Characterization of White Matter Injury in a Rat Model of Chronic Cerebral Hypoperfusion
Background and Purpose—Chronic cerebral hypoperfusion can lead to ischemic white matter injury resulting in vascular dementia. To characterize white matter injury in vascular dementia, we investigated disintegration of diverse white matter components using a rat model of chronic cerebral hypoperfusion.
Methods—Chronic cerebral hypoperfusion was modeled in Wistar rats by permanent occlusion of the bilateral common carotid arteries. We performed cognitive behavioral tests, including the water maze task, odor discrimination task, and novel object test; histological investigation of neuroinflammation, oligodendrocytes, myelin basic protein, and nodal or paranodal proteins at the nodes of Ranvier; and serial diffusion tensor imaging. Cilostazol was administered to protect against white matter injury.
Results—Diverse cognitive impairments were induced by chronic cerebral hypoperfusion. Disintegration of white matter was characterized by neuroinflammation, loss of oligodendrocytes, attenuation of myelin density, structural derangement at the nodes of Ranvier, and disintegration of white matter tracts. Cilostazol protected against cognitive impairments and white matter disintegration.
Conclusions—White matter injury induced by chronic cerebral hypoperfusion can be characterized by disintegration of diverse white matter components. Cilostazol might be a therapeutic strategy against white matter disintegration in patients with vascular dementia.
Chronic cerebral hypoperfusion induced by long- standing hypertension or severe carotid stenosis can lead to ischemic white matter injury.1,2 Integrity of the white matter, which consists of a complex structural unit, including neuronal axons, surrounding myelin with nodal or paranodal proteins at the nodes of Ranvier, and supportive glial cells, is critical to maintain brain function.3–5 Diverse components of white matter interact with each other, such as axon–glial connection5 or oligovascular signaling with a vascular component.4 Diverse nodal or paranodal proteins at the nodes of Ranvier are essential for the structural and functional stabilization of myelinated axons comprising white matter.3 Diffusion tensor imaging is a recently developed magnetic resonance imaging technique, which is highly sensitive to the directional diffusivities of water, where tissues are oriented according to particular directions of white matter tracts.6
For chronic cerebral hypoperfusion, rats subjected to permanent occlusion of the bilateral common carotid arteries (BCCAo) were used, which has been previously suggested as a possible animal model for vascular dementia.7,8 As time went on after the surgery, cerebral blood flow and metabolism decreased chronically in our previous study7 (Figure I in the online-only Data Supplement). In this study, we investigated cognitive impairment and its association with disintegration of diverse white matter components. Previously, we have shown that cilostazol, a phosphodiesterase III inhibitor, has beneficial effects against memory impairments in a rat model of chronic cerebral hypoperfusion.8 Its protective effect against white matter damage has been reported in other studies using an animal model of chronic cerebral hypoperfusion.9,10
Materials and Methods
All experimental procedures were approved by the Animal Experiment Review Board of Laboratory Animal Research Center of Konkuk University, and procedures were in accordance with the Stroke Therapy Academic Industry Roundtable criteria for preclinical stroke investigations. Rats were allocated randomly (Figure IIA in the online-only Data Supplement). Diverse behavioral tests and serial brain magnetic resonance imaging were performed (Figure IIB in the online-only Data Supplement). Detailed Experimental Methods are described in the online-only Data Supplement.
Sham-operated rats quickly learned to find the hidden platform in the Morris water maze task (Figure 1A). In the last session, it took a longer time and path for the rats with BCCAo to find the hidden platform (P=0.037 for time latency; P=0.039 for path length; and P=0.047 for search error). However, performances improved when rats were treated with cilostazol (P<0.05, compared with rats with BCCAo). Swimming speeds did not differ among the groups (P=0.373). In the second probe trial, the number of crossings still remained low for the rats with BCCAo, indicating poor memory (P=0.020, compared with sham-operated rats); however, the number of crossings increased when rats were treated with cilostazol (Figure 1B).
In the odor discrimination task (Figure 1C), it took more time for rats with BCCAo to retrieve the reward owing to impairments in learning and memorizing the rewarded odorant with a marginal statistical significance among 3 groups (P=0.054). When rats were treated with cilostazol, their performance improved (P=0.041, compared with rats with BCCAo). Percentage of correct trials across all 15 consecutive trials was significantly lower in rats with BCCAo (P=0.039, compared with sham-operated rats). However, performance improved to the level of sham-operated rats when rats were treated with cilostazol (Figure 1C).
In the novel object location test (Figure 1D), compared with 77.4% preference for the new location in the sham-operated rats, 52.6% preference in rats with BCCAo (P<0.001, compared with sham-operated rats) implicates that they chose objects randomly. When rats were treated with cilostazol, preference for the new location was increased to 73.6%, suggesting successful memorization (P=0.002, compared with rats with BCCAo).
Neuronal Morphology in the Hippocampus
Thionin staining of the hippocampus demonstrated no significant morphological changes of neurons in the CA1, CA3, or dentate gyrus in any group (Figure III in the online-only Data Supplement).
Neuroinflammation was significantly increased in rats with BCCAo (P=0.046, compared with sham-operated rats in the CD11b staining). When rats were treated with cilostazol, neuroinflammation was attenuated to the level of sham-operated rats (Figure 2A).
Oligodendrocyte and Myelin Basic Protein
The loss of oligodendrocytes and myelin basic protein (MBP) were prominent in rats with BCCAo (P=0.031 for oligodendrocytes and P=0.008 for MBP, compared with sham-operated rats). Cilostazol treatment did not protect the loss of oligodendrocytes in rats with BCCAo (P=0.044, compared with sham-operated rats); however, MBP levels were similar to those of sham-operated rats when rats were treated with cilostazol (P=0.089, compared with sham-operated rats; Figure 2B). Similar pattern was observed in diaminobenzidine staining for MBP (Figure IV in the online-only Data Supplement). Although there was a trend for protective effect of cilostazol on MBP attenuation in rats with BCCAo, however, it was not statistically significant compared with vehicle treatment.
Nodal and Paranodal Integrity
Ankyrin-G signals at the nodes of Ranvier and contactin-associated protein (Caspr) signals at the paranodal junction were distinct in the sham-operated rats (Figure 2C). However, ankyrin-G signals were widely dispersed along the paranodal area in rats with BCCAo. When rats were treated with cilostazol, ankyrin-G signals were mainly localized in the nodal areas in a similar pattern to the sham-operated rats. Increased colocalization coefficiency of ankyrin-G and Caspr in rats with BCCAo (P=0.007, compared with sham-operated rats) was normalized when rats were treated with cilostazol (Figure 2D).
White Matter Integrity in Serial Diffusion Tensor Imaging
The optic chiasm, corpus callosum, and both external capsules, where white matter tracts are the most abundant in the rat brain, were selected (Figure 3A). Compared with the baseline, fractional anisotropy in the optic chiasm in rats with BCCAo was significantly reduced 1 week after surgery (P=0.009). When rats were treated with cilostazol, an ongoing reduction of fractional anisotropy (P=0.034) in the optic chiasm and an ongoing elevation of mean diffusivity (P=0.024) with a concomitant elevation of axial diffusivity (P=0.044) and radial diffusivity (P=0.042) in the corpus callosum were halted 5 weeks after surgery (Figure 3B). Representative tractography (Figure 3C) and quantitative analyses (Figure 3D) show more abundant white matter tracts when rats were treated with cilostazol (P=0.002 for fiber length and P=0.001 for fiber density).
Cognitive Impairments Induced by Chronic Cerebral Hypoperfusion
Consistent with previous studies,7 we confirmed the presence of memory impairments using the water maze task and novel object test in a rat model of chronic cerebral hypoperfusion. Poor performance in the odor discrimination task suggests that cognitive functions associated with olfaction were also impaired in a rat model of chronic cerebral hypoperfusion. Dysfunction in olfactory threshold, identification, or discrimination is an early sign of cognitive impairment in patients with various neurodegenerative diseases.11 The poor performance of rats with BCCAo in the odor discrimination task might be comparable with the olfactory dysfunction in patients with diverse neurodegenerative diseases. To our knowledge, impaired odor discrimination in a rat model of chronic cerebral hypoperfusion has not been reported before. In addition, we confirmed our previous results about the protective effect of cilostazol against cognitive impairments in a rat model of chronic cerebral hypoperfusion.8
Loss of Oligodendrocytes and Attenuated Myelin Density Induced by Chronic Cerebral Hypoperfusion
Major components of the white matter are neuronal axons, the surrounding myelin sheath, and myelin-producing oligodendrocytes.3 In previous studies, neuroinflammation has been suggested as a key pathophysiology of white matter injury in animal models of chronic cerebral hypoperfusion.5,7,8 The loss of oligodendrocytes or compromised oligodendrogenesis has been reported in a mouse model of cerebral hypoperfusion.12,13 In our study, the loss of oligodendrocytes, attenuation of myelin density, and concomitant neuroinflammation were prominent. Neuroinflammation and the loss of myelin were attenuated by cilostazol treatment, but the loss of oligodendrocytes could not be prevented. However, previous studies using a similar rodent model of chronic cerebral hypoperfusion showed that cilostazol could protect oligodendrocytes.9,10 A longer time interval before histological study and different cilostazol treatment protocol in our study might have led to this discrepancy. Myelinated axons survived longer than oligodendrocytes in a mouse model of primary oligodendropathy, suggesting that the myelin sheath can remain supportive for axons independently from oligodendrocytes.14 The protective effect of cilostazol against white matter injury might be selective to myelin, but not to oligodendrocytes. However, the selective protective effect of cilostazol needs to be confirmed in other experimental design using different animal models.
Structural Derangement at the Nodes of Ranvier After Chronic Cerebral Hypoperfusion
Ankyrin-G is a key cytoskeletal molecule, which is present at the nodal area and binds various nodal or paranodal proteins.3,15 Proteolysis of ankyrin-G disrupts integrity of the nodes of Ranvier and detaches myelin from the axonal surface.15 Caspr is an adhesion molecule present in the axonal membrane at the paranodal area.3 In our study, signaling by ankyrin-G nodal protein was widely dispersed over the paranodal area obliterating the distinction of nodal or paranodal regions. Given that the anti–ankyrin-G antibody detects both the intact ankyrin-G and its proteolytic fragments, the ankyrin-G signal detected at the paranodal areas might represent the breakdown products of ankyrin-G, which spread from the nodes of Ranvier. In our study, the overlapping signal of ankyrin-G and Caspr decreased on cilostazol treatment. Cilostazol might be also protective against structural derangement at the nodes of Ranvier.
Disintegration of the White Matter Based on Diffusion Tensor Imaging
An increased radial diffusivity has been reported in animal models of dysmyelination or demyelination.6 Therefore, reduced fractional anisotropy and increased radial diffusivity in the optic chiasm of rats with BCCAo might suggest white matter disintegration. In our study, decreased axial diffusivity in the optic chiasm might also suggest the presence of concomitant axonal damage, as there is an association between axonal injury and axial diffusivity reduction in the mouse optic nerve after retinal ischemia.6 Compared with vehicle treatment, no further ongoing decrease was found in fractional anisotropy in the optic chiasm after cilostazol treatment.
Interpretation of the diffusion tensor imaging parameters in the white matter of brain parenchyma might be more complicated because of concomitant inflammation or edema around white matter tracts. Further increase of mean diffusivity in the corpus callosum 5 weeks after BCCAo surgery might reflect ongoing encephalomalacic changes and disintegration of the white matter. Neuroinflammation-associated water accumulation in the affected tissue might be attributed to the increases of radial diffusivity, axial diffusivity, and mean diffusivity. Even 5 weeks after BCCAo surgery, we found no such increase in the cilostazol-treated group.
In conclusion, white matter injury after chronic cerebral hypoperfusion can be characterized by disintegration of the white matter, which is demonstrated by the loss of oligodendrocytes, attenuation of myelin density, structural derangement at the nodes of Ranvier, and disintegration of white matter tracts. Protective effect of cilostazol against white matter disintegration might suggest a therapeutic strategy for patients with vascular dementia.
Sources of Funding
This article was supported by Konkuk University in 2015.
The online-only Data Supplement is available with this article at http://stroke.ahajournals.org/lookup/suppl/doi:10.1161/STROKEAHA.115.011679/-/DC1.
- Received October 2, 2015.
- Revision received October 29, 2015.
- Accepted November 5, 2015.
- © 2015 American Heart Association, Inc.
- Arai K,
- Lo EH.
- Reimer MM,
- McQueen J,
- Searcy L,
- Scullion G,
- Zonta B,
- Desmazieres A,
- et al
- Choi BR,
- Lee SR,
- Han JS,
- Woo SK,
- Kim KM,
- Choi DH,
- et al
- Kwon KJ,
- Lee EJ,
- Kim MK,
- Kim SY,
- Kim JN,
- Kim JO,
- et al
- Watanabe T,
- Zhang N,
- Liu M,
- Tanaka R,
- Mizuno Y,
- Urabe T.
- Miyamoto N,
- Tanaka R,
- Shimura H,
- Watanabe T,
- Mori H,
- Onodera M,
- et al
- Miyamoto N,
- Pham LD,
- Hayakawa K,
- Matsuzaki T,
- Seo JH,
- Magnain C,
- et al
- Reeves TM,
- Greer JE,
- Vanderveer AS,
- Phillips LL.