Blood pressure (BP) and heart rate were measured in conscious, freely moving Trpa1-ecKO and control Trpa1 ^fl/fl mice through surgical implantation of radiotelemetry transmitters as previously described (Feng et al., 2010; Li et al., 2014b; Thakore et al., 2020; Krishnan et al., 2022); mice were allowed to recover for 2 weeks. Animals were then chronically infused with Ang II (1.2 μg/kg/min) using osmotic minipumps, surgically inserted into the subcutaneous space, and fed a HS (8% NaCl) chow diet (Figure 1A). After 2 weeks of chronic Ang II/HS treatment, the NOS inhibitor L-NAME was introduced into the drinking water (120 mg/kg/day) to increase systemic BP further. Treatment was maintained for 28 days after minipump implantation, and BP, heart rate (HR), and locomotor activity were monitored daily (Figure 1A). Baseline systolic, diastolic, and mean arterial pressures, HR, and locomotor activity didn’t differ between Trpa1 ^fl/fl and Trpa1-ecKO mice. Ang II/HS treatment increased systolic, diastolic, and mean arterial pressure in control and Trpa1-ecKO mice, and BP was further elevated after adding L-NAME to the drinking water (Figures 1B–D). These responses didn’t differ between Trpa1 ^fl/fl and Trpa1-ecKO mice. The treatment didn’t alter HR or locomotor activity for either group (Figures 1E, F). Increased cardiac weight, suggestive of cardiac hypertrophy, was observed in Trpa1 ^fl/fl and Trpa1-ecKO mice treated with Ang II/HS/L-NAME compared to their untreated counterparts (Figure 1G). However no differences between Trpa1 ^fl/fl and Trpa1-ecKO mice were observed, suggesting that endothelial cell TRPA1 expression isn’t directly involved in cardiac hypertrophy associated with hypertension. These data indicate that Ang II/HS/L-NAME treatment induced severe hypertension and that genetic deletion of endothelial cell TRPA1 channels didn’t alter basal hemodynamics nor increases in BP in response to the hypertensive stimuli.

After the end of the treatment (28 days after minipump placement), a cohort of Trpa1 ^fl/fl mice was euthanized, and cerebral arteries were harvested. Trpa1 mRNA transcript (Figure 2A) and TRPA1 protein levels (Figure 2B) didn’t differ between normotensive and hypertensive animals. However, we found that mRNA levels of the ROS-producing enzymes Nox1, Nox2, and Nox4 were enhanced in cerebral arteries from hypertensive mice (Figure 2C). Using a chemiluminescent detection method, we also found that the capacity for superoxide anion (O₂ ⁻) production was elevated in cerebral arteries from hypertensive mice compared with normotensive animals (Figure 2D). We previously reported that administration of the NOX substrate NADPH dilated cerebral arteries by activating TRPA1 channels through lipid peroxidation (Sullivan et al., 2015). Here, we found that NADPH dilated cerebral arteries from hypertensive animals to a greater extent compared with normotensive mice (Figures 2E, F). The selective TRPA1 blocker HC-030031 attenuated NADPH-induced dilation (Figures 2E, F). These data suggest that the capacity for NOX-induced activation of TRPA1 channels and subsequent dilation of cerebral arteries are enhanced during hypertension.

Treatment with Ang II/HS/L-NAME was maintained for 28 days (after minipump implantation) or until mice exhibited signs of morbidity (hunched posture, squinted eyes, ruffled haircoat, dehydration, etc.), including clinical signs suggestive of stroke (circling behavior, head tilt, mobility impairment, muscular fasciculations, other motor dysfunction) (Iida et al., 2005) (Supplementary Videos S1, S2). Approximately 20% of mice from both groups completed the protocol without apparent morbidity (Figures 3A, B). Of the mice that didnt finish the study, about 40% were found deceased in the cage (mortality of unknown cause), 26%–31% displayed neurological dysfunction suggestive of stroke, and between 8% and 15% suffered from generalized, severe subcutaneous edema (Supplementary Figure S1) (Figure 3B). No differences in morbidities were observed between Trpa1^fl/fl and Trpa1-ecKO animals.

The volume of intracerebral hemorrhage lesions is commonly used as a prognostic indicator in stroke patients and is obtained using advanced diagnostic imaging methods, such as MRI (Broderick et al., 1993; Linfante et al., 1999; Kidwell et al., 2004). Due to the technical difficulties associated with high-resolution imaging of mouse brains in vivo and the acute onset of morbidity and mortality, lesion size was measured using a histological approach. Surviving mice and those removed from the protocol because of suspected stroke or subcutaneous edema were perfusion-fixed for histologic analysis. Their brains were removed, sectioned, and stained with hematoxylin and eosin (H and E) to identify sites of intracerebral hemorrhage (Figure 3C). Lesions were detected in 12 of 16 Trpa1^fl/fl mice and 13 of 15 Trpa1-ecKO animals (no significant difference between Trpa1^fl/fl and Trpa1-ecKO mice). Lesions were predominantly found in the cerebral cortex but were also detected in the basal ganglia, midbrain, brain stem, and cerebellum in brains of mice from both groups (Figures 3C and D). The number of lesions per brain did not differ between groups (Figure 3D), but the mean area of individual lesions was significantly smaller in the brains of Trpa1-ecKO mice compared to controls (Figure 3E). Lesions in the cerebral cortex, basal ganglia, and midbrain were significantly smaller in the brains of Trpa1-ecKO mice compared to control Trpa1^fl/fl mice. These data suggest that endothelial cell TRPA1 knockout reduces the size of intracerebral hemorrhages in hypertensive mice, but this effect doesn’t improve survival.

All animal procedures used in the study complied with the Guide for the Care and Use of Laboratory Animals and were approved by the Institutional Animal Care and Use Committees of the University of Nevada, Reno School of medicine, and Colorado State University. Endothelial cell-specific deletion of TRPA1 was achieved by initially crossing mice homozygous for loxP sequences flanking S5/S6 transmembrane domains of TRPA1 (floxed TRPA1; Strain number: 008,654; The Jackson Laboratory, Bar Harbor, Maine) with heterozygous Tek ^Cre mice which express cre recombinase (cre) driven by the promoter for the endothelial cell-specific receptor tyrosine kinase TEK (Strain number: 008,863; The Jackson Laboratory, Bar Harbor, Maine) to produce intermediate heterozygote mice which were used to generate Trpa1-ecKO mice, as previously described (Sullivan et al., 2015; Thakore et al., 2021). Mice homozygous for floxed TRPA1 and positive for cre were considered Trpa1-ecKO, and littermates homozygous for floxed TRPA1 but negative for cre were used as controls (Trpa1 ^fl/fl ). Mice were kept on a 12-h light/dark cycle and were fed standard chow and regular water ad libitum unless otherwise specified. Our prior study demonstrated that TRPA1 expression is undetectable by Western blotting in cerebral arteries from Trpa1-ecKO mice (Sullivan et al., 2015).

At 12 weeks of age, mice underwent telemetry probe implantation as previously described (Feng et al., 2010; Li et al., 2014b; Thakore et al., 2020; Krishnan et al., 2022). Briefly, anesthesia was induced by 4%–5% isoflurane carried in 100% O₂ (flow rate 1 L/min), after which anesthesia was maintained by adjusting isoflurane to 1.5%–2%; pre-operative analgesia was provided by a single subcutaneous injection of 0.05 mg/kg buprenorphine (Zoopharm, Windsor, CO). The neck was shaved and then sterilized with iodine. Under aseptic conditions, an incision (∼1 cm) was made to separate the oblique and tracheal muscles and expose the left common carotid artery. The catheter of a radio telemetry transmitter (PA-C10; Data Science International, Harvard Bioscience, Inc., Minneapolis, MN) was surgically implanted in the left common carotid artery and secured using non-absorbable silk suture threads. The body of the transmitter was placed in a subcutaneous pocket caudal to the right forelimb. After a 14-day recovery period, baseline BP, heart rate, and locomotor activity were recorded in conscious mice for 7 days. Data were collected daily every 10 s for 2 h (from 2 p.m. to 4 p.m.) for the duration of the study protocol and analyzed.

At 15 weeks of age, mice were surgically implanted with an osmotic minipump filled with Ang II (1.2 μg/kg/min; Sigma-Aldrich, St. Louis, MO) and fed 8% NaCl (HS; Envigo, Indianapolis, IN) chow ad libitum. After 2 weeks of Ang II and HS treatment, L-NAME (120 mg/kg/day; Sigma-Aldrich) was added to the drinking water. Treatment was continued for two additional weeks to induce severe hypertension and hemorrhagic strokes. Mice were monitored at least once daily. At the end of the treatment protocol, or sooner if mice showed behavioral signs of a stroke or other morbidity, mice were euthanized and perfusion-fixed in ice-cold 4% paraformaldehyde (PFA) solution for brain histology. Non-fixed brains were also collected from a cohort of Trpa1 ^fl/fl mice for ex vivo analysis.

Cerebral arteries from three animals were isolated and pooled together from control normotensive and hypertensive animals. RNA was extracted and purified using a RNeasy mini kit (Qiagen, Hilden, Germany). A reverse transcriptase reaction was performed using 200 ng RNA per group with a QuantiTect RT kit (Qiagen) to generate complementary DNA (cDNA) samples. Quantitative PCR was performed using a QuantiTect SYBR Green kit (Qiagen). Primers for mouse Gapdh (QT01658692, Qiagen), Trpa1 (QT01595937, Qiagen), Nox1 (QT00140091, Qiagen), Nox2 (QT00139797, Qiagen), and Nox4 (QT00126042, Qiagen) were used to determine respective mRNA levels. This process was repeated twice to generate results from three independent biological samples. The relative expression levels were calculated using the Pfaffl method (Pfaffl, 2001) as previously described (Crnich et al., 2010).

Cerebral arteries were isolated from control normotensive and hypertensive animals and snap-frozen in liquid nitrogen. RIPA lysis buffer (50 µl) (Thermo Fisher Scientific, Waltham, MA) containing a protease inhibitor cocktail (Thermo Fisher Scientific) was added to each artery sample and homogenized by sonication (20 x 1-s pulses) and mechanical disruption by a Fisher Scientific Tissuemiser (10 s) on ice. Samples were centrifuged at 13,000 rpm for 10 min, and the supernatant was transferred to a new tube. Protein concentration for each sample was determined using a BCA Protein assay (Thermo Fisher Scientific). Protein samples (10 ng) were added to SDS sample buffer and heated at 70°C for 10 min. Immediately after denaturation, proteins were separated by SDS-PAGE and transferred to nitrocellulose membranes. Membranes were blocked with 5% milk and 1% BSA in PBS containing 0.1% Tween and 0.02% sodium azide (PBS-TA) for 30 min at room temperature on a rocker and then exposed to a rabbit anti-TRPA1 antibody (1:500, ACC-037 , Alomone Labs, Jerusalem, Israel) in 5% milk, 1% BSA (PBS-TA) overnight at room temperature on a rocker. The same blot was probed independently with a rabbit β-actin antibody (1:1,000, ab8227 , Abcam, Cambridge, United Kingdom). The membranes were then washed with PBS-T for 3 × 5 min and exposed to a goat anti-rabbit secondary antibody (1:10,000, Invitrogen) in 5% milk, 1% BSA (PBS-T) for 2 h at room temperature on a rocker, followed by 5-min washes with PBS-T, incubated in Supersignal ECL substrate (Thermo Fisher Scientific) for 1–3 min, and imaged. Protein bands were quantified using ImageJ software (version 1.52n, National Institutes of Health, Bethesda, MD).

Cerebral arteries were isolated from control normotensive and hypertensive animals. RIPA lysis buffer (50 µL) containing a protease inhibitor cocktail was added to each artery sample and homogenized by sonication (20 × 1-s pulses) and mechanical disruption by a Fisher Scientific Tissuemiser (10 s) on ice. Samples were centrifuged at 13,000 rpm for 10 min, and the supernatant was transferred to a new tube. Protein concentration for each sample was determined using a BCA Protein assay (Thermo Fisher Scientific). The isolated protein (∼50–60 µl) was diluted to 400 µL with HEPES-buffered saline, and lucigenin reagent (5 μM; Cayman Chemicals, Ann Arbor, MI) was added. At this concentration lucigenin doesn’t produce excess superoxide and provides an accurate assessment of the rate of superoxide production (Skatchkov et al., 1999; Munzel et al., 2002; Guzik and Channon, 2005). 200 µL was added to each of two wells in a 96-well plate. Apocynin (100 µM) was added to one of the wells, and the other was treated with vehicle (DMSO). Basal luminescence was measured in relative light units (RLU) every minute for 10 min using a BioTek Synergy H1 microplate reader (Biotek, Winooski, VT). NADPH (200 μM, Sigma-Aldrich) was then dispensed into each well, and luminescence was measured every minute for 50 min. O₂ ⁻ production was calculated as the change in RLU/minute per µg protein. This ex vivo assay measures the capacity for O₂ ⁻ production, but not O₂ ⁻ level in cerebral arteries in vivo.

Isolated vessel experiments were performed as previously described (Sullivan et al., 2015; Wenceslau et al., 2021). Briefly, mouse middle cerebral arteries were isolated and placed in ice-cold MOPS-buffered saline. Arterial segments were then cannulated between two glass cannulas in a chamber (Living Systems Instrumentation, St. Albans City, VT) and secured with single-filament silk. Arteries were superfused with a physiological saline solution (PSS; 119 mM NaCl, 4.7 mM KCl, 21 mM NaHCO₃, 1.17 mM MgSO₄, 1.8 mM CaCl₂, 1.18 mM KH₂PO₄, 5 mM glucose, 0.03 mM EDTA) warmed to 37°C and bubbled with 21% O₂, 6% CO₂, and balanced N₂. Vessels were pressurized to 60 mmHg and gently stretched to simulate physiological conditions. Intraluminal pressure was lowered 20 mmHg and arteries were equilibrated for 15 min. Viability was assessed by superfusing a high [K⁺] PSS (60 mM KCl, 63.7 mM NaCl) to induce vasoconstriction. Follow a 15 min wash, intraluminal pressure was increased to 60 mmHg, and arteries were allowed to develop spontaneous myogenic tone. The % dilation was calculated as the % change in myogenic tone before and after the addition of NADPH.

Brains were removed from perfusion-fixed mice and stored overnight in 4% PFA. Brains were then serially dehydrated in 7.5%, 15%, and 30% sucrose (Sigma-Aldrich) solutions at 4°C and frozen in OCT sectioning medium at −80°C. The tissue was sectioned with a cryostat (Leica Biosystems, Wetzlar, Germany), generating 35 µm thick sections collected every 500 µm. Sections were mounted on glass slides (Superfrost Plus, VWR International, Radnor, PA) and refrigerated overnight. These sections were then stained with hematoxylin and eosin using the H&E staining kit (Abcam, Cambridge, United Kingdom), per the manufacturer’s instructions. Stained sections were sealed using the Cytoseal XYL mounting medium (Thermo Fisher Scientific) and allowed to dry for a minimum of 24 h. Sections were imaged using a bright field microscope (BZ-X710, Keyence Corporation, Osaka, Japan) at ×4 magnification.

Intracerebral hemorrhage lesions were identified as an abnormal appearance of blood within the brain tissue (appears red by eosin staining). Serial coronal sections of brains were viewed by light microscopy, and the number of hemorrhagic lesions within each of the following brain regions was recorded: cerebral cortex, basal ganglia, midbrain, brainstem, or cerebellum. ImageJ image analysis software was used to measure each lesion’s maximum area (mm²).

All summary data are presented as means ± SEM. Statistical analyses and graphical presentations were performed using GraphPad Prism software (version 9.5.0, GraphPad Software, San Diego, CA). The value of n refers to the number of mice per group used for radiotelemetry experiments, survival assays, lesion number quantification, qRT-PCR and Western blots. For lesion area quantification, n refers to the number of lesions from each mouse, and for isolated vessel experiments, n is the number of arteries per group. Statistical analyses were performed using Students paired or unpaired two-tailed t-test with or without Welch’s correction as appropriate, repeated measures or non-repeated measures two-way analysis of variance (ANOVA) with a Šidák correction for multiple comparisons, or Log-rank (Mantel-Cox) test. A value of p < 0.05 was considered as statistically significant.