti-rat endothelial cell antigen (RECA)-specific antibodies and rat anti-BrdU-specific antibodies. The antigen–antibody complexes were visualized with a mixture of donkey anti-mouse FITC-conjugated antibodies (1:3,000) and donkey anti-rat rhodamine-conjugated antibodies (1:5,000).
Cellular phenotyping by dual chemical staining
BrdU- and TUNEL-positive cells
In addition to proliferating cells, apoptotic cells also can incorporate BrdU . To detect apoptotic cells having BrdU-labeled nuclei, sections were first labeled with DIG-dUTP followed by alkaline phosphatase-conjugated sheep anti-digoxigenin-Fab fragment as described above. For BrdU detection, free-floating sections were pre-treated as described above and BrdU-positive nuclei were visualized by the horseradish peroxidase reaction with 0.05% diaminobenzidine (DAB) and 0.03% H2O2.
The number of labeled cells at the reperfusion times of 3, 7, 14 and 28 days was determined by counting cells on every tenth section in systematic random series across the entire infarcted volume. To this end, a sequence of confocal counting images of 161 × 242 × 25 μm, spaced 0.1 μm apart across a 25-μm-thick section and covering 30% of the infarcted area, was taken for fluorescently labeled cells . The resulting images were loaded into the 3-D analysis software “Volocity” (IMPROVISION, Coventry, UK) and computed using a Macintosh computer. The obtained images clearly indicated that most BrdU-labeled cells occurred, depending on the post-stroke stage, in dense, clonal-like clusters such that individual cells could not readily be distinguished, and hence an accurate count based on stereological methods could not be obtained (see Fig. 2g–i). However, in many instances, the 3-D reconstruction method allowed us to differentiate closely spaced cells and thus to count all BrdU-labeled cells within the 3-D spaces with reasonable accuracy and reliability. Nevertheless, at time points where proliferation was at its maximum, such as 7 days’ post-stroke, counting was less precise because of the many mitotic cells with a clonal-like appearance. The relative mean number of BrdU-positive cells was then calculated per group, time point, and age by multiplying the number of cells per section times 3.3 (the counting boxes that were quantitated covered one third of the area of each section) times the section interval of 10.
Counting of double-labeled cells
Because colocalization in one confocal plane sometimes may be misleading as to the number of cells colocalizing, we counted double-labeled cells by 3-D reconstruction as described above. By rotating the 3-D image, we were able to determine precisely the number of colocalized cells. The co-localized cells were given as percent of the average total number of cells within a counting cluster.
Counting of cells after chemical staining
The number of BrdU-, TUNEL-positive and dual-labeled (BrdU + TUNEL-positive) cells at the 3-, 7-, and 14-day reperfusion times was obtained by counting the cells in area units measuring 250 × 250 μm using a “random-systematic” protocol (random start point for a systematic series of every 10th section through the infarcted volume) using the NIH software PC Image for particle counting (NIH, Bethesda, MD). To eliminate false-positive particles, only cells and nuclei in a size range that included the smallest to the largest cells were enumerated. Furthermore, the optical dissector method was modified in that cells in sharp focus in the uppermost focal plane were not counted . To eliminate false-positive particles, only nuclei in a size range that included the smallest to the largest cells were enumerated. An integration of (BrdU + TUNEL-positive) cells was achieved by multiplying the number of cells per section times 3.3 (the counting boxes that were quantitated covered one-third of the area of each section) times the section interval . The co-localized cells were given as percent of the average total number of cells within a counting cluster.
For light microscopy, a Nikon Eclipse microscope (Duesseldorf, Germany) was used. Images (768 × 1,024 pixels) were captured electronically using a CCD camera (Optronics). The digital images were arranged and labeled using Adobe Photoshop and printed with a Kodak XLS 8000 digital printer. For a group of micrographs, the camera setting for exposure, gain, and contrast enhancement was the same. Confocal analysis of sections was done using a Nikon Eclipse microscope equipped with a laser device from Visitech (Munich, Germany). Fluorescence signals were detected using a Nikon E800 microscope at excitation/emission wavelengths 490/508 nm (FITC, green), 570/590 nm (rhodamine, red), and 650/668 nm (Cy5, blue).
The main effects of age, time, and manipulation (stroke vs. sham), as well as interactions, were evaluated by ANOVA followed by Tukey post hoc analyses using SPSS software (SPSS Inc., Chicago, IL). The level of significance was set at P ≤ 0.05, two-tailed test.
To mitigate problems of feeding in aged animals during the first 3 days’ post-stroke, we fed them with moistened, soft pellets. Nevertheless aged animals do lose some weight in the first week following stroke. The mortality rate was higher for aged rats (22%) than for younger rats (13%) during this time. For this reason, the initial group size was enlarged for aged rats to ensure comparable group sizes for analysis of the behavioral and histological outcomes. As previously reported, the hematologic parameters showed some age-associated differences, notably in blood pressure, but the differences were not statistically significant in this model .
Aged animals perform poorly in the first week post-stroke
The beam-walking task (Rotarod) assesses fine vestibulomotor function in the MCA occlusion model . After an abrupt decline in performance on the Rotarod at day 1 post-stroke, young rats began an improvement in performance that culminated in full recovery by day 12 post-stroke (Fig. 1a, filled circles). In contrast, the recovery of Rotarod performance by aged rats did not begin until day 3–4 post-stroke (Fig. 1a, open circles) and reached a maximum of only 70% by day 15 (Fig. 1a, filled circles).
Similar patterns of functional recovery for young and aged rats were observed for the other behavioral tests used. After an abrupt decline in performance on the symmetry of limb placement task at day 1 post-stroke, young rats began a linear increase in performance that culminated in full recovery by day 12 post-stroke (Fig. 1b, filled circles). The recovery in aged rats did not begin until day 3-4 post-stroke and reached about 85% of original limb placement symmetry (Fig. 1b, open circles) as compared to sham-operated rats (Fig. 1b, filled triangles). Aged rats recovered poorly in the inclined plane test (Fig. 1c, open circles), and the slope of recovery barely increased with time (Fig. 1c, open circles), while young rats recovered linearly up to 100% by day 14 (Fig. 1c, filled circles).
Infarct development and neuronal degeneration are accelerated in aged animals
The development of the infarct core was visualized by two complementary methods: Fluoro-Jade B, a marker for degenerating cells, and immunohistochemistry for NeuN, a sensitive indicator of neuronal viability .
Fluoro JadeB-staining confirmed that young rats had very few obviously degenerating neurons in the infarcted area at day 3 (Fig. 2g, lower panel). The number of degenerating neurons then increased rapidly through day 7 (Fig. 2h, lower panel) and reached a maximum at days 7–14 (fourfold vs. day 3; P < 0.001) (Fig. 2i, lower panel). Thereafter, the number of degenerating neurons declined progressively (not shown).
In contrast to young animals, on day 3, the necrotic zone of aged rats lacked NeuN immunopositivity in 28% of the ipsilateral cortical volume (Fig. 2e, d; upper panel). The infarcted area continued to expand, and by day 7 reached 41% of the ipsilateral cortical volume. Thus, the development of the infarct was more rapid in aged rats, but by day 7, the cortical infarcts were similar in size in both age groups, i.e. 37 ± 5.1% of total cortical volume in young rats and 41 ± 3.9% in aged rats (Fig. 2f, upper panel).
Fluoro JadeB-staining showed that aged rats had an unusually high number of degenerating neurons in the infarct core already on day 3 (3.5-fold vs. young rats; P < 0.001 (Fig. 2j, i, lower panel). Interestingly, thereafter the number of degenerating neurons did not rise further in aged animals (Fig. 2l, lower panel), although the infarcted area continued to expand, so that by day 7 the numbers of degenerating neurons were almost the same in both age groups (Fig. 2i, lower panel). While Fluoro-Jade B-staining (green) was largely confined to the infarct core (IC), NeuN-like immunoreactivity (blue) was mostly restricted to the periinfarct zone (PI) (Fig. 2m, lower panel). By this criterion, the area most affected by stroke was the parietal cortex, with the motor cortex being only slightly less damaged.
Post-ischemic apoptosis is accelerated in aged rats
The aged rats had a twofold increase over young rats (P < 0.02) in the number of apoptotic cells in the infarct core already at day 3 (Fig. 3c, e, gray-filled bars). At day 7, however, the ratio was reversed, i.e. the apoptotic cells in aged rats (Fig. 3d) were outnumbered by the apoptotic cells in young rats by 1.7-fold (P < 0.05). At this stage in the aged rats, apoptotic cells were detected only in the periinfarcted area (Fig. 3d). Thereafter the number of apoptotic cells decreased progressively from day 7 (Fig. 3d, e, gray-filled bars) to day 14 (Fig. 3e, gray-filled bars). From day 14 onward, the number of apoptotic cells fluctuated around basal levels in both age groups (not shown).
Post-ischemic cellular proliferation is prematurely increased in aged rats
In contrast to the late appearance of proliferating cells in young rats, in the infarcted zone there was a highly significant, fourfold excess (P < 0.001) of BrdU-positive cells in old rats at 3-day post-stroke (Fig. 4b, e). This difference remained at 7-day post-stroke (Fig. 4d) (4.8-fold, P < 0.0001; Fig. 4e), at which point the number of BrdU-positive cells peaked in both age groups before abruptly declining to control levels by days 14–28 (Fig. 4e). By day 28, there were few dividing cells in the infarcted hemisphere of aged rats, but even at this point a small number of BrdU-positive cells appeared to be migrating through the vascular wall from the circulation (Fig. 4f, inset). It should be noted that, as a result of the surgical procedure, the sham-control animals also showed a moderate increase in BrdU-positive cells in the sham-operated hemisphere at all timepoints.
Proliferating cells in aged rats emanate from the vasculature
The 3-D reconstruction of BrdU-positive cells in the infarcted area of aged rats at the 3-day timepoint revealed that most of the BrdU-positive cells (red) were actually within the vascular wall (primarily capillaries) (Fig. 5b, arrows), and only a few cells were detected in the extra-vascular tissue, mainly in the vicinity of the blood vessels. Away from the infarct itself, in the subventricular zone of aged animals, however, proliferating cells that were BrdU-positive were not spatially related to the cerebral capillaries (shown in green) (Fig. 5b, inset). Between days 3 and 7, whereas the overall number of BrdU-positive cells increased in the infarcted area, the number of BrdU-positive cells associated with the capillaries actually decreased in the old rats (Fig. 5c, capillaries shown in green, arrow) although BrdU-positive cells (red) were still present near the basal lamina (blue) of the capillaries (Fig. 5c, inset).
Phenotyping of proliferating cells
Phenotypic analysis of BrdU-positive cells after stroke in the brains of young and aged rats
% Cells colabeled
% Cells colabeled
Circulating blood cells
A small fraction of proliferating cells—4% for the young rats and 3% for the aged rats (Table 2)—also were positive for the lymphocytic marker CD8 (Fig. 6c, arrows), but only at day 3. There was no evidence of colocalization of BrdU with the circulating blood cells and hematopoietic progenitor cells blood cell markers, polymorphonuclear leukocytes (PMN) (Fig. 6f), CD34, CD45, CD11b, CD4 or myeloperoxidase a marker for immature myeloid cells (Table 2).
After multiple BrdU treatments, double labeling of cells with astrocytic (GFAP, red) and proliferation (BrdU, green) markers at day 7 post-stroke showed, in the periinfarcted area, many reactive astrocytes (Fig. 7c, arrows), some being in a mitosis-like state (Fig. 7c, arrowheads). Although the astrocytic reaction to stroke in old rats was stronger than in young rats , there were, surprisingly, fewer newly generated (BrdU-positive) astrocytes in the periinfarcted region of aged rats (13%, Table 2) (Fig. 7d, arrows) compared to young rats (23%, Table 2) at the same time point.
Apoptotic cells also incorporate BrdU into their DNA
At day 7, apoptotic (TUNEL-positive) cells from young rats seldom incorporate BrdU (less than 1% of the counted cells, Table 2) (Fig. 7e). Unlike young rats, at an early post-stroke stage (3 days), aged rats had an appreciable number of apoptotic cells that also incorporated BrdU in the region of the infarct core (12% of the counted cells, Table 2) (Fig. 7f, arrows).
Aging is a major risk factor for stroke, yet the effects of senescence on the response of the brain to ischemia remain poorly understood. Our present findings show that aging is associated with poor functional recovery from mild cerebral ischemia, that the development of an infarct is accelerated in aged rats, and that the lesion is associated with the early appearance of proliferating, BrdU-labeled cells that emanate from the vasculature, as well as a large number of prematurely degenerating and apoptotic cells.
Following infarction, sensorimotor function was impaired in most animals, but the young rats began to recover after a brief period of only 1–2 days. In contrast, behavioral recovery in the aged animals did not commence until 2–4 days post-infarct. Furthermore, young animals fully recovered after 10–15 days, whereas the aged rats only recovered to about 70% of pre-stroke sensorimotor functionality during the same period. No further recovery was noted after day 15 post-stroke. We hypothesize that rapid infarct development in aged rats, together with the premature appearance of proliferating, degenerating and apoptotic cells, contributes to the long-lasting performance deficits seen in aged rats. The paradoxical increase in recovery with a concomitant increase in infarct size might be explained by the compensatory contribution of the healthy, contralateral hemisphere to sensorimotor recovery after stroke, especially in young rats [9, 18, 28, 31]. The differential cellular response to stroke in young and aged subjects is discussed below.
One factor that may contribute to the rapid development of the infarct in aged animals is the early, fulminant phagocytic activity of brain macrophages . Activated macrophages generate free radicals, the production of which is augmented in aged subjects following cerebral ischemia [16, 53]. A related consideration is that the vulnerability of brain tissue to traumatic injury , and to oxidative stress in particular, also increases with age [5, 17].
Another cellular event that contributes to early infarct development in aged rats is augmented apoptosis . Aging increases the susceptibility of the central nervous system to apoptotic events . The particular vulnerability of the aged brain to apoptosis is confirmed by our study; we found that aged rats had considerably more apoptotic cells 3 days after ischemia than did young rats. The large number of co-localized TUNEL/BrdU-positive cells in aged rats suggests that both apoptotic cells and cells undergoing repair are labeled by terminal deoxynucleotidyl transferase . Furthermore, it is important to note that the apoptotic cells include not just neurons, but also many glial cells.
Finally, our data show that not only are cells dying earlier in the infarct zone of aged rats, but there are also more newly generated cells at this time. Pulse-labeling with BrdU shortly before sacrifice revealed a dramatic increase in cells undergoing mitosis and/or repair  that appeared to be migrating through the cerebral vessels in the infarcted area. Significantly, the number of BrdU-positive cells in the infarcted hemisphere of aged rats at days 3 and 7 vastly exceeded that of young rats. In young rats, transiently labeled, mostly non-neuronal cells in the infarcted hemisphere peaked at day 7 post-stroke, in accordance with a previous report .
The reasons for the premature accumulation of BrdU-positive cells in the lesioned hemisphere of aged rats remain unknown. Blood parameters such as levels of O2, CO2 and blood pressure did not differ significantly in young and aged rats after mild ischemia  and are unlikely to be involved. Rather, we hypothesize that two age-associated factors could be important: (1) a decreased plasticity of the cerebral vascular wall  and (2) an early, precipitous inflammatory reaction to injury . The increased fragility of aged blood vessels due to decreases in the distensible components of the microvessels such as elastin  may lead, upon ischemic stress, to fragmentation of cerebral capillaries that would promote the leakage of hematogenous cells into the infarct area [32, 58].
Phenotypic identification of BrdU-positive cells at sites of injury is not a trivial undertaking, and most reports have focused on only one cell type. By analyzing a number of cell-specific markers, we are able to conclude that the majority (>60%) of the BrdU-positive cells are either apoptotic cells, activated macrophages, oligodendroglia progenitors, reactive astrocytes or (to a lesser extent) CD8-positive T cells. Earlier studies have shown that activated microglia and reactive astrocytes increase both in number and reactivity in normal older subjects. For example, the levels of glial fibrillary acidic protein (GFAP) and its mRNA increase in the astrocytes of rats as a function of age , and microglial cells also display an age-associated augmentation of reactivity in a variety of mammalian species [39, 63, 67].
Paradoxically, the number of proliferating astroglial cells is lower in aged rats than in young rats. We hypothesize that the chronically increased state of astroglial activation prior to stroke in the aged brain accelerates the formation of the glial scar in aged rats, whereas scar formation is delayed in younger animals by the later proliferation and activation of astrocytes.
Like neurons, oligodendrocytes are highly sensitive to ischemic injury . However, following damage to the CNS, oligodendrocyte progenitor cells become active, migrate to the lesion site and initiate the remyelination of denuded or regenerating axons, a process that peaks between 7 and 14 days’ post-injury [15, 25]. Although oligodendrocyte progenitor cells have been reported to increase in the ischemic penumbra after reperfusion in young rats relative to aged rats , we could not find such an age difference. However, it should be kept in mind that we counted the number of oligodendrocyte progenitors having a BrdU-positive nucleus, i.e. only proliferating oligodendrocyte progenitor cells.
Some dual-labeled cells in aged rats were both BrdU- and TUNEL-positive, confirming recent work showing that BrdU can be incorporated not only into mitotic cells, but also into cells undergoing DNA fragmentation and repair . Therefore, our results may reflect an early attempt by aged rats to repair stroke-induced DNA damage. Why degenerating cells in young rats do not incorporate BrdU to the same extent as old rats is an issue that warrants further study.
Whether any of the proliferating cells ultimately will become neurons is controversial, but recent publications indicate that this is usually not the case. For example, transplantation of bone marrow cells (pre-labeled with BrdU) into the ischemic area at 1 day after reperfusion of the adult rat brain indicates that, at 14-day post-stroke, only a small proportion of exogenous cells express either neuronal (NeuN) or astrocytic (GFAP) markers . Similarly, genetically modified hematopoietic cells enter the CNS and differentiate into microglia  or perivascular cells [22, 62] but do not adopt a neuronal phenotype in the infarcted area by 14–28 days after ischemia.
Factors that contribute to the early emergence of the infarct in aged rats include a high rate of cellular degeneration and the premature accumulation of proliferating and apoptotic cells in the infarcted area. Temporally, modulating the cytogenic response in the aged brain could engender a more permissive environment for the preservation of neurons and their connections .
This research was supported by a grant from Deutsche Forschungsgemeinschaft (DFG) to CK (Ke 599/1–1), by NIH RR-00165 (LCW) and by a grant from “Prof. Dieter Platt Stiftung” to APW.