Exercise Improves Cerebral Blood Flow and Functional Outcomes in an experimental mouse model of Vascular Cognitive Impairment and Dementia (VCID)

Abstract

Vascular cognitive impairment and dementia (VCID) are a growing threat to public health without any known treatment. The bilateral common carotid artery stenosis (BCAS) mouse model is valid for VCID. Previously, we have reported that remote ischemic postconditioning (RIPostC) during chronic cerebral hypoperfusion (CCH) induced by BCAS increases cerebral blood flow (CBF), improves cognitive function, and reduces white matter damage. We hypothesized that physical exercise (EXR) would augment CBF during CCH and prevent cognitive impairment in the BCAS model. BCAS was performed in C57/B6 mice of both sexes to establish CCH. One week after the BCAS surgery, mice were randomized to treadmill exercise once daily or no EXR for four weeks. CBF was monitored with an LSCI pre-, post, and four weeks post-BCAS. Cognitive testing was performed for post-BCAS after exercise training, and brain tissue was harvested for histopathology and biochemical test. BCAS led to chronic hypoperfusion resulting in impaired cognitive function and other functional outcomes. Histological examination revealed that BCAS caused changes in neuronal morphology and cell death in the cortex and hippocampus. Immunoblotting showed that BCAS was associated with a significant downregulate of AMPK & pAMPK and NOS3 & pNOS3. BCAS also decreased red blood cell (RBC) deformability. EXR therapy increased and sustained improved CBF and cognitive function, muscular strength, reduced cell death, and loss of white matter. EXR is effective in the BCAS model, improving CBF and cognitive function, reducing white matter damage, improving RBC deformability, and increasing RBC NOS3 and AMPK. The mechanisms by which EXR improves CBF and attenuates tissue damage need further investigation.

Introduction

Each year there are approximately 10 million new cases of dementia, of which about half arise or are contributed to vascular origins [1]. Vascular cognitive impairment and dementia (VCID) is an inclusive term for the wide range of neurological damage and cognitive impairment caused by vascular disease and low cerebral blood flow (CBF) [2]. Low CBF with hypoperfusion of brain white matter, specifically in the periventricular regions and centrum semiovale [1, 3], triggers a cascade of events that cause white matter damage in the brain and impaired cognition. This is likely due to the limited arterial supply in these areas, resulting in a watershed area in the deep white matter [4, 5].

As of this time, there has been no known specific treatment for VCID, and treatment of risk factors such as hypertension and diabetes are advised. Regular or chronic exercise (C-EXR) has been linked to improving recovery or preventing decline from VCID as it has been shown to increase CBF and, in an observational study, reduce the progression of cognitive decline in patients with white matter disease [1, 6, 7]. Exercise is advantageous for cognitive function in aging [8–15] and structural changes in the brain, mainly in the hippocampus [9, 14, 16], a domain that plays a potential role in learning and memory [17–19].

Moreover, a cascade of inflammatory responses is a critical factor in secondary brain injury caused by cerebral ischemia. Studies have found that after ischemic brain damage, microglial cells in the brain and macrophages that infiltrated the central nervous system (CNS) due to damage to the blood-brain barrier will rapidly migrate to the inflammatory zone. In the damaged environment, these cells are activated through the classical pathway to undergo M1 polarization and produce TNF-α, IL-6, and inducible nitric oxide, which promote inflammatory responses and neuronal death [20, 21]. In the later stage of neuroinflammation, macrophages that are polarized into the M2 phenotype by the alternative activation pathway can secrete IL-10, and other anti-inflammatory factors, which produce the regeneration of the axons. Therefore, regulating the activation status of macrophages can be a new treatment strategy for VCID [22–25].

To simulate VCID, bilateral common carotid artery stenosis (BCAS) in mice was performed. The BCAS model in mice is a valid model for VCID and does not lead to visual loss as occurs in rats. BCAS induces white matter damage, decreasing CBF, and causing similar white matter degeneration and cognitive deficiencies [26]. This model can replicate cerebral small vessel disease (SVD) with inflammation, blood-brain barrier damage, and changes in the brain’s small vessels [27]. This study tested the effects of long-term exercise in male and female mouse models of chronic cerebral hypoperfusion induced by BCAS. We examined the behavioral and pathological outcomes, NOS3, and AMPK signaling.

Materials and methods

Animals and Experimental Groups

Adult (20 ±1 week old) male & female C57BL/6J house-bred mice in AU’s AAALAC accredited facility were maintained in a standard 12 h light-dark cycle, temperature-controlled room (22 ± 1 °C), with access to food and water ad libitum. The Institutional Animal Care and Use Committee of Augusta University (AU) approved the experimental procedures per the National Institute of Health guidelines. Thirty mice of each sex were divided into three groups: (1) A Sham-operated group for procedures of BCAS (Sham, N=10); (2) BCAS (No Exercise, N=10); (3) BCAS treated with daily treadmill running Exercise (EXR) for one-month post-BCAS surgery (BCAS + EXR, N=10). A blinded investigator measured the outcomes. Cognitive function and CBF changes were considered the primary outcomes. Functional outcomes and CBF were determined one month after BCAS with 1MO-EXR. After that, blood was collected through either cardiac puncture for flow cytometry or RBC deformability estimation. Brains were dissected for histopathology and biochemical studies (Schematic representation of the experimental study plan in Supplemental Fig. 1).

Bilateral Carotid Artery Stenosis Surgical Procedure

BCAS-induced CCH was performed as described previously [28, 29]. In short, animals were anesthetized with 2% isoflurane with oxygen using a small-animal anesthesia system. A midline cervical incision exposed the common carotid arteries (CCAs) and was carefully separated from the cervical sympathetic and vagal nerves. Customized microcoils, designed to mimic a VCID model in the mice made of steel wire with an inner diameter of 0.18 mm, were twined by rotating around both rights and left CCA at 30 min intervals. The incision was closed eventually. During the surgery, body temperature was monitored and maintained at 37 ± 0.5 °C with a heating platform. Sham-operated mice underwent a similar surgery but without carotid coiling.

Treadmill Running Exercise

One week following BCAS surgery, mice were habituated to 10 min of treadmill-running (Columbus Instruments, Exer-6M, Ohio USA) at 10–15 m/min and 5 % grade for 2 consecutive days to observe the mice after surgery if able to be running on a treadmill. From day 3, a treadmill running exercise (at 15 m/min and 5 % grade) was applied five days per week with 30 minutes duration for one month. The animals in the non-exercise groups remained on the treadmill for the same period without running.

Cerebral Blood Flow by Laser Speckle Contrast Imager

Cerebral blood flow (CBF) was measured using a high-resolution Laser Speckle Contrast Imager (LSCI) (PSI system, Perimed Inc.) at different time points, as indicated in the figure and as previously reported by us[28, 30]. Mice were placed on a warming pad and thermostatically controlled at around 37°C to avoid the effect of body temperature during the measurement of CBF.

Samples and tissue preparation

Following four weeks of treadmill running exercise, animals were deeply anesthetized with isoflurane and perfused transcardially with cold saline using a butterfly needle with a syringe. Brains were removed and divided into two parts; one was snap-frozen with liquid nitrogen for western blotting, and the other half was placed in 4% paraformaldehyde for paraffin sectioning.

Behavioral study: Functional outcomes

Novel object recognition (NOR) test

Behavioral assessment by novel object recognition (NOR) test was performed as reported earlier by us [28, 30]. The 2-trial novel object recognition task was also performed. A mouse was placed in an enclosed box (40 × 40 inch) with two identical objects within a 4-inch diameter circle and located apart. The mouse was then removed from the environment for a set amount of time, and 1 of the two previously used (familiar) objects was replaced with a novel object that differed from the familiar object in texture and appearance. The mouse’s behavior on exposure to the novel object was then recorded. This test is based on the natural tendency of mice to investigate a novel object rather than a familiar one, which reflects the use of learning and recognition memory processes. The capability of the mouse to discriminate between a familiar versus novel object was determined as the discrimination index, DI=(Tn-Tf)/(Tn+Tf), where Tn is the time spent by the mouse with the novel object and Tf indicates the time spent with the familiar object.

Spatial working memory test on the Y maze

A spatial working memory test on the Y maze was performed as described by us and others [28, 31]. Briefly, a mouse was placed in one arm and allowed to explore for 7 min freely. Mouse behavior was monitored, recorded, and analyzed by a webcam (C920, Logitech, Newark, CA) and the Any-Maze software (Stoelting, Wood Dale, IL). A mouse was considered to have entered an arm if the whole body (except for the tail) entered the arm and exited if the entire body (except for the tail) exited the arm. If an animal consecutively entered three different arms, it was counted as an alternating triad. Because the maximum number of triads is the total number of arm entries minus 2, the score of alternation was calculated as “the number of alternating triads/ (the total number of arm entries minus 2)”.

Beams walk test

Motor balance and coordination tests were performed using the balance beam reported earlier [32] with some modifications. Mice were trained before the actual test three times. During training, mice were encouraged to keep moving across the beam by prodding, poking, or pushing it from behind with gloved fingers. Training trials were repeated until each animal crossed the beam three times without stopping or turning around. Three tests were performed per mouse to cross a beam on final testing. The time (in seconds) to cross the beam is counted by a timer (one at 25 cm that starts a timer and one at 100 cm that stops the timer). A soft bed/cloth is stretched below the beam above the tabletop to cushion any falls. Trials in which the animal stopped or turned around were repeated. The average of the tests was calculated in seconds.

Wire Hanging test

This test was applied to determine grasping ability, forelimb strength, and coordination movements described earlier [33]. The hanging time of fall was measured three times (in seconds) for each mouse. Couple of trials served to familiarize the mouse with the testing evaluation. Later, three tests were taken at five-minute intervals to allow a recovery period. The hanging time measurements start from the point when the mouse was hanging free on the wire and last with the animal falling to the cage below the cord. No attempt was made to force compliance during any of the various trials. If an animal adopted a strategy that permitted an extended hanging time, this was allowed, and the actual latency to fall time was recorded (in secs). In cases where an animal could not hang from the wire for at least one second, a fall time of zero was noted for that trial.

Open-field test

Mice were placed in a 40 × 40 × 40 cm box for 10 minutes, and the activity was recorded digitally. Distance traveled, mean velocity, and time spent in the center zone were determined using Ethovision XT video tracking software (Noldus Information Technology).

Histopathology

A standard paraffin block was used for histopathological staining to get the coronal sections of thickness 6 µm. Each slide was mounted with three sections (4–5 mice/group). Histochemical staining for fiber density of Luxol-Fast-Blue (LFB, a demyelination marker)-neutral red staining was performed to detect the severity of WM lesion, as we previously published [28]. Briefly, de-waxed rehydrated sections were immersed in the LFB solution (Solvent Blue 38, Sigma) at 60°C overnight. Excess stain was removed by 95% ethanol treatment followed by washing with deionized water. Grey and white matter differentiation was initiated with the treatment of 0.05% aqueous lithium carbonate (Sigma) for 30 seconds, followed by 70% ethanol until the nuclei were decolorized. Sections were immersed in neutral red solution (Sigma) for 30 min and washed in deionized water. They were dehydrated in an ethanol gradient (70 – 100%), cleared in xylene and mounted with cytoseal. For neuronal cell morphology, sections were stained with Hematoxylin & Eosin (H&E) and Nissl staining using cresyl violet (CV). The severity of white matter (WM) lesion was graded into four levels based on Klüver-Barrera (Luxol Fast Blue) staining by an investigator blind to the experimental condition as no stain (0), low stain (1), moderate stain (2) and high stain (3). The analysis corrected the percentage /cell numbers and normalized them to those found in the sham groups.

Western blot

Brain tissue from all groups was isolated and homogenized with radioimmunoprecipitation assay (RIPA) buffer using protein inhibitors (100 µL/mL). Protein concentrations were quantified using a BCA Protein Assay kit (Thermo Scientific), according to the manufacturer’s guidelines, and evaluated for expression of active eNOS (p-S1177-eNOS) and active AMPK (p-T172-AMPK) by Western blotting as previously described [34, 35]. In brief, equal volumes of homogenized tissues with equal protein content (50 μg total protein) were separated by 4–15% gradient sodium dodecyl sulfate-polyacrylamide gel electrophoresis and transferred to a polyvinylidene fluoride (PVDF) membrane. Blots were incubated overnight at 4°C in primary antibody 1:8,000 anti–β-actin (Sigma-Aldrich, USA, A5441); 1:500 anti-eNOS (Cell signaling, USA, 32027S); 1:500 anti-p-S1177 eNOS (Cell signaling, USA, 9571); 1:500 anti-AMPK (Cell signaling, USA, 2532S); 1:500 anti-p-T172-AMPK (Cell signaling USA, 2535) followed by a 1-hour incubation at room temperature with HRP-conjugated anti-rabbit or anti-mouse secondary antibody (Jackson Immunoresearch). Blots were visualized using ECL plus Western Blotting Detection System (Bio-Rad), and densitometry analysis was performed using ImageJ software.

Flow cytometry

Whole Blood (100 µl) was collected via cardiac puncture. Cells were incubated with antibodies against the conjugated cell surface marker for Glycophorin A (GPA: Bioss USA, Cat#bs-2575R) for 20 minutes in the dark on ice. Following a PBS wash, cells were fixed and permeabilized using a Fixation/Permeabilization Concentrate (Affymetrix eBioscience USA Cat#88–8824) and then incubated with antibodies for intracellular labeling of p-eNOS (BD Bioscience Cat#560103), p-AMPK (Bioss USA Cat#bs-4002R) and Hemoglobin Beta, HGB, (Santa Cruz Biotech, USA Cat#sc-21757) for 20 minutes in the dark on ice. After a final wash, cells were analyzed using a four-color flow cytometer (FACS Calibur, BD Biosciences) and CellQuest software (BD Biosciences), as described previously [36]. Isotype-matched controls were analyzed to set the appropriate gates for each marker. To minimize false-positive events, the number of double-positive events detected with the isotype controls was subtracted from the number of double-positive cells stained with corresponding antibodies (not isotype control), respectively. Viable cells were visibly differentiated from debris by gating on live cells with high forward scatter (FSC) and positivity for specific antibodies. Single staining was performed for compensation controls, controls to check for fluorescence spread, and isotype controls were used to determine the level of nonspecific binding. Cells expressing a specific marker were reported as a percentage of the number of gated events.

Plasma nitrite measurement

Plasma nitrite levels were measured by NO-specific chemiluminescence, as described previously [37]. Briefly, 100 ul of plasma were mixed with twice volumes of ethanol (100%) and kept at ‒20 C for 60 min, followed by centrifuging at 13000 rpm for 10 min to remove protein as a pellet. The supernatant (100 ul) was taken and injected to measure nitrite. The nitrite levels were measured by NO Analyzer 280i (GE Analytical Instruments, CO, USA). The level of nitrite was expressed in nanomolar.

Magnetic Resonance Induction (MRI) acquisition and Analysis (ASL MRI test White matter degeneration)

All MRI experiments were conducted using a 7 Tesla horizontal magnet with a clear bore of 20 cm in diameter interfaced to a Bruker Advance console (BioSpec 70/20, Paravision 6.1). Animals underwent arterial spin labeling (ASL) imaging acquired through a flow-sensitive alternating inversion recovery (FAIR) sequence, together with the associated T1 map generated from spin-echo T1 weighted images using a multiple TR (repetition time) to determine the CBF. The following parameters were used to acquire FAIR images; TR 11780.4 ms, TE 46.4 ms, FOV 2.4 cm, matrix 128×128, TI 1400ms, slice thickness 2.4mm, NEX 20 and corresponding T1 images; TR 5500 to 326 ms, TE 8.5 ms, FOV 2.4 cm, matrix 128×128, FA 90, slice thickness 2.4mm, NEX 1. For 3D diffusion tensor imaging, we have used the following parameters: single-shot EPI, slice thickness 0.8 mm, TR 1000ms, TE 28.82 ms, BW 300 kHz, seven directions, B values of 0 and 1000, matrix 128×128, FOV 2.4 cm.

RBC deformability

RBC deformability test was determined as a function of the elongation index (EL) at various shear stresses (0–20 Pascal pressure) with the help of RheoScan-AnD 300, a laser-assisted ektacytometry instrument integrated into a dedicated software (RheoMeditech, South Korea). Blood samples were drawn from each group of mice under isoflurane-anesthetized with a cardiac puncture to measure erythrocytic deformability (EI-value) at the termination of the study. Briefly, 6-μL of fresh blood was mixed with 600-μL polyvinylpyrrolidone (PVP) solution in saline (300 Osmolality, Osm), an isotonic medium of optimum viscosity. A 500-μL aliquot was immediately transferred to a disposable deformability test kit (K01 Kit; RheoMeditech, South Korea). The test kit was placed inside the RheoScan for automated readout, data, and image collection per the manufacturer’s instructions, which takes less than two minutes.

Statistical Analysis

All statistical analysis was performed using GraphPad Prism version 9 (GraphPad Software, San Diego, CA, USA). A one- or two-way ANOVA was used to examine differences between the groups (Sham, BCAS, BCAS+EXR). If the overall test for the one-way ANOVA was statistically significant, a Tukey–Kramer or Bonferroni multiple comparison tests were used to examine differences between the three groups in both sexes. Statistical significance was accepted at the 95% (P < 0.05) confidence level, and individual P values, determined from multiple comparison tests, are indicated in the text and figure legends.

Results

Effect of chronic exercise (C-EXR) on Cerebral Blood Flow (CBF) after BCAS-induced cerebral hypoperfusion

The effect of treadmill exercise on CBF of Sham, BCAS, and BCAS+EXR in both sexes was examined. In males, there was no significant difference at baseline before sham & BCAS surgery (100 ± 3.4 vs 102 ± 4.7 respectively: 99.57 ± 4.76 % change of sham). As expected, there was a considerable reduction in CBF post-BCAS surgery as compared to sham surgery (60.58 ± 2.4 vs 100 ± 3.2 respectively: 60.76 ±1.68, % change of sham). In females, there was no significant difference at baseline before surgery (sham 100±5.36 vs. BCAS 103.2±6.8; 109.9±3.54, % change of sham), whereas there was a considerable difference post-BCAS surgery (48.08±1.98; 56.23±1.9, % change of sham) to sham surgery (100±4.26, % change of sham). However, there was a considerable difference between the CBF of all groups in both sexes one month after C-EXR. In males, in comparison to the nonexercised BCAS group, there was a significant increase (p<0.0001; 58.48±3.66 vs 91.32±2.29 %) in CBF after one month of C-EXR therapy in the BCAS+EXR group, whereas no substantial changes (100±4.50, % change of sham) in sham groups. In females, in comparison to the nonexercised BCAS group, there was a significant increase (p<0.0001, 54.92±2.51 vs. 109±5.59, % change of sham) in CBF after one month of C-EXR therapy in the BCAS + EXR group, whereas no substantial changes (100±2.23, % change of sham) in sham groups (Fig. 1).

Fig. 1.

Measurement of cerebral blood flow changes by laser speckle contrast imager (LSCI) pre- (Pre), immediately post- (Post), at one-month post- (1 MO) sham or BCAS surgery in both sexes. a (males) & b (females) Representative contrast and photo overlay images of cerebral perfusion, c (male) & d (female) % change of cerebral perfusion with respect to the sham, where “blue” line indicates sham group, “red” symbols indicate BCAS group while “green” symbols are for BCAS + EXR group. A two-way ANOVA followed by Tukey’s multiple comparisons test was performed for statistical analysis. Data presented as mean ± SEM from 9 to 10 mice/group; ****p < 0.0001; ns, not significant.

Effect of C-EXR on Neurobehavioral and functional outcomes

One-month chronic treadmill exercise (C-EXR) improved cognition in male (Fig. 2a, b) and female (Fig. 2e, f) mice with BCAS-induced chronic hypoperfusion. Results from the NOR test (for spatial memory) (Fig. 2a, b & e, f) and the Y-maze test for working memory (Supplemental Fig. 2 male A-B & female C-D) showed that the sham group was more attracted to novel objects than nonexercised BCAS group. The BCAS group was significantly lesser exploration time for a novel object (TN) and short discrimination index (DI) in both genders. This indicates that BCAS causes impairment of memory. However, BCAS groups treated with regular exercise daily for one month showed improved cognitive function compared to non-treated BCAS groups which was not significantly different from sham controls. Moreover, C-EXR significantly increased the entries’ alternations in the arms of Y-maze compared to the sham group tested at 1MO after BCAS. It indicates that exercise improves both spatial and working memory.

Fig. 2.

Exercise improves spatial memory and muscular strength in both genders.

After BCAS, animals were significantly less time of exploration (TN) with novel objects to nonexercised BCAS as compared to BCAS+EXR groups (males a, females e) with a reduction in discrimination index (DI) (males b, females f). Values are indicated as mean ± SEM from 9 to 10 mice/group. For males TN, **p=0.0068 sham vs. BCAS; *p=0.0474 BCAS vs. BCAS+EXR. For females TN, ****p<0.0001 sham vs. BCAS; ***p<0.0005 BCAS vs. BCAS+EXR. For males DI, ***p=0.0001 sham vs. BCAS, ****p<0.0001 BCAS vs. BCAS+EXR. For females DI, **p=0.0036 sham vs. BCAS; *p=0.0426 BCAS vs. BCAS+EXR.

Similarly, for muscular/motor strength, after BCAS, animals spent significantly more time crossing on a beam by nonexercised BCAS as compared to BCAS+EXR groups (males c, females g) with a reduction in wire hanging time (males d, females h). Values are indicated as mean ± SEM. For males beam walk crossing, **p=0.0075 sham vs. BCAS; *p=0.0147 BCAS vs. BCAS+EXR. For females beam walk crossing, **p=0.0079 sham vs. BCAS, **p=0.0015 BCAS vs. BCAS+EXR. For males wire hanging, *p=0.0448 sham vs. BCAS; *p=0.0166 BCAS vs. BCAS+EXR. For females wire hanging, ***p=0.0001 vs. sham; **p=0.0013 BCAS vs. BCAS+EXR; ns, not significant.

We next tested muscular/motor strength with beam walk crossing (Fig. 2c, g) and wire hanging test (Fig. 2d, h) in both sexes after one month of C-EXR. In male mice, the beam walk crossing test and wire hanging test showed significant differences (p<0.05, Fig. 2c, d) between BCAS and BCAS+EXR groups at one month, whereas there was no difference between sham and BCAS+EXR. In female mice, the beam walk crossing test and wire hanging test showed significant differences (p<0.05, Fig. 2g, h) between BCAS and BCAS+EXR groups at one month, whereas there was no difference between sham and BCAS+EXR.

We also tested the open field test for distance traveled, velocity and frequency in the center zone to determine if BCAS-induced chronic hypoperfusion leads to anxiety and depression in these young-middle age groups of animals. The mice subjected to sham versus BCAS surgery and BCAS +EXR group did not show significant changes in locomotor activity in either sex (Supplementary Fig. 2A-C for males & 2D-F for females). It indicates that young middle-aged mice have no symptoms of anxiety and depression after 1MO of BCAS.

Effect of C-EXR on BCAS-induced WM degeneration, neuronal cell damage, and cell morphology

Luxol fast blue (LFB)/Klṻver-Barrera staining was studied to observe the effect of C-EXR on myelin integrity. The results showed that myelin damage (including myelin fiber disorder and loss of myelin sheath) appeared in the corpus callosum and cingulum bundle region of the BCAS group one month after BCAS surgery, compared with the sham group (Fig. 3a, b for male & 3c, d for female). In addition, relative to the BCAS group, C-EXR significantly reduced myelin damage, as shown by decreasing WM grading scores (Fig. 3b for males & 3d for females). Furthermore, to determine whether long-term exercise protects from neuronal cell damage in this BCAS-induced VCID, we used CV & H&E staining to examine neuronal damage in the cortical and hippocampal (CA1 and CA3) regions. In the present study of CV, neurons were identified morphologically by their larger, pale nuclei surrounded by darkly stained cytoplasm containing Nissl bodies. In the sham group, neurons in the indicated regions were well stained with CV. In the nonexercised BCAS group, we found that CV-positive neurons tended to be either decreased or dispersed in both cortex and hippocampus CA3 of independent. In contrast, with the exercise therapy in BCAS+EXR groups, CV-positive neurons were ameliorated or in the usual shape (Supplementary Fig. 4A-C; magnified images supplementary Fig. 5A-B & 6 for cortex & Fig. 7 for CA1 hippocampus region; For male cortex: 53.24 ± 4.75 vs 87.05 ± 7.91 %; for female cortex: 55.17 + 4.66 vs 81.61 + 5.26 %; for male CA3: 46.6 + 6.25 vs 97.09 + 8.81 %; for female CA3: 54.84 +7.05 vs 81.45 + 6.42 % BCAS vs BCAS+EXR respectively relative to % of sham ).

Fig. 3.

Exercise maintains white matter in the corpus callosum (CC) and cingular gyrus (Cg) regions of both genders. The intensity in Klüver-Barrera staining after BCAS was decreased than after the sham operation while it was protected by treadmill exercise for both sexes. a (male) & c (female) Representative image (left panel, 10×, scale bar=100 µm) from cross-sections of the corpus callosum & cingular region of the Sham, BCAS, and BCAS+EXR groups with the Luxol fast blue (LFB)/Klṻver-Barrera staining and its higher magnification (right panel, 20×, scale bar=50 µm) showing significant degeneration of white matter after BCAS and its prevention by exercise therapy with 1MO in both sexes. b (male) & d (female), Histogram showed grading scores of WM degeneration and their protection by exercise. Values are indicated as mean ±SE from 4–5 mice/group. ***p=0.0006 sham vs BCAS; **p=0.006 BCAS vs BCAS+EXR; **p=0.0032 sham vs BCAS; **p=0.0095 BCAS vs BCAS+EXR; ns, not significant.

Fig. 7.

Exercise increases RBC deformability in both genders. (a & e) Elongation index (EI) measured at 3.0 Pa, reflecting shear stress in the microcirculation as Major vessels. **p=0.0014 BCAS vs BCAS+EXR; *p=0.0273 BCAS vs BCAS+EXR; ns, not significant. (b & f), Elongation index (EI) measured at 10 Pa, reflecting shear stress in the microcirculation as Minor vessels *p=0.0308 BCAS vs BCAS+EXR; *p=0.0265 BCAS vs BCAS+EXR; ns, not significant. (c & g), Elongation index (EI) was measured at 17 Pa, reflecting shear stress in the microcirculation as Microvessels. *p=0.0326 BCAS vs BCAS+EXR; *p=0.0265 BCAS vs BCAS+EXR; ns, not significant. (d & h), Maximum Elongation index (EImax) reflects shear stress in the microcirculation as Max Deformability. **p= 0.0042, *p=0.0212 BCAS vs BCAS+EXR; ns, not significant. Values are indicated as mean ±SEM, n=5/group.

Next, the brain slides were stained with Hematoxylin and Eosin (H&E) staining. Pyramidal cells in the hippocampal CA3 (Supplementary Fig. 8) and CA1 (Supplementary Fig. 9, higher magnification) area of the Sham group were dense and arranged neatly in a hierarchical order. The cell morphology and construction were of good integrity and clarity. The nuclei were rounder and larger with clear nucleoli. Compared with the Sham group, the pyramidal cells of the nonexercised BCAS group were either scattered or cells were whitish in color. Some pyramidal neurons displayed karyopyknosis, with a loss of the nucleolus, a pyknotic cytoplasm, hyperchromatic features, and losing their stain colors. In the EXR-treated animals, the neuronal cells were prominently well shaped, with an improvement in the cellular and nuclear morphology (Supplementary Fig. 9).

Effect of C-EXR alters phosphorylation of NOS3, AMPK & nitrite production after BCAS surgery in a gender-independent manner

The flow cytometry analysis demonstrated that BCAS significantly decreased the expression level of p-NOS3 and p-AMPK in a gender-independent fashion in the normal sham control mice (Fig. 4). One month of C-EXR was able to reverse the effects of BCAS and restore the expression of p-NOS3 and p-AMPK towards the normal level in both sexes (Fig. 4a and 4b, the bottom panels). Further, as shown (Fig. 4a, i-iii & 4b, i-iii), glycophorin A, pNOS3, and pAMPK positive cells were significantly reduced within the nonexercised BCAS group compared to sham groups in a gender-independent manner. Importantly, one-month exercise therapy significantly increased expression in the BCAS+EXR group compared to the BCAS group except for glycophorin A (although its trends showed an increase with exercise). Despite showing a direction of increasing trends in the histogram, whole blood analysis demonstrated no significant difference in the total frequencies of MQs (M1 and M2) as well as the ratio of M1/M2 MQs (Supplementary Figs. 10A, i-iv for male; 10B, i-iv for females) among all groups in a gender independent fashion.

Fig. 4.

Flow cytometry graphs show in male (a) and female (b) a significant increase in the pNOS3 and pAMPK counts with one month of EXR therapy in BCAS+EXR groups compared to the BACS group. Histograms on the left upper panel for pNOS3 and pAMPK and the right upper panel for M1 & M2 for both male (a) and female (b) show the isotype controls (shaded grey-filled graphs). Histogram at the left bottom panel (4a, i-iii) glycophorin, pNOS3 and pAMPK for male; and right bottom panel (4b, i-iii) glycophorin, pNOS3 & pAMPK for female. The X axis is the relative fluorescence (light scatter intensity) and Y axis is the number of events (cell counts). For dot plots, the scales are relative fluorescence (light scatter intensity) on both axes. Values are indicated as means ±SEM from 5 mice /group for each sex and were analyzed by a one-way ANOVA test followed by Bonferroni’s multiple comparison tests. Comparison of each group for male 4a (i) **p=0.0011 sham vs BCAS; *p=0.0157 BCAS vs BCAS+EXR; 4a (ii) ***p=0.0007 sham vs BCAS; *p=0.0201 BCAS vs BCAS+EXR; 4a (iii) ****p<0.0001 sham vs BCAS; ***p=0.0009 BCAS vs BCAS+RXR; ns, not significant; and comparison of each group for female 4b (i) **p=0.0014 sham vs BCAS; *p=0.0373 BCAS vs BCAS+EXR; 4b (ii) **p=0.0016 sham vs BCAS; *p=0.0199 BCAS vs BCAS+EXR; 4b (iii) ****p<0.0001 sham vs BCAS; ***p=0.0005 BCAS vs BCAS+RXR; ns, not significant.

We next analyzed the expression of AMPK & pAMPK (Fig. 5), NOS3 & pNOS3 (Fig.6) using the Western blotting technique. In males, there were significantly upregulated total AMPK (t-AMPK) and phospho-AMPK (pAMPK) (Fig. 5a, p=0.0001) expression in BCAS+EXR groups compared to BCAS groups. However, the ratio of pAMPK/tAMPK compared with the nonexercised BCAS group, a significant (Fig. 5b, p=0.01001) upregulation of the ratio was observed in the BCAS+EXR group. In females, the BCAS+EXR group, compared with the nonexercised BCAS group, showed a significant expression of AMPK (Fig. 5c, p=0.0283), and a phospho-AMPK (Fig. 5c, p=0.0001) was observed. However, the ratio of pAMPK/tAMPK compared with the nonexercised BCAS group, a significant (Fig. 5d, p=0.0109) upregulation of the ratio was observed in the BCAS+EXR group. As AMPK is a direct activator of NOS3, we evaluated if the long-term exercise-mediated AMPK activation may translate into increased activation of NOS3. We analyzed the expression of total NOS3 (tNOS3) and phospho-NOS3 (pNOS3) by Western Blotting (Fig. 6). In males, there was a significant upregulation in the t-NOS3 (Fig. 6a, p=0.0009) and pNOS3 (Fig. 6a, p=0.0002) in BCAS+EXR groups compared to BCAS groups. However, the ratio of pNOS3/tNOS3 compared with the nonexercised BCAS group showed a significant (Fig. 6b, p=0.0374) upregulation of the ratio in the BCAS+EXR group. In females, there were significant differences in the tNOS3 (Fig. 6c, p=0.0001) and pNOS3 (Fig. 6c, p=0.0001) levels in BCAS+EXR groups compared to BCAS groups. However, the ratio of pNOS3/tNOS3 compared with the nonexercised BCAS group showed a significant (Fig.6d, p=0.0353) upregulation of the ratio in the BCAS+EXR group. These results further suggest that exercise positively regulates the AMPK/NOS3 pathway through its phosphorylation during BCAS-induced hypoperfusion.

Fig. 5.

BCAS-induced chronic hypoperfusion downregulates AMPK & pAMPK. a (in males) & c (in females) Representative immunoblotting images of AMPK & pAMPK. b (in males) and d (in females). Protein levels of AMPK and pAMPK in brain homogenates were detected by immunoblotting (n = 4). All bars represent means ± SEM; A one-way ANOVA followed by Bonferroni’s multiple comparisons test was performed for statistical analysis. The results are expressed as a ratio of phospho-AMPK/total AMPK for each sex with respect to sham control. Comparison of each group for male: p=0.0439 sham vs BCAS; *p = 0.0100 BCAS vs BCAS+EXR and comparison of each group for female p=0.2564 sham vs BCAS; *p = 0.0109 BCAS vs BCAS+EXR; ns, not significant.

Fig. 6.

BCAS-induced chronic hypoperfusion downregulates NOS3 & pNOS3. a (in males) & c (in females) Representative immunoblotting images of NOS3 & pNOS3. b (in males) and d (in females). Protein levels of NOS3 and pNOS3 in brain homogenates were detected by immunoblotting (n = 4). All bars represent means ± SEM. A one-way ANOVA followed by Bonferroni’s multiple comparisons test was performed for statistical analysis. The results are expressed as a ratio of phospho-NOS3/total NOS3 for each sex with respect to sham control. Comparison of each group for male: p=0.3280 sham vs BCAS; *p = 0.0374 BCAS vs BCAS+EXR; and comparison of each group for female: p=0.1170 sham vs BCAS; *p = 0.0353 BCAS vs BCAS+EXR; ns, not significant.

We next evaluate plasma nitrite production, where treadmill running exercise for 4 weeks post-BCAS significantly increased the plasma nitrite levels compared to nonexercise BCAS groups of animals in both sexes (Supplementary Fig. 11; males A & females B). However, no significant changes were shown between the sham and BCAS+EXR groups. It indicates that exercise increases nitrite levels in the blood.

Effect of C-EXR on RBC deformability both in male and female mice after BCAS surgery

Chronic hypoperfusion induced by BCAS decreases CBF, and its consequences have a harmful effect on RBC morphology. Exercise plays a potential role in maintaining RBC deformability. RBC deformability was measured as elongation index (EI) at 3pa (reflecting shear stress in the microcirculation as Major Vessels), 10pa (reflecting shear stress in the microcirculation as Minor Vessels), 17pa (reflecting shear stress in the microcirculation as Micro Vessels) and maximum elongation (EImax, reflecting shear stress in the microcirculation as Max Deformability) in both the sexes in all groups at the termination point of the experiment. In males, 1MO of EXR therapy significantly increased RBC deformability as major vessels (p=0.0014, Fig. 7a), minor vessels (p=0.0308, Fig. 7b), microvessels (p=0.0326, Fig. 7c), and maximum deformability (p=0.0042, Fig. 7d) in BCAS+EXR groups compared to the BCAS. However, no significant differences were shown between sham from BCAS groups. This was similar in female animals as 1MO of EXR therapy significantly increased RBC deformability as major vessels (p=0.0273, Fig. 7e), minor vessels (p=0.0265, Fig. 7f), microvessels (p=0.0265, Fig. 7g) and maximum deformability (p=0.0212, Fig. 7h) in BCAS+EXR groups compared to the BCAS. However, no significant differences were shown between sham from BCAS groups. Although, trends of the histogram show decreasing deformability in the BCAS groups.

Effect of C-EXR on Magnetic Resonance Induction (MRI) acquisition

We next tested cerebral blood flow (CBF) and fractional anisotropy using MRI. Using our previously used techniques, we acquired FAIR images using a 7T small animal MRI system (Bruker) [38–41]. Image post-processing for CBF was performed using ImageJ coupled with in-house designed ImageJ macroscripts, where M0, T1, and inversion (ms) data were applied to create maps (CBF) to indicate ml/100/min. We drew identical rectangular region of interest (ROI) on both hemispheres and noted the rCBF. The average CBF value of each animal was normalized to the average values of the sham-operated animals (average rCBF/average rCBF of sham) x100). Normalized rCBF of hemispheres was compared among the groups of male animals (Sham, BCAS, and BCAS+EXR groups). Decreased blood flow noted in nonexercised BCAS groups compared to shame operated animals. However, no significant differences were shown in sham groups compared to BCAS+EXR groups (Fig. 8a, c). The same MRI system was used to acquire DTI per our previously published methods [41]. DTI was analyzed using the vendor-supplied software Paravsion 6.1 (Bruker Inc.), from which the associated diffusion-weighted, FA, and ADC images were generated, and the values were noted. Similar to the above mentioned rCBF, all values were normalized to the average values of sham-operated animals (p=0.0105; 90.76 ±2.7 vs.105.7 ± 3.4 % BCAS vs BCAS+EXR respectively: % change of sham).

Fig. 8.

Exercise treatment improves CBF & ameliorates white matter degeneration in the BCAS-induced VCID model. (a) representative image shows a region of interest (ROIs) performed in the determining cerebral blood flow (CBF) images obtained from arterial spin labeling (ASL) magnetic resonance perfusion imaging in Sham, BCAS, and BCAS + EXR of male mice. The CBF values in the striatum were calculated from bilateral white dot lines ROIs. Bar graph showing the percentage change of CBF. (b) Visualization of the region of interest (ROIs) based DTI analysis in Sham, BCAS, and BCAS + EXR of male mice. Bar graph showing the percentage change of fractional anisotropy (FA): corpus callosum (CC) on the effect of BCAS microsurgery on DTI parameters. Data represented as mean ± SD. n =4 mice per group. *p<0.5, **p < 0.01 sham vs BCAS, ***p<0.001 BCAS vs BCAS+EXR; ns, no significance.

Discussion

Ours is one of the first reports where treadmill exercise (C-EXR) post-surgery in the BCAS-induced VCID murine model improved CBF, reduced cognitive impairment, improved motor function, and reduced WM damage and hippocampal neuronal. Exercise improved RBC deformability and activated RBC AMPK and NOS3. These findings were not sex-specific.

We assessed improvement between the one-month treadmill exercise and BCAS-sham exercise groups through behavioral tests of cognition and motor function. In this study, each test showed significant improvement in behavioral and functional outcomes in the one-month exercise group over the BCAS-sham exercise group. This is a good indicator that exercise can improve physical and cognitive impairment resulting from chronic hypoperfusion in the BCAS model and could be beneficial in as a therapy for patients at risk for dementia.

Ample evidence of basic studies showing the beneficial effect of exercise in protecting cognitive impairment has allocated its positive impact to a turnabout of pathological change in the cerebral white matter [42–46]. Recent studies revealed that structural alterations of the white matter are linked with a risk of disease advancement in AD patients [47, 48], and that restoration of myelin may help treat dementia. Some studies have shown the beneficial effect of exercise on white matter in dementia models, including mice [49, 50] and rats [51, 52]. To the best of our knowledge, only a few previous studies examined effectiveness of exercise in preventing white matter degeneration caused by chronic cerebral hypoperfusion [51, 52]. Similar to the current study, prior literature using a rodent model of vascular dementia, exercised rodents showed improved CBF and protected WM degeneration in their corpus callosum and cauda putamen and performed better in behavioral tests [52–54]. Additionally, MRI data supports our finding that BCAS-induced hypoperfusion leads to loss of white matter consistent with those published elsewhere [55]. However, our results are inconsistent with one published report in the BCAS model [56], where there were no significant differences in myelin density, markers used as myelin basic protein in the BCAS model between sedentary and exercise groups in biochemical analysis. This could be explained by the fact that pooled tissues homogenate determines may miss significant differences seen in in the histopathological studies. Our previous reports also revealed that, remote ischemic conditioning (RIC) an exercise mimetic therapy improved white matter protection [28, 30] with increased CBF and plasma NO.

This study suggests that physical or chronic exercise promotes the activation of phosphorylated AMPKα and its effect and interaction with phosphorylated NOS3, which is functionally linked to NO bioavailability. This increase in blood phosphorylated NOS3 suggests that blood cell NOS3 may be important in mediating some of the benefits of physical exercise.

RBC deformability is the ability of red blood cells to change their shape in response to external forces. It can involve various reactions, including expansion and curvature alteration [57]. RBC deformability is a crucial measure of the health of the RBCs, and improved RBC deformability has protective effects on the health of the body [58]. Some of this RBC deformability has been linked to NOS3 regulation by the peripheral vascular bed, which was determined by a biochemical test for NOS3 [59]. Although, the upregulation of AMPK & NOS3 expression shown in the exercise mice group compared to the nonexercised BCAS group suggests the increased RBC deformability was due to the involvement of AMPK/NOS3 mediated factor in the effectiveness of the treadmill running exercise. Other biomarkers that we studied were measured by flow cytometry. Flow cytometry can effectively test individual cell types for different surface receptors in whole blood with good sensitivity and minimal interference [60].

It has been shown that macrophages (M1 or M2) are associated with inflammation and white matter damage [61, 62]. In the present study, we found no significant difference in the M1 and M2 macrophages that treadmill exercise groups. This may be due to small number of samples.

The current studies do not rule out the contribution of other known mechanisms of physical exercise. Nonetheless, our studies provide additional insight and suggest that CBF and NOS3 interaction with AMPK may play an important role.

Although we have demonstrated that treadmill exercise ameliorates CBF, functional outcomes, white matter damage, and activates NOS3 and AMPK caused by cerebral hypoperfusion, there are some limitations and caveats to this study. First, we have not yet confirmed how the biochemical analysis documented in the specific region of exercised mice tissues contributes to alleviating cognitive decline. In addition, time-course studies would be necessary with some genetic background mice, such as NOS3 knock-out and NOS3 knock-out in red blood cells mice with exercise to determine the role of NOS3 and RBC specific NOS3 in the beneficial effect of exercise. Second, we did not determine the partial oxygen (PO2) values for the specific region, such as the corpus callosum, after exercise. Third, systemic factors released into the circulation during exercise are still undetermined. Since differences in age and gender may alter the dynamics of systemic humoral factors, as mentioned in the published literature [63–65], it will be essential to determine if our results could be replicated in old male and female mice.

In conclusion, we have demonstrated that treadmill exercise reduces cognitive decline in mice with prolonged cerebral hypoperfusion that mimics the pathophysiology of VCID. This beneficial effect of exercise may be associated partly with an increase of NOS3/AMPK in the blood, leading to reduced cognitive impairment.

Fig. 9.

Representative hypotheses show how might exercise protect against VCID. Several pathways might explain how exercise protects the brain and prevents the development of VCID and other memory impairments, such as Alzheimer’s disease. In mice, exercise improves vascular health and increases the amount of CBF with other factors in the brain, which promotes neurogenesis, the survival of new neurons, and reduces and protects white matter degeneration.

Abbreviations

1MO

One month

AMPK

Adenosine monophosphate-activated protein kinase

BCAS

Bilateral carotid artery stenosis

CBF

Cerebral blood flow

CCAs

Common carotid arteries

C-EXR

Chronic exercise or regular exercise

CCH

Chronic Cerebral hypoperfusion

DTI

Diffusion tensor imaging

EXR

Exercise

FAIR

Flow sensitive alternating inversion recovery

eNOS/NOS3

Endothelial nitric oxide synthase

pAMPK

Phospho-AMPK

peNOS/pNOS3

Phospho-eNOS or Phospho-NOS3

LSCI

Laser Speckle Contrast Imager

LFB

Luxol Fast Blue

NO

Nitric oxide

RBC

Red blood cells

VCID

Vascular contributions to cognitive impairment and dementia or Vascular cognitive impairment and dementia

WM

White matter