All animal procedures reported in this manuscript were performed upon approval by the Institutional Animal Care and Use Committee. The University of Tennessee Health Science Center is accredited by the Association for Assessment and Accreditation of Laboratory Animal Care.

Sprague-Dawley rats (50 g) of both sexes were purchased from Envigo and acclimated for 3 days upon arrival to the University of Tennessee Health Science Center. Then, animals were placed on a high-cholesterol diet (2% cholesterol) in standard rodent food (TD.07684, Harland-Teklad) ad libitum. Rats were used for experimentation after 18 weeks on high-cholesterol diet. By this point, rat blood lipid panel is known to exhibit a significant elevation of total cholesterol due to increase in low-density lipoprotein count [10]. Cholesterol amount per mg of protein within cerebral arteries also increases significantly [10]. While being placed on high-cholesterol diet, rats were randomized into DHA-receiving versus control groups. Animals in the DHA supplementation group received daily oral administration of DHA into the oral cavity at 14 mg/kg of weight using pipette dispenser as described [13]. Placebo group was receiving an equivalent volume of water dispensed via pipette. Daily DHA supplementation results in sexually dimorphic (males only) increases in DHA level per mg of protein within cerebral arteries while females have already higher DHA levels prior to receiving DHA regimen [13].

Cranial window experiments were performed as described by our group [14]. Briefly, male and female rats were initially anesthetized with ketamine/xylazine mixture (91/9 mg/kg of weight, IP injection). Every 15 min, lack of sensation was verified with toe pinch and maintenance ketamine dose was injected (50 mg/kg of weight). Females in estrus were excluded from experimentation until they reach next stage of the estrous cycle. A catheter was inserted in the carotid artery using standard veterinary technique to ensure that the infused alcohol was directed toward cerebral circulation. A cranial window was then made on the side where the catheter had been inserted. The area above the zygomatic arch between the ear and the eye was cleared of hair using Nair hair removal cream. Skin and underlying tissue were dissected out. Then, the bone was removed using a Dremel 4000 instrument. Middle cerebral artery pial branches were monitored using a Leica MC170 HD microscope with a mounted camera (M125 C) connected to a computer monitor. Alcohol was diluted to reach 50 mM in 0.9% NaCl and administered via bolus injection into catheter at 1 mL/250 g of weight at a rate 1 mL/min. Cranial window images before and after alcohol infusion administration were acquired every 60 s. At the end of the experiment rats were euthanized via thoracotomy.

Rats and mice were used as tissue donors. Sprague-Dawley rats (250–300 g) were purchased from Envigo and acclimated for at least 3 days upon arrival at the University of Tennessee Health Science Center. Slo1Y450F knock-in (KI) breeders were initially acquired from the laboratory of Dr. Gregg Homanics (University of Pittsburgh). Mouse colony was then established at the University of Tennessee Health Science Center where mice were bred to render wild-types (C57BL/6J) and slo1Y450F KI on C57BL/6J background from the same parental pair. For cerebral artery diameter monitoring ex vivo, 8–12 weeks-old mice were used.

On the day of the experiment, vaginal swabs were performed to establish the stage of estrous cycle in females. Similar to cranial windows in vivo, females in estrus were excluded from experimentation ex vivo until they reached the next stage of the estrous cycle. For cerebral artery harvesting, animals were deeply anesthetized with isoflurane using drop in a jar approach. Upon loss of toe pinch sensitivity, rats and mice were decapitated using guillotine and sharp scissors, respectively. Brains were immediately placed into an ice-cold physiological saline solution (PSS) of the following composition (mM): 119 NaCl, 4.7 KCl, 1.2 KH₂PO₄, 1.6 CaCl₂, 1.2 MgSO₄, 0.023 EDTA, 11 glucose, 24 NaHCO₃. Middle cerebral arteries were dissected out under microscope (Nikon SMZ645) and de-endothelialized by passing an air bubble into the artery lumen for 90 s. First, one end of the artery was cannulated on a hollow glass cannula within perfusion chamber, this cannula was connected to plastic tubing through which air bubble was pushed to remove endothelium. After this, physiological saline was perfused through the artery lumen in a gentle manner to flush cellular debris. This approach has been validated to remove functional endothelium by the lack of arterial dilation in response to the endothelium-dependent vasodilator carbachol as opposed to endothelium-independent vasodilator sodium nitroprusside [4]. Immediately following de-endothelialization, artery’s second end was cannulated on the other cannula within perfusion chamber (Living Systems Instrumentation). The chamber was continuously perfused at a rate of 1.65 mL/min with PSS that was being equilibrated at pH 7.4 with a 21/5/74% mix of O₂/CO₂/N₂ gases and maintained at 35 °C–37 °C. Artery diameter was monitored via camera attached to an inverted microscope (Nikon Eclipse TS100) with a 10X objective. The external wall diameter was continuously monitored and plotted on a computer screen over time using the automatic edge-detection software (IonWizard by IonOptics). When required for pressurization, intralumenal pressure was changed using different levels of elevation of attached reservoir filled with PSS. Changes in pressure were monitored using a pressure transducer (Living Systems Instruments). Arteries were first incubated at an intravascular pressure of 10 mm Hg for 10 min. Then, intravascular pressure was increased to 60 mmHg and held steady throughout the experiment to trigger development and ensure maintenance of the myogenic tone. The latter was considered present if arteries responded to intralumenal pressure with constriction [15]. Continuous dilation of the artery in response to intralumenal pressure (60 mmHg) was used as a criterion to remove such recording from analysis. Drugs from stocks were prepared in PSS to reach their desired concentration. Drug-containing PSS was continuously pumped into artery perfusion chamber.

DHA ethyl ester for feeding in vivo was purchased from Sigma Aldrich. Single-use aliquots of DHA for experiments ex vivo were purchased from Cayman. Cholesterol, methyl-β-cyclodextrin (MβCD), ethanol (ethyl alcohol, 190 proof) and all other chemicals were purchased from Sigma-Aldrich.

On the day of the experiment, ethanol was diluted to 50 mM. For cranial windows in vivo and pressurized arteries ex vivo, ethanol was diluted in 0.9% NaCl and PSS, respectively. For enrichment of cerebral arteries with cholesterol ex vivo, we followed previously described method [12, 16, 17]. Briefly, cholesterol powder was added to MβCD in PSS to render 0.625:5 mM cholesterol:MβCD (1:8) molar ratio. Resulting suspension was incubated on shaking water bath at 37 °C overnight and filtered the next morning prior to use. Pressurized cerebral arteries were perfused extraluminally with cholesterol:MβCD complex for 15 min to render cholesterol elevation within artery wall [12, 16]. For DHA perfusion ex vivo, stock solution of DHA was first prepared in dimethyl sulfoxide (DMSO), and then further diluted in PSS to the desired concentration. Final content of DMSO in artery bathing solution did not exceed 0.04%.

Analysis of cranial window images obtained in vivo was performed using ImageJ software by the investigator who was blind to experimental group identity. After all diameter measurements were documented, animal group identity was revealed and data from different rats were grouped accordingly. Diameter traces obtained from pressurized cerebral arteries ex vivo were analyzed using IonWizard 4.4 software (IonOptix). Pressurized artery basal diameter (i.e., diameter before drug application) was determined by averaging diameter values during 3 min of continuous recording immediately before drug application. Artery diameter in presence of a specific drug was determined immediately before the perfusion was switched to a washout. Alcohol-induced constriction was measured as percent change of artery diameter by alcohol from basal diameter (pre-alcohol) level.

Statistical analysis was performed using free online tool Statskingdom. With number of observations in experimental groups <10, the mode of data distribution could not be established with certainty. Thus, non-parametric Mann-Whitney test (2-tailed) was used to establish statistical significance. For datasets with n ≥ 10, testing of the mode of distribution (Gaussian versus non-Gaussian) was performed using the free online tool Statskingdom. If distribution was Gaussian, unpaired t-test (2-tailed) was used to determine significance of differences between the groups. If data distribution was not Gaussian, statistical significance was probed using non-parametric Mann-Whitney test (2-tailed). Data from mouse arteries was analyzed by two-way analysis of variance (ANOVA) with genotype (wild-type versus slo1Y450F KI) and treatment (DHA versus vehicle control) as independent factors. In all cases, statistical significance was set at p < 0.05.

Final plotting of data was conducted using Origin 2023 software (OriginLab). Data are expressed as mean ± standard error.

We have previously established that either high-cholesterol diet or DHA oral supplementation protects against alcohol-induced constriction of cerebral arteries [10, 13]. To determine whether combining high-cholesterol diet with daily DHA oral supplementation offers additional benefit against alcohol effect on cerebral artery diameter, we performed cranial windows on rats in vivo. Prior to cranial window surgery, rats were maintained on a high-cholesterol diet (2% cholesterol) for 18 weeks ad libitum. For the duration of high-cholesterol diet, experimental groups of males versus females also received daily gavages with DHA. In control groups, daily supplementation via gavage consisted of distilled water that was isovolumic to gavages with DHA as quantified per animal’s weight. After cranial window installation, arteriole diameter was documented every minute, with changes in arteriole diameter over time being presented as a fold-change from initial diameter, that is, the diameter observed during 1st minute of observation (Figure 1). Baseline variations in arteriole diameter were established over the course of 10 min. Then, alcohol was probed at 50 mM injected into cerebral circulation via catheter in the carotid artery. As expected, high-cholesterol diet protected against alcohol-induced constriction. In males, fold-change in arteriole diameter from the initial timepoint only reached 1.13 ± 0.09 nine minutes after alcohol was infused (Figure 1B). In females, this value was 1.12 ± 0.08 (Figure 1C). Both male and female values of fold-change in diameter by alcohol were not statistically different from baseline variations.

In experimental group of males that were receiving daily DHA oral gavages during the course of high-cholesterol diet, fold-change in arteriole diameter from the initial timepoint averaged 0.95 ± 0.04 nine minutes following alcohol infusion, and it was not statistically different from control group (high-cholesterol diet with daily gavages consisting of distilled water) (Figure 1B). Of note, fold-changes in artery diameter from the initial timepoint were statistically lower in DHA group at a single time point following alcohol infusion (20 min, Figure 1B). In experimental group of females that were receiving daily DHA oral gavages during the course of high-cholesterol diet, fold-change in arteriole diameter from the initial timepoint averaged 1.04 ± 0.07 nine minutes following alcohol infusion, and it was not statistically different from control group (high-cholesterol diet supplemented with daily gavages consisting of distilled water) (Figure 1C). We did not detect statistically significant differences in fold-change arteriole diameter from the initial timepoint between control and DHA groups of females, albeit averaged values in DHA group were consistently lower than in control (Figure 1C).

In neither males nor females, DHA oral supplementation during a high-cholesterol diet exerted a clear effect on alcohol-induced changes in arteriole diameter compared to control group. These findings suggest that cholesterol- and DHA-driven protection of cerebral artery against alcohol-induced constriction may converge on a shared mechanistic pathway (ceiling effect).

We have previously established that either cholesterol-enrichment of arteries ex vivo or their bathing in DHA-containing solution protects against alcohol-induced constriction of cerebral arteries [10, 13]. To determine whether combining cholesterol-enrichment ex vivo with application of DHA-containing solution offers additional benefit against alcohol effect on cerebral artery diameter, we probed pressurized middle cerebral arteries ex vivo. Middle cerebral arteries from rats were dissected out and pressurized to 60 mmHg to develop myogenic tone and allow assessment of pharmacological responses under physiologically relevant conditions [4, 18]. After arteries developed myogenic tone, they were enriched with cholesterol by incubation in a MβCD-cholesterol complex as described [16]. This procedure increases artery cholesterol content 1.5–2.0-fold [12, 16], and was previously shown to protect middle cerebral arteries of rat against alcohol-induced constriction [10]. Moreover, cholesterol level within artery wall remains elevated after MβCD-cholesterol bathing is replaced with regular physiological saline solution. Following removal of MβCD-cholesterol complex from the bath, cholesterol-enriched arteries were bathed in 3 μM DHA, a concentration previously shown to protect against alcohol-induced constriction of ex vivo pressurized middle cerebral arteries from rat [13]. Artery diameter prior to DHA probing in males and females averaged 174 ± 5 μm (n = 33) and 217 ± 8 μm (n = 26), respectively. These diameters did not differ significantly from baseline diameters before DMSO application (see below) which averaged 171 ± 5 μm in males (n = 45) and 198 ± 8 μm in females (n = 32) (p = 0.71 and 0.1 by 2-tail t-test for DHA versus DHA groups from males and females, respectively). Alcohol was probed at physiologically and toxicologically relevant concentrations as follows: 5, 17.5, 50, 75, and 100 mM. In control group (cholesterol-enrichment without DHA), alcohol was probed in a manner that was time-matched to experimental group. Moreover, vehicle control to bathing with DHA (0.04% DMSO) was probed with 17.5, 50, and 100 mM alcohol. Vehicle perfusion (0.04% DMSO) did not alter baseline arterial diameter or alcohol responses.

As expected, high-cholesterol diet precluded alcohol-induced constriction across the entire range of alcohol concentrations in both sexes (Figures 2A,B,D,E). Addition of DHA did not produce statistically significant changes in cerebral artery diameter beyond those observed with cholesterol-enrichment alone despite the fact that our recent observations using the same experimental setting in cholesterol-naïve (not subjected to any manipulation on naturally occurring cholesterol levels) arteries clearly pointed to a protective effect of DHA against alcohol-induced constriction in middle cerebral arteries from male Sprague-Dawley rats [13]. This lack of additive or synergistic effect further reinforces the idea that cholesterol- and DHA-driven protection against alcohol constriction may converge on a shared mechanism. Moreover, that lack of additivity in cholesterol and DHA action during in vivo and ex vivo testing suggests that common pathway of interaction with alcohol is independent of systemic circulation and complex neuronal integrity.

To assess the contribution of endothelial-derived vasoactive factors into this pathway, experiments were repeated following physical removal of the endothelium by intraluminal passing of air bubbles [4, 19]. Artery diameter prior to DHA probing in males averaged 179 ± 7 μm (n = 30), while in females it was 207 ± 10 μm (n = 30). Artery diameter prior to DMSO probing in males averaged 172 ± 6 μm (n = 42), while in females it was 206 ± 10 μm (n = 34). There were no statistically significant differences between these groups (p = 0.48 by 2-tail t-test for DHA versus DHA groups from males; p = 0.85 by 2-tail Mann-Whitney test for DHA versus DHA groups from females). Removal of the endothelium and probing of cholesterol, DHA, and alcohol in deendothelialized arteries (Figures 2C,F) rendered an outcome that was similar to data obtained from arteries with intact endothelium (Figures 2B,E): cholesterol enrichment precluded alcohol-induced constriction across the entire range of alcohol concentration in both sexes, and DHA application to cholesterol-enriched arteries did not evoke any changes to lack of artery response to alcohol (Figures 2C,F). These findings indicate that the potentially shared mechanism underlying cholesterol- and DHA-dependent protection against alcohol-induced cerebrovascular constriction does not require a functional endothelium.

Based on the hypothesis that cholesterol and DHA share a common mechanism underlying their protection against cerebral artery constriction by alcohol, we examined the role of tyrosine 450 (Y450) in the cerebrovascular BK channel-forming α subunit (slo1). Our previous work showed that DHA-driven protection against alcohol-induced constriction of cerebral arteries required the presence of BK channel-forming α subunits [13]. Notably, tyrosine 450 substitution to phenylalanine (Y450F) in this subunit disabled cholesterol-dependent protection against alcohol [12]. If cholesterol- and DHA-driven protections against alcohol shared a molecular pathway, then slo1Y450 would be a strong candidate to mediate DHA protection against alcohol constriction of cerebral arteries. If proven to mediate DHA protection against alcohol, slo1Y450 could then be postulated as a key structural mediator upon which cholesterol, DHA and alcohol actions on cerebral arteries converge.

To test the role of Tyr450 in DHA-driven protection against alcohol, middle cerebral arteries isolated from C57BL/6J mice expressing wild-type BK channel (slo1) and global knock-in with slo1Y450F substitution on C57BL/6J background were pressurized ex vivo at 60 mmHg. Endothelium was removed by passing an air bubble through artery lumen immediately prior to pressurization. Following development of myogenic tone, there were no differences in average diameter between strain and sexes assigned to DHA versus DMSO probing. Namely, artery diameters in C57BL/6L males in DHA group averaged 123 ± 1 μm (n = 8) versus 125 ± 13 μm in DMSO group (n = 7) (p = 0.91 by 2-tail Mann-Whitney test). Artery diameters in slo1Y450F males in DHA group averaged 104 ± 11 μm (n = 7) versus 100 ± 17 μm in DMSO group (n = 6) (p = 0.73 by 2-tail Mann-Whitney test). Artery diameters in C57BL/6L females in DHA group averaged 132 ± 17 μm (n = 6) versus 107 ± 11 μm in DMSO group (n = 6) (p = 0.34 by 2-tail Mann-Whitney test). Artery diameters in slo1Y450F females in DHA group averaged 130 ± 18 μm (n = 6) versus 131 ± 2 μm in DMSO group (n = 6) (p = 0.48 by 2-tail Mann-Whitney test). Arteries were perfused with either 3 μM DHA or vehicle control (0.04% DMSO). Alcohol was probed at 50 mM, and alcohol-induced constriction was measured as percent change of artery diameter from pre-alcohol levels. In wild-type male mice, DHA protected against alcohol-induced constriction. Namely, in vehicle control, alcohol-induced change in artery diameter from pre-alcohol level averaged −1.03% ± 0.42% (Figures 3A,B). In presence of DHA, however, alcohol-induced change in artery diameter averaged 0.46% ± 0.20% (p = 0.03 by two-tailed Mann-Whitney test when compared to vehicle control, Figure 3B). In slo1Y450F knock-in males, alcohol-induced change in artery diameter averaged −0.61% ± 0.28% (Figures 3A,B). In presence of DHA, alcohol-induced change in artery diameter averaged −0.91% ± 0.10% (p = 0.31 by two-tailed Mann-Whitney test when compared to vehicle control; Figure 3B). Further analysis of data from male arteries using two-way ANOVA revealed significant main effect of DHA treatment (p = 0.04) and trending effect of genotype (p = 0.09). Most important, a significant interaction between genotype and DHA treatment was detected (p = 0.004), underscoring that the loss of antagonism on alcohol-induced cerebral artery constriction by DHA treatment is actually determined by the substitution of slo1Y450.

In females, alcohol-induced constriction was not readily observed. In vehicle control, alcohol-induced changes in diameter averaged 0.29% ± 0.59% (Figures 3C,D). In presence of DHA, fold-change in artery diameter by alcohol averaged 0.61% ± 0.35% (p = 0.94 by two-tailed Mann-Whitney test when compared to vehicle control, Figure 3D). Remarkably, middle cerebral arteries from slo1Y450F knock-in females exhibited a pronounced constriction by alcohol: averaged change in diameter by alcohol in this group reached −3.87% ± 0.69% (Figures 3C,D). DHA treatment significantly attenuated this response, reducing the change in artery diameter to 0.27% ± 0.24% (p = 0.002 by two-tailed Mann-Whitney test when compared to vehicle control, Figure 3D). Consistent with pair-wise comparisons, further analysis of data from female arteries using two-way ANOVA revealed significant main effects of genotype (p = 0.001) and DHA treatment (p = 0.002). A significant interaction between genotype and DHA treatment was also detected (p = 0.0008).

Collectively, our findings indicate that Y450 within the BK channel slo1 subunit is a sexually dimorphic sensor of DHA protection against alcohol. Furthermore, combined with previous reports on the critical role of slo1Y450 in mediating cholesterol protection against alcohol in cerebral arteries, slo1Y450 emerges as a potential relay station for cholesterol, DHA, and alcohol actions on cerebral artery diameter.