Male C57BL/6J mice (8 weeks old, 22–25 g) were purchased from GemPharmatech Co., Ltd. (Nanjing, Jiangsu, China) and housed under specific pathogen-free (SPF) conditions with ad libitum access to standard rodent chow and water. Animals were maintained in a controlled environment at 25 $\pm$ 1 °C and 65 $\pm$ 5% relative humidity, under a 12-h light/dark cycle. The mice were assigned to one of three conditions: (i) sham operation (sham), (ii) SAH induction alone (SAH), or (iii) SAH induction followed by WBV treatment (SAH-WBV).

The SAH model was induced via chiasmatic cistern injection as previously described [24]. Briefly, mice were anesthetized with 5% isoflurane and maintained on 1.5% isoflurane (R510-22-10, RWD, Shenzhen, Guangdong, China) during surgery. Donor mice were positioned supinely, and their thoracic skin was disinfected. Arterial blood was obtained via cardiac puncture of the left ventricle using an insulin syringe. Donor mice were then euthanized by cervical dislocation. Experimental mice were anesthetized, positioned prone in a stereotaxic apparatus, and the scalp was disinfected and incised to expose the anterior fontanelle. A 1 mm diameter hole was drilled approximately 4.5 mm anterior to the fontanelle and 2 mm lateral to the midline. An insulin needle containing donor blood was inserted at a 45-degree angle to a depth of ~4 mm, and 50 µL of arterial blood was slowly injected. The needle was left in place for 2 min prior to its removal, and the skull hole was sealed with bone wax. The surgical area was sterilized, and the skin was sutured. Sham group mice underwent the same procedure without blood injection. All mice recovered in a temperature-controlled environment. For the ligation group, a neck incision exposed the trachea, and the left dCLN was ligated. The control group had dCLN exposure without ligation.

SAH severity was evaluated 24 h post-SAH using a grading scale. The scale assesses six behavioral domains (spontaneous activity, symmetric limb movement, forelimb extension, vertical climbing, trunk touch response, vibrissae tactile sensation), each scored from 0 to 3. Total scores range from 0 to 18. An independent researcher performed all tests and grading. SAH model mice scoring $\le$9 were excluded based on validated criteria.

WBV intervention was administered using a commercial platform (Thermo Cyclo Pad, Model TCP-2021; Niagara Medical, Brisbane, Australia). The parameter was set to a frequency of 28 Hz and a peak-to-peak displacement amplitude of 0.3 mm. Before each session, mice underwent 30-min environmental acclimatization. During WBV, SAH and sham groups were individually housed in ventilated plexiglass chambers (10 $\times$ 10 $\times$ 15 cm) mounted on the platform to standardize posture and restrict locomotion. The SAH-WBV group received continuous vertical sinusoidal vibrations, while the SAH group was placed on an inactive platform. This protocol was administered once daily for 3 consecutive weeks, starting 7 days post-surgery, at consistent circadian times.

Spontaneous activity and anxiety-like behaviors were assessed using an open field apparatus (Model: ZH-SBS Type; Anhui Zhenghua Biological Instrument Equipment Co., Ltd., Huaibei, Anhui, China). Mice freely explored in a blue box (50 $\times$ 50 $\times$ 25 cm) with a defined central zone (20 $\times$ 20 cm). Each mouse was placed in the center and allowed to explore for 5 min. Total distance moved and time spent in the central zone were automatically tracked. The arena was cleaned with 70% ethanol between trials. Data were analyzed using the ZH-SBS Animal Fine Behavior Analysis System (v2.01, Anhui Zhenghua Biological Instrument Equipment Co., Ltd.).

Spatial working memory was evaluated using a standardized Y maze with three identical opaque arms (35 $\times$ 10 $\times$ 15 cm) arranged at 120°. Each mouse was placed in the central hub under 50 lux illumination and allowed to explore freely for 8 min. Spontaneous alternation was defined as consecutive entries into three distinct arms without repetition. Mice with fewer than 8 total arm entries were excluded. Sessions were recorded and analyzed using the ZH-SBS Animal Fine Behavior Analysis System, with arm entry defined as $\ge$90% body displacement into an arm.

The Morris water maze protocol included a 6-day acquisition phase and a probe trial. During acquisition (Days 1–5), mice underwent four trials/day (30-min inter-trial interval) in a circular pool (120 cm diameter, 50 cm height, 22 $\pm$ 1 °C water) containing a hidden platform (10 cm diameter) positioned 1 cm below the water surface in the target quadrant. Spatial cues surrounded the pool. Each trial lasted up to 60 s; mice finding the platform remained for 15 s, while others were guided to it. On Day 6, the platform was removed for a 60-second probe trial. Data, including the time spent in the target quadrant and the number of crossings of the platform area, were recorded. Data were analyzed via the ZH-SBS Animal Fine Behavior Analysis System.

Gait analysis was performed using the CatWalk XT 10.6 system (Noldus Information Technology, Wageningen, Netherlands). Mice were acclimatized for 15–20 min before testing. Each mouse was placed on a 1.1 m glass walkway under dim lighting (15 lux), with paw contacts visualized via total internal reflection of 532 nm green light. A high-speed CMOS camera (FASTCAM Mini UX100, Photron, San Diego, CA, USA) recorded paw prints at 100 Hz. Animals completed a 7-day acclimatization protocol with three daily trials (5–10 min/trial, $\ge$2 h inter-trial interval) before formal testing. Gait parameters were extracted using manufacturer software (run speed variation threshold 60%, complete step sequence required). Analyzed metrics included spatiotemporal parameters (stride length, stance duration, swing speed), dynamic parameters (print area, maximum contact pressure), and interlimb coordination (regularity index, phase dispersion). Raw data and intensity profiles were exported for analysis.

Mice were deeply anesthetized with avertin (240 mg/kg, intraperitoneal injection, DW3101, Dowobio, Shanghai, China) and transcardially perfused with 50 mL ice-cold phosphate-buffered saline (PBS) followed by 50 mL 4% paraformaldehyde (A500684-0500, Sangon Biotech, Shanghai, China). Brains were dissected and post-fixed at 4 °C for 24 h. Tissues were dehydrated through a graded ethanol series (70%–80%–90%–100%, 1 h each), cleared in xylene (10023418, Sinopharm, Shanghai, China), and embedded in paraffin (39601095, Leica, Richmond, IL, USA). Serial coronal sections (5 µm thick) were cut using a rotary microtome (RM2235, Leica). For hematoxylin and eosin (H&E) staining, deparaffinized sections were stained with Mayer’s hematoxylin (SL7050, Coolaber, Beijing, China) for 5 min, differentiated in 1% acid alcohol, rinsed in tap water, counterstained with eosin Y (SL7060, Coolaber) for 2 min, dehydrated, cleared, and mounted with neutral balsam (G8590, Solarbio, Beijing, China). For Nissl staining [25], sections were incubated in 0.1% cresyl violet (C9140, Solarbio) at 50 °C for 10 min, differentiated in 95% ethanol with 0.1% acetic acid (10000218, Sinopharm), dehydrated, cleared, and coverslipped with distyrene plasticizer xylene (DPX, MM1410, MKBio, Shanghai, China). Sections were imaged under bright-field illumination using a Leica microscope (DM2500, Leica) with a 20$\times$/0.7 NA objective lens and a DFC450 CCD camera.

Total RNA was isolated from dura mater using RNAiso™ Plus reagent (9109, TAKARA, Kyoto, Japan). First-strand cDNA was synthesized from 1 µg total RNA using HiScript® II Q RT SuperMix (R223-01, Vazyme, Nanjing, Jiangsu, China). qPCR was performed using NovoStart® Universal Fast SYBR qPCR SuperMix (E401-01A, Novoprotein, Shanghai, China) on a QuantStudio™ 6 Pro system (Applied Biosystems, Carlsbad, CA, USA). Reactions were run in triplicate under: 95 °C for 30 s, 40 cycles of 95 °C for 10 s, and 60 °C for 30 s. Gene expression was normalized to Gapdh and calculated via the 2^{-Δ⁢Δ⁢Ct} method. Primers were designed via PrimerBank (https://pga.mgh.harvard.edu/primerbank/) and synthesized by GenScript Biotech Co., Ltd. (Nanjing, Jiangsu, China). Primer sequences (5^′-3^′) were as follows: Gapdh (NM_001289726.2): Forward: AGGTCGGTGTGAACGGATTTG; Reverse: GGGGTCGTTGATGGCAACA (size: 95 bp). Podoplanin (Pdpn, NM_001290822.2): Forward: GTTTTGGGGAGCGTTTGGTTC; Reverse: CATTAAGCCCTCCAGTAGCAC (size: 154 bp). Lymphatic vessel endothelial hyaluronan receptor 1 (Lyve-1, NM_053247.4): Forward: CAGCACACTAGCCTGGTGTTA; Reverse: GGAAAGGTCTTGCACGAGGTC (size: 87 bp).

The tissues were homogenized in chilled RIPA lysis buffer (KGB5204-100, KeyGEN Biotech, Nanjing, Jiangsu, China) for 5 min using a homogenizer (F6/10, Jingxin, Shanghai, China). Subsequently, the homogenate was centrifuged at 12,000 rpm for 10 min at 4 °C, and the supernatant was transferred to a clean 1.5 mL tube (CFT801015, Jet Biofil, Guangzhou, Guangdong, China). After running sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) gels (ET10010LGel, ACE, Changzhou, Jiangsu, China) in an ice-cold buffer, the proteins were transferred onto a PVDF membrane (WJ002, EpiZyme, Shanghai, China). Immunoblots were probed with the indicated antibodies. The primary antibodies include: Syrian hamster anti-PDPN (1:300, ab11936, Abcam, Cambridge, UK), rabbit anti-LYVE-1 (1:500, ab14917, Abcam), and rabbit anti-GAPDH (1:3000, B8016, Biodragon, Suzhou, Jiangsu, China).

For immunofluorescence staining, the tissue sections were blocked with a solution containing 5% bovine serum albumin (BSA, HYB780, HUAYUN, Guangzhou, Guangdong, China) and 0.3% Triton X-100 (A110694-0100, Sangon Biotech) for 1 h at room temperature (RT). Subsequently, they were incubated sequentially: first with the primary antibody overnight at 4 °C, followed by incubation with a fluorescent-dye-conjugated secondary antibody for 1 h at RT. The primary antibody used in this study was rabbit anti-LYVE-1 (1:500, ab14917, Abcam). The appropriate Alexa Fluor-conjugated secondary antibody was a donkey anti-rabbit antibody (1:1000, abs20021, Absin, Shanghai, China). The nuclei were counterstained with 4′,6-diamidino-2-phenylindole (DAPI, C1002, Beyotime, Shanghai, China) at RT. Fluorescent images were captured using a confocal laser-scanning microscope (LSM880, Zeiss, Oberkochen, Germany).

Mice were anesthetized with 5% isoflurane in 100% oxygen (1 L/min) and secured in a digital stereotaxic frame (68811, RWD) with 2% isoflurane during imaging. The dorsal cervical skin was shaved and aseptically prepared. Ocular lubrication (19-7223, HOYON, Shenzhen, Guangdong, China) was applied. After a midline incision exposed the calvaria, bregma coordinates were zeroed. The left lateral ventricle was targeted (AP –0.35 mm, ML +1.0 mm, DV –2.3 mm). In different experiments, either Evans Blue (EB, 10% w/v, A602025-0005, Sangon Biotech) or Alexa Fluor 647-conjugated ovalbumin (OVA-647, O34784, Invitrogen, Carlsbad, CA, USA) was injected using a microinjection pump at a rate of 0.2 µL/min. The needle remained in place for 5 min post-injection. EB fluorescence was imaged 4 h post-SAH, and OVA-647 fluorescence was imaged 12 h post-injection using an IVIS® Lumina Series III (PerkinElmer, Waltham, MA, USA). Radiant efficiency was analyzed.

Peripheral blood mononuclear cells (PBMCs) were isolated via density gradient centrifugation (725431, NEST Biotechnology, Wuxi, Jiangsu, China). For surface staining, cells were incubated with an FITC-conjugated anti-cluster of differentiation 4 (CD4) antibody (clone RM4-5, 11-0042-82, eBioscience, San Diego, CA, USA; 1:100, 30 min, 4 °C). Cells were washed with PBS, then fixed/permeabilized using the Forkhead box protein P3 (Foxp3)/Transcription Factor Staining Buffer Set (00-5523-00, eBioscience). Intracellular staining used anti-Foxp3 PE (clone FJK-16s, 12-5773-82, eBioscience; 60 min, 4 °C, dark). Data were acquired on a BD FACSAria™ III with FACSDiva™ software (v8.0.1, BD Biosciences, Franklin Lakes, NJ, USA), collecting $\ge$5000 events in the lymphocyte gate. Tregs were defined as CD4⁺/Foxp3⁺ population. Data were analyzed using FlowJo™ (v10.8.1, Agilent, Santa Clara, CA, USA).

Analyses were conducted in GraphPad Prism (Version 10.0.2, GraphPad Software, San Diego, CA, USA). Normality was assessed via Shapiro-Wilk tests ($\alpha$ = 0.05); homogeneity of variance was confirmed by Brown-Forsythe tests. Multi-group comparisons meeting parametric assumptions used one-way ANOVA with Tukey’s post hoc test. Non-normal data used Kruskal-Wallis test with Dunn’s correction. Continuous variables are expressed as mean $\pm$ SEM, with n values indicated in figure legends. Significance was set at p 0.05 (*p 0.05, **p 0.01, ***p 0.001, and ****p 0.0001).

The SAH model was established via chiasmatic cistern injection, and the long-term effects of WBV in mice following SAH were examined (Fig. 1A). Within 1 h post-injection, visible blood accumulation in basal cisterns was observed in the SAH group but not in sham controls (Fig. 1B). Neurological function assessed 24 h post-SAH via the Modified Garcia test showed no significant score differences between SAH and sham groups (Fig. 1C). Motor coordination assessed via the beam walking test revealed significantly prolonged traversal times in SAH mice compared to sham controls (Fig. 1D). H&E staining revealed cerebral edema and dilated perivascular spaces in the cortex and hippocampus of SAH mice (Fig. 1E). WBV treatment was initiated 7 days after SAH induction and administered using a vertical sinusoidal vibration platform operating at a frequency of 28 Hz and an amplitude of 0.3 mm (Fig. 1F). SAH-WBV mice received daily 30-min sessions for 21 consecutive days; control SAH mice were placed on an inactive platform. Open field, Y-maze, and Morris water maze tasks were conducted 3 weeks after WBV intervention. The open field test revealed significant decreases in total distance moved and center time in SAH mice compared to sham controls (Fig. 1G–I), indicating impaired locomotion and anxiety-like behavior. WBV intervention effectively reversed these deficits (Fig. 1G–I). The Y-maze spontaneous alternation task showed no significant differences among groups (Fig. 1J). In the Morris water maze acquisition phase, SAH mice showed impaired spatial learning, with a slower reduction in escape latency across days compared to sham controls. Notably, SAH-WBV mice showed improved learning from days 3–5 (Fig. 1K). During the probe trial, SAH-WBV mice spent significantly more time in the target quadrant than the SAH group, indicating restored spatial memory (Fig. 1L). Collectively, these findings suggest that WBV therapy alleviates cognitive and affective deficits in SAH mice.

Fig. 1.

WBV alleviates SAH-induced cognitive and affective deficits. (A) Experimental timeline. (B) Macroscopic brain morphology. Scale bar: 2 mm. (C) Modified Garcia score (sham, n = 6; SAH, n = 12). (D) The traversal time in the beam walking test (sham, n = 5; SAH, n = 12). (E) H&E staining for perivascular space pathology. Representative images from three replicates. Scale bar: 25 µm. (F) WBV parameters: frequency, 28 Hz; amplitude, 0.3 mm; 30 min/day for 21 days. (G–I) Open field test: [G], Representative locomotor activity graphic; [H], Total distance traveled; [I], Center zone dwell time (sham, n = 11; SAH, n = 6; SAH-WBV, n = 8). (J) Spontaneous alternation rate in the Y-maze test (sham, n = 12; SAH, n = 6; SAH-WBV, n = 8). (K,L) Morris water maze: K, Escape latency during acquisition; L, Probe trial latency (sham, n = 6; SAH, n = 6; SAH-WBV, n = 6). Abbreviations: SAH, subarachnoid hemorrhage; SAH-WBV, SAH mice treated with whole-body vibration; H&E, hematoxylin and eosin. Data are presented as mean $\pm$ SEM. ns, p $>$ 0.05, *p 0.05, **p 0.01, ****p 0.0001.

To investigate the effect of WBV on gait in post-SAH mice, we performed gait analysis (Fig. 2A). Spatiotemporal parameters and dynamic parameters were recorded (Fig. 2B,C). Key gait features, including run duration, average speed, and regularity index, were analyzed. SAH significantly altered run duration, regularity index, and front paw contact area. WBV treatment partially reversed these impairments, significantly improving the regularity index and hind paw contact area compared to the SAH group (Fig. 2D–J). Taken together, these findings indicate that WBV improves locomotor coordination and paw motor control in SAH mice.

Fig. 2.

WBV improves gait behavior in SAH mice. (A) CatWalk XT gait analysis. (B,C) Paw profiles: B, Raw grayscale prints; C, Heatmaps of paw intensity (a.u.) (n = 6 mice/group). (D–J) Statistical analysis of gait parameters: [D], Speed (cm/s); [E], Cadence (steps/s); [F], Run duration (s); [G], Regularity index (%); [H], Maximum variation (%); [I], Front paws (cm); [J], Hind paws (cm). Data are presented as mean $\pm$ SEM. Statistical significance: ns, p $>$ 0.05, **p 0.01, ***p 0.001. RF, right forelimb; RH, right hindlimb; LF, left forelimb; LH, left hindlimb.

MLV drainage is critical for cerebral homeostasis [11]. To assess MLV-mediated blood clearance post-SAH, we examined dCLNs in sham and SAH mice. At 4 h post-injection, erythrocyte infiltration was observed in dCLNs of SAH mice, but not saline-injected controls (Fig. 3A). To evaluate MLV drainage efficacy, 2% Evans blue was injected into the lateral ventricle. In vivo imaging showed reduced fluorescence in dCLNs and increased fluorescence in MLVs of SAH mice compared to sham controls, indicating greater cerebral Evans blue retention (Fig. 3B–D). To further investigate the impact of lymphatic dysfunction on SAH progression, unilateral lymphatic ligation was performed 24 h pre-SAH. Beam walking performance and lateral ventricle area were quantified. Lymphatic ligation exacerbated beam walking deficits in SAH mice (Fig. 3E). Moreover, ligated mice showed lateral ventricle area enlargement post-SAH (Fig. 3F,G). Overall, these findings suggest that meningeal lymphatic drainage via dCLNs is crucial for early blood clearance post-SAH, and impaired MLV function aggravates pathology.

Fig. 3.

SAH impairs meningeal lymphatic drainage. (A) Erythrocyte accumulation in dCLNs. (B) In vivo imaging of 2% Evans blue (5 µL) in MLVs and dCLNs after intracerebroventricular injection for 4 h. (C,D) Quantified fluorescence intensity in MLVs (C) and dCLNs (D) (n = 3 mice/group). (E) Traversal time in the beam walking test (n = 5 mice/group). (F) Nissl staining results. Scale bar: 200 µm. (G) Quantification of the lateral ventricle. Data are presented as mean $\pm$ SEM. Statistical significance: *p 0.05, **p 0.01, ***p 0.001. MLV, meningeal lymphatic vessel; dCLN, deep cervical lymph node.

To investigate the impact of WBV on meningeal lymphatic drainage during SAH recovery, we performed real-time in vivo imaging by injecting OVA-647 (5 µL) into the lateral ventricle following a 3-week WBV intervention. Fluorescent signals were detected in MLVs and dCLNs 12 h post-injection. Quantitative analysis showed significantly higher fluorescence intensity in the periventricular and cervical regions of SAH mice compared to the sham group (Fig. 4A–D). Notably, WBV intervention decreased fluorescence intensity in both MLVs and dCLNs (Fig. 4A–D), suggesting that WBV enhanced macromolecule clearance via the MLVs and dCLNs.

Fig. 4.

WBV promotes meningeal lymphatic drainage in SAH mice. (A,B) Fluorescence intensity in MLVs 12 h post-intracerebroventricular injection of OVA-647 (5 µL). (C,D) Fluorescence intensity in dCLNs. Data are presented as mean $\pm$ SEM. Statistical significance: *p 0.05, **p 0.01, ***p 0.001. OVA-647, Alexa Fluor 647-conjugated ovalbumin.

To explore the mechanisms underlying WBV-enhanced drainage, we assessed Pdpn and Lyve-1 expression, lymphatic markers essential for valve development. The mRNA levels of Pdpn and Lyve-1 were significantly upregulated in MLVs and dCLNs after WBV treatment in SAH mice (Fig. 5A,B). Consistently, the protein levels of PDPN and LYVE-1 were also increased in MLVs after WBV intervention (Fig. 5C,D; the original Western blot WB images can be found in the Supplementary Materials-WB results). Furthermore, immunofluorescence results showed that the LYVE-1-positive density and branching complexity of meningeal lymphatic vessels were increased in the SAH-WBV group compared to the SAH group (Fig. 5E–H). Overall, these findings indicate that WBV promotes meningeal lymphangiogenesis in MLVs and dCLNs.

Fig. 5.

WBV promotes meningeal lymphangiogenesis in SAH mice. (A,B) Relative Pdpn and Lyve-1 mRNA levels in MLVs (A) and dCLNs (B). Normalized to sham (n = 3 mice/group; 3 technical replicates). (C,D) Relative PDPN and LYVE-1 protein levels in MLVs. (E,F) Representative images of staining for LYVE-1 in the MLVs of mice. Scale bar: 100 or 400 µm. (G,H) The LYVE-1-positive density and branch complexity in the MLVs. Data are presented as mean $\pm$ SEM. Statistical significance: ns, p $>$ 0.05, *p 0.05, **p 0.01, ***p 0.001. Lyve-1, lymphatic vessel endothelial hyaluronan receptor 1; Pdpn, podoplanin; DAPI, 4^′,6-diamidino-2-phenylindole‌; GAPDH, glyceraldehyde-3-phosphate dehydrogenase.

Tregs have been reported to promote tissue repair after neural injury [26]. Given that SAH recovery involves extensive tissue regeneration, we hypothesized that WBV might modulate Treg dynamics. To evaluate this hypothesis, we quantified the peripheral blood Treg population in mice after a three-week WBV intervention. Flow cytometry analysis revealed a significant increase in CD4⁺FoxP3⁺ Tregs post-SAH compared to sham controls, further enhanced by WBV treatment (Fig. 6A,B). As Treg cells primarily function through interleukin (IL)-10 secretion, we measured IL-10 levels in the sham, SAH, and SAH-WBV groups. Interestingly, IL-10 levels were significantly higher in the SAH group than in the sham control (Fig. 6C). Moreover, WBV treatment further increased IL-10 levels in SAH mice (Fig. 6C), consistent with the observed expansion of Treg cells in peripheral blood. These findings indicate that Treg expansion may be a mechanism underlying WBV-mediated neuroprotection.

Fig. 6.

WBV increases the peripheral CD4⁺Foxp3⁺ Treg population and IL-10 levels in SAH mice. (A) Flow cytometry of PBMCs: gated on CD4⁺ Tregs. (B) Frequency of CD4⁺Foxp3⁺ cells (n = 4 independent experiments). (C) The IL-10 levels (n = 4). Data are presented as mean $\pm$ SEM. Statistical significance: **p 0.01, ***p 0.001, ****p 0.0001. IL, interleukin; CD4⁺Foxp3⁺, cluster of differentiation 4 and forkhead box protein P3 positive.