Evaluation of the Electroacupuncture Effect on Cerebral Blood Flow in APP/PS1 Mice by Using the Laser Speckle Technique

Alzheimer's disease (AD) is a progressive neurodegenerative disorder characterized by the progressive cognitive impairment and memory decline. It processes the pathological features of the deposition of extracellular beta-amyloid (Aβ) plaques, intracellular neurofibrillary tangle (NFTs), and neuronal loss¹. While these classic features have been the primary focus of therapeutic research, growing evidence underscores the critical role of decreased cerebral blood flow (CBF) in the pathogenesis and progression of AD². CBF reduction occurs in the early stage of AD onset and is associated with the severity of cognitive dysfunction³. Therefore, the treatment strategy for improving cerebrovascular damage is an effective way to alleviate AD symptoms and improve the course of the disease.

Electroacupuncture (EA), a modern technique that integrates traditional acupuncture with electrical stimulation, has shown potential in improving cognitive function in AD. Studies suggest that EA may exert its beneficial effects through multiple mechanisms, such as reducing Aβ deposition, attenuating neuroinflammation, and promoting synaptic plasticity⁴. Notably, there is increasing speculation that the neuroprotective effects of EA may be closely linked to its ability to modulate CBF⁵. However, direct, real-time evidence demonstrating the impact of EA on cerebral hemodynamics in AD models remains limited. The lack of such evidence hinders a comprehensive understanding of how EA improves brain function.

Technological advancements in optical imaging have provided powerful tools for investigating cerebral hemodynamics. As a fast, wide-field optical technique, LSCI delivers high spatiotemporal resolution for visualizing volume-integrated tissue blood flow maps⁶, enabling real-time monitoring of electroacupuncture-induced perfusion dynamics. While skull exposure is required for optimal optical access, LSCI avoids invasive tissue penetration. LSCI was introduced in the 1990s and has been widely applied in neuroscience in the past few decades⁷. Like all imaging techniques, LSCI has inherent limitations: its optical penetration depth is constrained, favoring superficial cortical vasculature while limiting access to deep brain regions (e.g., hippocampus, basal ganglia), and it quantifies relative blood flow (perfusion units) rather than absolute CBF values, with accuracy subtly influenced by vascular morphology and tissue optical scattering^8,9. However, these limitations do not undermine LSCI's utility but define its specialized application in real-time assessment of superficial cortical blood flow. For our study focusing on EA-induced cortical perfusion changes, LSCI provides reliable, actionable data. Specifically, we established a standardized LSCI protocol with targeted ROI delineation and optimized parameters to quantify CBF in APP/PS1 mice, complemented by Morris Water Maze tests to link cerebrovascular improvements with cognitive outcomes.

This protocol was approved by the Animal Ethics Committee of Beijing University of Chinese Medicine (BUCM20250708-005), and it was in accordance with all guidelines for the Care and Use of Laboratory Animals of China.

1. Preparation

1. Twelve 6-month-old male APP/PS1 mice and six 6-month-old male C57bl/6 mice were used in this experiment.
2. All mice were housed in the Animal Experiment Center of Beijing University of Chinese Medicine, with three mice per cage. They had free access to sterile drinking water and standard pellet diet.
3. The housing environment was maintained on a 12 h light/dark cycle, with the ambient temperature controlled at 23 ± 2 °C.

2. Animal grouping and interventions

1. Randomly allocate twelve 6-month-old male APP/PS1 mice (weight 27.88 g ± 0.48 g) to two experimental groups (n = 6/ group): the Alzheimer's disease (AD) model group and the electroacupuncture (EA) treatment group.
2. Use six 6-month-old male C57BL/6 mice (weight 29.90 g ± 0.81 g) as the wild-type (WT) group.
3. For the WT control and AD model groups, gently restrain the mice in the well-ventilated mouse restrainers and administer no experimental interventions.
4. For the EA treatment group, gently restrain the mice in the same mouse restrainer as used for the WT control and AD model groups, which limits excessive movement while ensuring unobstructed breathing. Under this standardized restraint condition, insert the disposable sterile acupuncture needles (0.25 mm × 13 mm) into the Baihui (GV20) and Yintang (GV29) acupoints (Figure 1A) on the head, directed toward the nasal region, with a consistent depth of 2-3 mm. Connect the positive and negative terminals of the EA device to these two acupoints, respectively, and deliver stimulation for 15 min (Figure 1B). Configure the stimulation parameters as a disperse-dense wave pattern, with an intensity of 0.1 mA and a frequency of 2 Hz.
5. Administer the interventions to each group once every 48 h, with each session lasting 15 min, for a total duration of 30 days.
   NOTE: Perform acupoint selection, localization, and needling procedures in accordance with T/CAAM 0002-2020: Names and Locations of Commonly Used Acupuncture Points for Laboratory Animals Part 3: Mice¹⁰, a standard issued by the China Association for Acupuncture and Moxibustion on May 15, 2020.

3. MWM test

NOTE: 24 h following the 30-day intervention period, subject all mice across the three experimental groups to the Morris Water Maze (MWM) test, which included both the hidden platform (place navigation) trial and probe trial^11,12. Implement a 1-day acclimation period prior to formal testing to minimize stress, allowing mice to habituate to the experimental room, MWM apparatus, and handling procedures.

1. Preparatory procedures
   1. Position the MWM apparatus and its associated signal acquisition/processing system in a sound-attenuated experimental room. Maintain the ambient temperature of the testing environment at approximately 25 °C, and standardize lighting conditions as follows: mount four 60-W LED tube lights above the four quadrants of the maze at a height sufficient to avoid capture by the ceiling camera. Opaque drapes surrounded the entire testing area to block external environmental interference and prevent unintended spatial cues from affecting the mice's navigation.
   2. Place a circular white tank (90 cm diameter × 50 cm height) at the center of the MWM setup. Connect a ceiling-mounted video camera to an automated tracking system and recorder for continuous data capture throughout testing.
   3. Divide the tank into four equal regions, marked north (N), south (S), east (E), and west (W), and further segment into four congruent quadrants, including northeast (NE), northwest (NW), southwest (SW), and southeast (SE).
   4. Affix distinctly shaped visual shapes (square, triangle, and circle) to the outer wall of each quadrant, positioned within the mice's line of sight to serve as spatial reference points.
   5. Fill the tank with 30 cm of water maintained at 24 ± 1 °C with an electric heater; add white edible pigment to render the water opaque.
2. Hidden platform trial
   1. Fix a submerged platform in the SE quadrant for the duration of this trial.
   2. Randomly release each mouse into the pool from three alternate quadrants (NE, NW, SW), with initial orientation facing the tank wall for consistency.
   3. Allow each mouse a maximum of 60 s to locate the hidden platform independently.
   4. Record escape latency (the time from release to platform ascent) for each trial to assess spatial learning.
      ​NOTE: Each mouse completed three trials per day over five consecutive days. If a mouse failed to find the platform within 60 s, it was gently guided to the platform and allowed to remain there for 10 s to reinforce spatial memory before concluding the trial.
3. Probe Trial
   1. Remove the platform.
   2. Release each mouse once into the pool facing the tank wall and permit 60 s of unconstrained swimming.
   3. Record key outcome measures: average swimming speed and the number of crossings over the former platform location.
4. Post-test care
   1. After each MWM session, dry the mice thoroughly with soft towels and gently warm them using an electric heater to prevent hypothermia.

4. Laser speckle contrast imaging

1. Randomly select three mice from each group and monitor their CBF in real time and in full field using the LSCI system.
2. Animal preparation: Perform all animal operations under anesthesia. Anesthetize mice with 1.25% tribromoethanol injection (0.02 mL/g) intraperitoneally to ensure continuous spontaneous breathing and place them on a horizontal operating plate. Place anesthetized mice on a thermostatically controlled heating pad maintained at 37 °C to stabilize core body temperature. Shave the scalp and carefully dissect the underlying muscles/meninges to expose the skull; keep the exposed cranial surface moist with 0.9% sterile physiological saline throughout the procedure to prevent desiccation.
3. Imaging setup: Position the system's scanning head perpendicular to the exposed cranial window at a fixed distance of 20 cm. Enable the auto-focus and auto-gain functions to optimize image clarity and contrast, with gain values automatically maintained within a range of 140-160 across all experiments. Configure additional standardized imaging parameters as follows: exposure time = 20 ms, time constant = 1.0 s, operating mode = Temporal, filter = 250 frames, sample interval = 5000 ms, and image resolution = 1032 × 772.
4. Data acquisition: Acquire continuous speckle images at a 25 Hz sampling rate, and synchronize collection to the preconfigured temporal and filter parameters.
5. ROI analysis: Use the analysis software to define regions of interest (ROIs) over the prefrontal cortex, specifically targeting the anterior and middle cerebral artery territories within this cortical area. Delineate ROIs over these vascular regions, and extract mean flux values (expressed in perfusion units) from each ROI for subsequent statistical comparison.

5. Statistical analysis

1. Perform statistical analyses using SPSS 27.0.
2. Analyze group differences in the hidden platform, probe trial outcomes, and CBF were analyzed using the Student's t-test. Define statistical significance as P < 0.05, and set high statistical significance at P < 0.01.

Effects of EA on spatial learning and memory in AD model mice

The results of the Morris water maze test are shown in Figure 2. Figure 2A shows the escape latency during the hidden platform test over five days. The escape latency of AD group showed a fluctuating trend, while the WT and EA groups decreased steadily. The escape latency of the AD group was significantly longer than that of the WT group from Day 4 to Day 5 (P < 0.01). The EA group had a significantly shorter latency than the AD group from Day 4 to Day 5 (P < 0.05).

Figure 2B shows the results of the probe test. The swimming speed was lower in the AD group compared to the WT group (P < 0.05). No significant differences were found in the platform crossover number.

Effects of EA on CBF in AD model mice

The results of cerebral blood flow (CBF) measurement using laser speckle contrast imaging are shown in Figure 3.

Figure 3A shows the representative laser speckle images of each group. Figure 3B shows the quantitative analysis of CBF. The CBF in the AD group was significantly reduced compared to the WT control group (P < 0.05). The CBF was significantly elevated in the EA treatment group relative to the AD model group (P < 0.05). 

Figure 1: Localization of target acupoints and EA electrode connection. (A) Location of acupoints Baihui (GV20) and Yingtang (GV29). (B) Connection of positive and negative EA electrodes to GV20 and GV29 acupoints. Please click here to view a larger version of this figure.

Figure 2: Results of the Morris water maze test(n=6). (A) Changes in the escape latency of mice among the different groups in the hidden platform. Quantification of escape latency in each group. ** P < 0.01 vs. WT group; #P < 0.05 vs. AD group. (B)Changes in the swimming speed and platform crossover number of mice among the different groups in the probe trial. Quantification of escape latency in each group. * P < 0.05 vs. WT group. Please click here to view a larger version of this figure.

Figure 3: Results of the Cerebral blood flow(n=3). (A)The color image and flux image of each group. (B)Quantification of the cerebral blood flow in each group. * P < 0.05 vs. WT group. Please click here to view a larger version of this figure.

The present study demonstrates that long-term EA intervention at Baihui (GV20) and Yintang (GV29) acupoints significantly ameliorates spatial learning deficits in APP/PS1 transgenic mice, as reflected by the shortened escape latency in the Morris Water Maze hidden platform test. This finding is consistent with previous reports that EA improves cognitive function in various AD models via mechanisms such as reduced Aβ deposition and attenuated neuroinflammation¹³. Notably, acupoint selection was grounded in Traditional Chinese Medicine theory, which provides a traditional rationale for its application in managing cognitive disorders¹⁴.

A core finding of this study is the confirmation that CBF is significantly reduced in APP/PS1 transgenic mice compared to wild-type controls, as quantified by LSCI. This observation aligns with extensive clinical and preclinical literature documenting cerebral hypoperfusion as a hallmark feature of AD pathogenesis. CBF reduction is reported in both early and progressive disease stages and is closely linked to cognitive decline^2,3. Importantly, our results further demonstrate that long-term EA intervention at GV20 and GV29 effectively elevates CBF in AD model mice, reversing the pathological hypoperfusion observed in the untreated AD group. This finding provides direct experimental evidence that EA exerts a cerebrovascular regulatory effect in AD, complementing its established cognitive-enhancing properties. The restoration of CBF by EA may contribute to the maintenance of neuronal microenvironment homeostasis, facilitating the delivery of oxygen and nutrients to vulnerable cortical regions and potentially mitigating downstream pathological cascades, including Aβ aggregation and neuronal loss, which are hypothesized to be key mechanistic underpinnings of the observed improvements in spatial cognitive function.

LSCI is a valuable tool for cerebrovascular research, offering distinct advantages including non-invasiveness, real-time full-field visualization, and high spatiotemporal resolution that enable reliable assessment of superficial cerebral blood perfusion^6,12. Nevertheless, inherent technical limitations of LSCI should be acknowledged. First, its penetration depth is restricted to superficial cortical layers, precluding direct assessment of CBF in deep brain regions, such as hippocampus, which is critically involved in cognitive processes in AD¹⁵. Second, conventional LSCI applications lack the capacity for precise, region-specific or vessel-specific CBF quantification, which may limit the granularity of hemodynamic analysis. To address these constraints and enhance the objectivity and reproducibility of CBF assessment, the present study established a standardized technical protocol. Instead of random or subjective region-of-interest (ROI) selection, we targeted the anterior and middle cerebral artery territories within the prefrontal cortex for ROI delineation. This systematic approach ensures consistent evaluation of CBF in major cerebrovascular territories relevant to AD, minimizing biases associated with arbitrary ROI selection and providing a more robust quantification of EA-mediated hemodynamic changes. Future studies could integrate LSCI with complementary imaging modalities (e.g., two-photon microscopy, functional MRI) to overcome depth limitations and capture microcirculatory or deep-brain perfusion dynamics, thereby further refining our understanding of EA's cerebrovascular regulatory mechanisms in AD.