Age-Dependent Exacerbation of White Matter Stroke Outcomes
A Role for Oxidative Damage and Inflammatory Mediators
Background and Purpose—Subcortical white matter stroke (WMS) constitutes up to 30% of all stroke subtypes. Mechanisms of oligodendrocyte and axon injury and repair play a central role in the damage and recovery after this type of stroke, and a comprehensive study of these processes requires a specialized experimental model that is different from common large artery, gray matter stroke models. Diminished recovery from stroke in aged patients implies that damage and repair processes are affected by advanced age, but such effects have not been studied in WMS.
Methods—WMS was produced with focal microinjection of the vasoconstrictor N5-(1-iminoethyl)-L-ornithine into the subcortical white matter ventral to the mouse forelimb motor cortex in young adult (2 months), middle-aged (15 months), and aged mice (24 months).
Results—WMS produced localized oligodendrocyte cell death with higher numbers of apoptotic cells and greater oxidative damage in aged brains than in young-adult brains. Increased expression of monocyte chemotactic protein-1 and tumor necrosis factor-α in motor cortex neurons correlated with a more distributed microglial activation in aged brains 7 days after WMS. At 2 months, aged mice displayed increased white matter atrophy and greater loss of corticostriatal connections compared with young-adult mice. Behavioral testing revealed an age-dependent exacerbation of forelimb motor deficits caused by the stroke, with decreased long-term functional recovery in aged animals.
Conclusions—Age has a profound effect on the outcome of WMS, with more prolonged cell death and oxidative damage, increased inflammation, greater secondary white matter atrophy, and a worse behavioral effect in aged versus young-adult mice.
Subcortical white matter stroke (WMS) constitutes up to 30% of all stroke subtypes,1,2 and has devastating clinical consequences. These infarcts produce focal deficits with incomplete recovery and are the second leading cause of dementia.3,4 Despite an abundance of preclinical literature featuring animal models of large artery, gray matter stroke, there have been few studies of white matter repair and recovery, a disparity which can be attributed to a lack of an effective animal model of WMS.
In addition, most studies using models of ischemic stroke have been performed on young animals. Ischemic stroke occurs mostly in older individuals, and recovery in aged patients is diminished compared with young patients.5,6 Aged animals differ from young animals in their neurophysiology, and pathophysiological features of brain ischemia differ across the lifespan.7,8 Furthermore, remyelination efficiency in white matter lesions decreases with age.9 Thus, studies using young animals to model ischemic stroke may have not fully estimated the effects of ischemia on brain tissue in aged subjects.
We have recently developed a model of subcortical stroke in white matter below the mouse forelimb motor cortex that models many aspects of this disease in humans. The model includes an induction of focal ischemia resulting in cell death and white matter destruction at the infarct core, and a peri-infarct zone of partial myelin, axon, and oligodendrocyte loss.10,11 This model has been adapted here to study how the damage and functional consequences after white matter ischemia differ between the young and aged brain.
All experiments were performed in accordance with National Institutes of Health animal protection guidelines and were approved by the University of California at Los Angeles Animal Research Committee. Thirty-two 3-month-old C57/BL6 mice (Jackson), twenty-four 15-month-old, and twenty-eight 24–month-old C57/BL mice (National Institute on Aging) were used in this study. Cohorts of animals (n=6–8 per group) survived to 1, 2, 7, and 56 days after stroke or sham procedure.
White Matter Stroke
White matter stroke was produced as described10 with modifications. The vasoconstrictor N5-(1-iminoethyl)-L-ornithine (27 mg/mL in sterile physiological saline; Millipore) was injected via micropipette through the cortex at an angle of 45° with a dorso-posterior to ventro-anterior injection path into the white matter underlying the forelimb motor cortex. This procedure promotes intense vasoconstriction at the site of injection, leading to focal ischemia.10,12 Three injections (each of 200 nL N5-(1-iminoethyl)-L-ornithine solution) were made in the following coordinates: A/P: +0.5, 0, −0.5; M/L: −0.96 (for all 3 injections); and D/V: −1.2, −1.15, −1.1. The pipette was left in situ for 5 minutes after injection to allow proper diffusion. Control animals underwent a sham procedure during which a craniotomy was done similarly to stroke animals, with no subsequent injections.
Tissue Processing, Microscopy, and Stereology
The tissue was processed as described previously10,11 (See detailed staining methods in the online-only Data Supplement). High-resolution confocal images and Z-stacks were acquired (Nikon C2 confocal system). Area measurements of the infarct core, Iba-1–positive cells, neutrophils, and cell nuclei tagged by terminal deoxynucleotidyl transferase dUTP nick-end labeling were stereologically quantified using the optical fractionator probe and neuroanatomical quantification software (Stereoinvestigator, MBF Bioscience; Williston, VT). Striatal axonal projections stained with NF200 were quantified with intensity profiles generated by ImageJ software (National Institutes of Health).
Mice were randomly allocated to treatment groups. All tests were analyzed blindly to stroke condition or age. The required number of animals per group was determined by a power analysis, with the expected variance and change in performance in the pasta matrix and grid-walking tasks predicted based on preliminary studies in young-adult animals. A group size of n=8 was predicted to be sufficient to detect a statistically significant result in ANOVA with α=0.05 and power >0.8. All data are expressed as mean±SEM. For cell quantification, axonal degeneration, white matter measurements, and behavioral testing, differences between age and treatment groups were analyzed using 1-way or 2-way ANOVA, with level of significance set at P<0.05, with Tukey honestly significant difference post hoc analysis (Excel and SigmaStat).
Infarct Volume, Oligodendrocyte Death, and Oxidative Damage After WMS
To assess the differences in initial tissue damage after WMS between the young and aged brain, animals from the 2 age groups were euthanized 24 and 48 hours after stroke. Infarct volume was similar 24 and 48 hours after WMS, and no difference was detected between young-adult and aged animals (Figure I in the online-only Data Supplement; P>0.3; n=8). Terminal deoxynucleotidyl transferase dUTP nick-end labeling staining 24 hours after stroke revealed extensive cell death throughout the infarct core in young-adult and aged mice (Figure 1A). In both age groups, the apoptotic cells were localized to the white matter lesion and were not detected in surrounding cortical areas, confirming the focal nature of this stroke type. The number of apoptotic cells did not differ significantly between young and aged animals at the 24-hour time point (Figure 1A and 1D; P>0.1; n=8). However, 48 hours after WMS, the number of apoptotic cells was significantly higher in aged animals compared with young animals, resulting in a total higher number of apoptotic cells detected in aged animals (Figure 1A and 1D; P<0.01; n=8; terminal deoxynucleotidyl transferase dUTP nick-end labeling–positive cells over 48 hours equaled 924±46 in young adults; 1166±52 in aged; P<0.05). Double immunostaining for terminal deoxynucleotidyl transferase dUTP nick-end labeling and the oligodendrocyte marker, Olig2, confirmed that cell death was primarily induced in oligodendrocytes (Figure 1C). Staining for 8-hydroxydeoxyguanosine revealed cells positive for this oxidative DNA adduct within the lesion. Similar to the observed pattern of apoptosis, there was no difference between the age groups in the number of cells displaying oxidative damage 24 hours after WMS, but the number was higher in aged animals 48 hours after stroke (Figure 1B and 1D; P<0.03; n=8).
Inflammatory Mediators and Reactive Cells After White Matter Stroke
The inflammatory response to stroke was assessed at 24 hours and 7 days, the latter being a time of maximal microglia/macrophage response in young adults with a related WMS.11 Ly6G (neutrophil)/Iba-1 (microglia/macrophage) staining 24 hours after WMS showed neutrophil infiltration in the infarct core and reactive microglia surrounding the lesion (Figure IIA in the online-only Data Supplement). The numbers of neutrophils and Iba-1–positive microglia were similar between the age groups (Figure IIB in the online-only Data Supplement; Iba-1, P>0.3; neutrophils, P>0.3; n=6). Seven days after stroke, a greater infiltration of activated microglia densely filled and surrounded the infarct core (Figure 2A). The density of Iba-1–positive cells in the lesion did not differ significantly between the age groups (Figure 2C; P>0.2; n=8). Interestingly, although the increase in Iba-1 staining after stroke in young animals was limited to the infarct area, in aged animals, there was a higher density of Iba-1–positive cells throughout the motor cortex compared with aged controls and young stroke animals (Figure 2B and 2D; control young-adult versus aged, P<0.05; WMS young-adult versus aged, P<0.005; control aged versus WMS aged, P<0.01; n=8).
To test whether upregulation of inflammatory mediators could have contributed to microglial activation in the cortex, expression of monocyte chemotactic protein-1 (MCP1) and the proinflammatory cytokine tumor necrosis factor-α (TNFα) was measured. Seven days after stroke, there was a robust MCP1 expression in reactive astrocytes in or around the infarct core (Figure IIIA in the online-only Data Supplement). The number of MCP1-positive astrocytes did not differ between age groups (Figure IIIB in the online-only Data Supplement; P>0.3; n=6). Interestingly, MCP1 could be transiently detected in a subset of motor cortex neurons only in aged brains 24 hours after stroke. Neuronal expression of MCP1 was not present in stroked young-adult animals or control animals of either age group, and was undetectable in stroked aged brains by day 7 (Figure 3C). At day 7, TNFα was expressed in neurons in cortical areas adjacent to the stroke lesion in both young-adult and aged mice (Figure 3A). When quantified in 3 regions at increasing distances from the infarct core, TNFα-positive neurons were more numerous and extended further away from the infarct core in aged brains compared with young-adult brains (Figure 3B; P>0.2 in region 0–300 μm; P<0.01 in regions 300–400 μm and 400–500 μm; young-adult WMS versus aged WMS; n=6).
Long-term Tissue Outcome of White Matter Stroke: Myelin Loss and Axonal Degeneration
Immunostaining for myelin basic protein revealed an area of myelin loss in the infarct core and a narrowing of the white matter medial to the stroke site at late time points after stroke (56 days; Figure 4). To quantify this volume change, the thickness of the white matter was measured at 10 equal increments from the midline (1) to the center of the infarct lesion (or equivalent point in controls: 10). A local decrease in the thickness of the white matter after stroke was observed in both young-adult and aged animals. However, it was markedly more pronounced in aged animals (Figure 4B; young-adult WMS versus control P<0.04 at interval #8; aged WMS versus control P<0.005 at intervals #6–9; WMS young-adult versus aged P<0.009 at intervals #6–9; n=8).
Immunostaining for the axonal marker NF200 revealed that the loss of myelin in the white matter was accompanied by loss of axonal projections from the cortex to the dorso-medial striatum after stroke. These axonal bundles include corticostriatal projections from the ipsilateral and contralateral cortex that play an important part in controlling motor programs and movement.18 Quantification of axonal staining in the region of corticostriatal projections (See detailed axonal measurement methods in the online-only Data Supplement) indicated a loss of axonal bundles at a late time point after stroke that was more pronounced in aged animals compared with young animals (Figure 5C; young-adult WMS, P<0.04 compared with control; aged WMS, P<0.01 compared with control; aged WMS, P<0.03 compared with young-adult WMS; n=8).
Functional Impairments After White Matter Stroke
To test whether the greater degenerative loss of white matter tracts and corticostriatal projections in aged WMS correlated with functional impairments, the baseline performance of young-adult and aged mice was assessed before and after stroke in 2 behavioral tasks that rely on forelimb motor control. A third, middle-aged group of animals was similarly tested in these tasks, to better evaluate whether the effect of age on functional impairment is graded with advancing age.
The pasta matrix reach task is specifically aimed at evaluating strength, reach capacity, and dexterity of the mouse forelimb.14,19,20 Baseline performance in the pasta matrix reach task was similar across the 3 age groups (Figure IVA in the online-only Data Supplement; P>0.1; n=8). After WMS, there was a significant decrease in performance in all age groups, but middle-aged and aged mice performed significantly worse than young-adult mice (Figure 6A; week-1 poststroke performance was 62.2±11.3% of baseline in young adults; 41.1±5.7% in middle-aged and 48.2±5.7% in aged; P<0.02; n=8). Functional recovery and return to control performance levels were observed in young-adult animals starting 4 weeks after stroke (Figure 6A, left; weeks 1–3, P<0.05; weeks 4–8, P>0.1; n=8), and in middle-aged animals 6 weeks after stroke (Figure 6A, middle; weeks 1–5, P<0.005; weeks 6–8, P>0.06; n=8). In contrast, no improvement in performance was observed in aged animals throughout the 8 weeks of testing (Figure 6A, right; weeks 1–8, P<0.001; n=8).
Similar to the pasta matrix reach task, there were no differences between age groups in baseline performance in the grid-walking task (Figure IVB in the online-only Data Supplement; P>0.3; n=8). Analysis of forelimb foot faults revealed a significant increase in the number of foot faults recorded in middle-aged and aged, but not young-adult mice, after WMS (Figure 6B; 7 days after stroke, young adults, P>0.3; middle-aged, P<0.005; aged P<0.01; n=8). Eight weeks after stroke, there was a functional recovery and return to control performance levels in middle-aged mice but not aged mice (Figure 6B, middle and right; 56 days after stroke, middle-aged, P>0.1; aged, P<0.01; n=8). These behavioral results indicate an age-related worsening of initial motor performance after stroke and a decline in recovery with age.
In the aged brain, white matter stroke produces more prolonged oligodendrocyte cell death and oxidative damage, more distributed microglial activation and expression of cytokines MCP1 and TNFα, increased white matter atrophy, and greater loss of corticostriatal connections compared with the young-adult brain. When comparing young-adult, middle-aged, and aged animals, stroke has an age dose–response, with progressively greater functional impairment and a greater initial behavioral deficit from young-adult to middle-aged and then aged animals.
The greater oligodendrocyte death and oxidative damage in aged animals, despite a similar areal extent of the lesion across age groups, may be caused by the ischemic injury and oxidative stress that follow white matter stroke and are more severe in aged animals.21 The axons in the white matter contain abundant mitochondria, a main source of reactive oxygen species after ischemia.22 Mitochondrial activity and integrity decrease during aging, and endogenous antioxidant system activities exhibit an age-dependent decline.23,24 Both processes may contribute to an increased susceptibility to ischemia, making mitochondria-mediated oxidative stress a predominant pathway of white matter injury in the aged brain.7
The increase in oligodendrocyte cell death after stroke in aged animals likely contributes to axonal degeneration through secondary demyelination. Loss of normal myelination damages axonal integrity and leads to axonal dysfunction.25,26 In addition, the ischemic insult has a direct effect on axons that may be more severe or occur through distinct mechanisms in aged animals compared with young animals.21,27 The present data show that a greater sensitivity of myelinated axons to ischemia is not reflected in an initially larger infarct zone in WMS, but in a substantially greater secondary loss of these axons with age.
There is an age-specific induction of several interacting arms of neuroinflammation after WMS. In the aged brain, WMS induces MCP1 and TNFα in neurons distant to the infarct, but the axons of which project through the ischemic area. Such a remote neuronal induction of MCP1 to axonal injury has been reported in nerve axotomy.28 These 2 cytokines attract and activate microglia and macrophages29 and may underlie the larger numbers of microglia/macrophages found in the cortical regions above a WMS in the aged brain. TNFα induces a more injurious response in tissue macrophages,30 greater damage to myelin and oligodendrocytes,31 and can result in greater injury in hypoxic central and peripheral insults.32,33 A more widespread and damaging inflammatory response in the cortex of aged animals has been reported in other models of injury, such as chronic stress or systemic LPS injection.34,35
The observed loss of corticostriatal projections likely underlies, at least in part, the motor impairment observed in this WMS model. The greater impairment observed in middle-aged and aged mice in the first week after stroke, and the diminished functional recovery in aged mice at longer time points, can be attributed respectively to greater initial damage and inefficient repair processes that result in lasting white matter atrophy. In addition to a more efficient white matter repair process in young brains, the behavioral recovery observed 4 weeks after stroke in young-adult mice, and 6 to 8 weeks after stroke in middle-aged mice, may be achieved through axonal sprouting and formation of new patterns of neuronal connections in adjacent brain regions.36,37 The gene expression profile that drives this reorganization differs significantly between the young and aged brain,38 and this may further contribute to the diminished recovery after WMS in aged animals.
In summary, white matter stroke causes increased inflammation, oxidative tissue injury, and delayed white matter atrophy in aged animals, which translate into an age-dependent exacerbation of behavioral motor deficits. The use of a clinically relevant model is likely crucial in translational research aimed at developing therapeutic interventions for WMS. The present studies use an extensive and long-duration multimodal analysis of the cellular and behavioral phenotypes of WMS in the aged brain, setting the stage for future pharmacological approaches to mitigate this progressive damage.
We thank Michael Reitman, Wenli Mui, and Meghan Yetzke for excellent technical assistance.
Sources of Funding
This work was supported by National Institutes of Health grant RO1 NS071481 (Dr Carmichael).
The online-only Data Supplement is available with this article at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA.113.001796/-/DC1.
- Received June 5, 2013.
- Revision received June 13, 2013.
- Accepted June 18, 2013.
- © 2013 American Heart Association, Inc.
- Schneider AT,
- Kissela B,
- Woo D,
- Kleindorfer D,
- Alwell K,
- Miller R,
- et al
- Srikanth V,
- Beare R,
- Blizzard L,
- Phan T,
- Stapleton J,
- Chen J,
- et al
- Sim FJ,
- Zhao C,
- Penderis J,
- Franklin RJ
- Hinman JD,
- Rasband MN,
- Carmichael ST
- Clarkson AN,
- Overman JJ,
- Zhong S,
- Mueller R,
- Lynch G,
- Carmichael ST
- Overman JJ,
- Clarkson AN,
- Wanner IB,
- Overman WT,
- Eckstein I,
- Maguire JL,
- et al
- Baltan S
- Lipton P
- Tekkök SB,
- Goldberg MP
- Matute C,
- Domercq M,
- Pérez-Samartín A,
- Ransom BR
- Henry CJ,
- Huang Y,
- Wynne AM,
- Godbout JP