Differential Effects of High Fat Diets on Resilience to H2O2-Induced Cell Death in Mouse Cerebral Arteries: Role for Processed Carbohydrates

High fat, western-style diets increase vascular oxidative stress. We hypothesized that smooth muscle cells and endothelial cells adapt during the consumption of high fat diets to become more resilient to acute oxidative stress. Male C57Bl/6J mice were fed a western-style diet high in fat and processed carbohydrates (WD), a high fat diet that induces obesity (DIO), or their respective control (CD) and standard (SD) diets for 16 weeks. Posterior cerebral arteries (PCAs) were isolated and pressurized for study. During acute exposure to H2O2 (200 µM), smooth muscle cell and endothelial cell death were reduced in PCAs from WD, but not DIO mice. WD selectively attenuated mitochondrial membrane potential depolarization and vessel wall Ca2+ influx during H2O2 exposure. Selective inhibition of transient receptor potential (TRP) V4 or TRPC3 channels reduced smooth muscle cell and endothelial cell death in concert with the vessel wall [Ca2+]i response to H2O2 for PCAs from CD mice and eliminated differences between CD and WD. Inhibiting Src kinases reduced smooth muscle cell death along with [Ca2+]i response to H2O2 only in PCAs from CD mice and eliminated differences between diets. However, Src kinase inhibition did not alter endothelial cell death. These findings indicate that consuming a WD, but not high fat alone, leads to adaptations that limit Ca2+ influx and vascular cell death during exposure to acute oxidative stress.


Introduction
Western-style diets (WD) are high in fat and processed carbohydrates, which promote obesity [1]. In turn, obesity leads to excessive levels of reactive oxygen species (ROS), thereby inducing oxidative stress in humans and animals [2,3]. Furthermore, consuming processed carbohydrates can augment oxidative stress in obesity [4]. Oxidative stress is associated with cerebral deficits such as memory and behavioral impairments [5], which are linked to apoptosis of neuronal and vascular cells [6]. The consumption of a high fat diet is also a key risk factor for stroke [7,8]. Acute oxidative stress and apoptosis are consequences of ischemic stroke where, upon reperfusion, ROS damage neurons [9] and vascular [10] cells in the brain. Therefore, to limit damage to the cerebral vasculature and the parenchyma it supplies, greater understanding is needed with respect to how the consumption of high fat diets affects the susceptibility of smooth muscle cells (SMCs) and endothelial cells (ECs) to apoptosis when exposed to acute oxidative stress.
H 2 O 2 is common to multiple pathways of ROS production [11] and elicits apoptosis via the intrinsic pathway [12], which is triggered by an overload of intracellular Ca 2+ concentration ([Ca 2+ ] i ) leading to increases in mitochondrial Ca 2+ content and depolarization of mitochondrial membrane potential (∆Ψ m ) [13]. Loss of ∆Ψ m facilitates release of cytochrome C into the cytosol, where it interacts with the apoptosis-activating factor and caspase 9, thereby activating caspase 3, the death protease mediating cell death. In arteries supplying the brain, acute H 2 O 2 exposure promotes Ca 2+ entry through transient receptor potential (TRP) channels, which are critical for eliciting cell death through intrinsic apoptosis [14]. As shown in arteries of skeletal muscle, consuming a WD leads to cellular adaptations that reduce Ca 2+ through TRP channels and thereby attenuates cell death [15]. Whether cerebral arteries adapt to consuming a WD in a similar manner is unknown.
In the present study, we tested whether SMCs and ECs of posterior cerebral arteries (PCAs) from mice consuming a diet high in fat and processed carbohydrate would develop resilience to acute oxidative stress from H 2 O 2 by evaluating cell death, [Ca 2+ ] i and ∆Ψ m responses to H 2 O 2 . To determine whether processed carbohydrate was an integral dietary component to developing resilience, a high fat diet that induces obesity (DIO) [16] but is low in processed carbohydrate was evaluated for reference. Complementary experiments tested whether TRP channels mediate the protection of ECs and SMCs of PCAs during H 2 O 2 exposure.

Animal Care and Use
Experimental procedures were reviewed and approved by the University of Missouri Animal Care and Use Committee (Protocol #17720). Male mice were used for all experiments because vessels from females are more resilient to oxidative stress [15]. Mice were housed on a 12:12 h light-dark cycle at ∼23 • C with fresh water and food available ad libitum. Mice were anaesthetized with ketamine and xylazine (100 kg −1 and 10 mg·kg −1 , respectively; intraperitoneal injection) and killed by decapitation.

Cell Death
Prior to cannulation, pipettes were filled with PSS containing the membrane-permeant nuclear dye Hoechst 33,342 (1 µM; Cat. #H1399, Fisher) to identify all cell nuclei and propidium iodide (2 µM; Cat. #4170, Sigma), which permeates membranes of dead and dying cells; respective dyes were thereby introduced into the vessel lumen [12,15]. Time controls have verified that mouse PCAs studied under these conditions exhibit <1% cell death after 50 min when not exposed to H 2 O 2 [14]. Pressurized PCAs were equilibrated for 20 min in PSS containing vehicle alone or with a pharmacological agent, then exposed to 200 µM H 2 O 2 (Cat. #H1009, Sigma) for 50 min. Following H 2 O 2 exposure, superfusion with fresh PSS resumed while the PSS containing the nuclear dyes perfused the lumen (0.1 mL min −1 , 10 min). Luminal perfusion was halted during H 2 O 2 exposure because luminal flow reduces cell death and the SMC monolayer does not restrict EC access to H 2 O 2 delivered from the bath [12].
Cell death was quantified as described [12,14,15]. Fluorescent images of Hoechst 33,342 (blue) and PI (red) were acquired with a 40× water immersion objective (numerical aperture (NA) = 0.80) using appropriate filters and coupled to a DS-Qi2 camera on an E800 microscope using Elements software (version 4.51; all from Nikon). Z-stack images were obtained through the upper half of a vessel segment. EC nuclei were identified by having an oval shape oriented parallel to the vessel axis while SMC nuclei are and thin and orientated perpendicular to the vessel axis.

Mitochondrial Membrane Potential
Pressurized PCAs were loaded from the bath with the mitochondrial-targeted ∆Ψ m fluorescent indicator tetramethylrhodamine methyl ester (10 nM in PSS; TMRM, Cat. #T668, Fisher) for 30 min preceding H 2 O 2 exposure [20,21] and throughout the protocol. TMRM accumulates in the mitochondrial matrix due to the electronegative potential within these organelles; thus, the intensity of fluorescence decreases with depolarization of ∆Ψ m [22]. Fluorescence images were acquired as described in Section 2.4 at 1 min intervals for 30 min with excitation at 543/22 nm and emission at 592/40 nm. The protonophore carbonyl cyanide 4-(trifluoromethoxy)phenylhydrazone (FCCP, 10 µM; Cat. #C6827, Sigma) was used as a positive control to verify changes in ∆Ψ m [23].

Ca 2+ Photometry
A pressurized PCA was positioned on an inverted microscope (Eclipse TS100, Nikon) and incubated in a static solution of Fura 2-AM dye (diluted to 1 µM in PSS (final [DMSO] = 0.5%); Cat. #F14158, Fisher) for 40 min. Under these conditions, the dye is primarily incorporated into SMCs. Superfusion with PSS was then resumed for 20 min to wash out excess dye. Using a Nikon Fluor 20× objective (NA = 0.45), the vessel was excited at 340 and 380 nm with emission acquired at 510 nm using an IonOptix system with IonWizard 6.3 software [12]. After recording baseline fluorescence, 200 µM H 2 O 2 was added to the superfusion solution. F 340 /F 380 ratios were recorded at 10 Hz for 30 s at 5 min intervals (to limit photobleaching of Fura 2 dye) during 50 min exposure to H 2 O 2 and the ensuing 30 min wash with control PSS.

Data Analysis and Statistics
The intensity of DCFH fluorescence was evaluated using Image J software (version 1.52a; National Institutes of Health, Bethesda, MD, USA) in a region of interest (ROI; 80 µm × 300 µm) located in the center of a vessel. After subtracting background fluorescence, values for ROS generation (i.e., fluorescence accumulation in arbitrary units) reflect the change (∆) from baseline within the ROI over time: ∆ = (fluorescence at x min − fluorescence at 0 min), where x denotes 5 min intervals. The rate of ROS generation (dF/dt) was determined using linear regression, where F is fluorescence and t is time (min). Live and dead cell nuclei were counted manually using Image J software within a 80 × 300 µm ROI, which contained~50 ECs and~50 SMCs [14]. Cell death was calculated as: (# of red nuclei/# of blue nuclei) × 100%. Quantification of ∆Ψ m was evaluated within an 80 × 300 µm ROI by assessing TMRM fluorescence relative to the initial baseline fluorescence (F/F 0 ). [Ca 2+ ] i values within the microscope field of view (~300 × 300 µm) are expressed as the change in F 340 /F 380 (∆340/380) from baseline (0 min) at each 5 min interval following subtraction of background fluorescence recorded before dye loading. Student's t tests or ANOVA (Prism 9, GraphPad Software, La Jolla, CA, USA) were used to analyze data as appropriate with Bonferroni's test for post hoc comparisons. p < 0.05 was considered statistically significant. Summary data are displayed as means ± SE, where n indicates the number of vessels (each from a separate mouse) in an experimental group.

Effects of High Fat Diets on Vascular Oxidative Stress
High fat diets result in weight gain, insulin resistance, and vascular oxidative stress [17,18,27]. In the present study, both WD and DIO augmented weight gain ( Table 1). The baseline ROS production (DCFH fluorescence accumulation) was greater in PCAs from WD vs. CD mice (Figure 1a,b). There was also a trend (p = 0.06) for DIO to elevate ROS production above that of SD mice (Figure 1c,d).

WD, but Not DIO, Increases Resilience to H2O2
Cell death prior to H2O2 exposure was minimal for all groups (1-3%). Following H2O2 exposure (200 µM; 50 min), SMC and EC death were reduced in PCAs from mice fed WD vs. CD mice (Figure 2a-d). In contrast, cell death in PCAs from mice fed DIO was not different from those fed SD (Figure 2e,f), which also had a low incidence of cell death. Depolarization of ΔΨm is a key signaling event mediating cell death [28,29]. Changes in ΔΨm were evaluated with TMRM fluorescence. In PCAs from CD mice, H2O2 progressively depolarized ΔΨm (Figure 3) by ~60% over 30 min (Figure 3a). In PCAs from WD mice, depolarization to H2O2 was reduced to ~30%, illustrating a protective effect of this diet. In contrast, ΔΨm depolarization to H2O2 was not different between PCAs from DIO and SD mice ( Figure 3b); both were similar to the response of PCAs from CD mice. There were no significant differences in baseline TMRM fluorescence among PCAs from mice Depolarization of ∆Ψ m is a key signaling event mediating cell death [28,29]. Changes in ∆Ψ m were evaluated with TMRM fluorescence. In PCAs from CD mice, H 2 O 2 progressively depolarized ∆Ψ m (Figure 3) by~60% over 30 min (Figure 3a). In PCAs from WD mice, depolarization to H 2 O 2 was reduced to~30%, illustrating a protective effect of this diet. In contrast, ∆Ψ m depolarization to H 2 O 2 was not different between PCAs from DIO and SD mice (Figure 3b); both were similar to the response of PCAs from CD mice. There were no significant differences in baseline TMRM fluorescence among PCAs from mice consuming respective between diets. In PCAs from CD mice, FCCP was used as a positive control to depolarize ∆Ψ m (∆F/F 0 = −0.66 ± 0.03, n = 3), and in the absence of H 2 O 2 or FCCP, TMRM fluorescence remained stable for 30 min (∆F/F 0 = −0.05 ± 0.02, n = 4).

WD Attenuates the [Ca 2+ ]i Response Induced by H2O2
Excessive levels of [Ca 2+ ]i contribute to ΔΨm depolarization and vascular cell death [14,30,31]. For PCAs from CD mice, [Ca 2+ ]i increased progressively during H2O2 exposure and nearly recovered during washout (Figure 4a); the peak [Ca 2+ ]i response to H2O2 was reduced by ~50% in PCA from WD mice. In contrast, the [Ca 2+ ]i response to H2O2 was not different between PCAs from DIO vs. SD mice (Figure 4b). Baseline [Ca 2+ ]i was not different between groups and [Ca 2+ ]i remains constant throughout the ~90 min protocol in the absence of H2O2 [14]. Because DIO was not different from SD for cell death (Figure 2e [14]. Because DIO was not different from SD for cell death (Figure 2e

WD Diminishes Ca 2+ Influx through TRP Channels Induced by H2O2
TRPC3 and TRPV4 channels are integral to Ca 2+ entry and cell death in response to H2O2 exposure [14]. Inhibition of TRPV4 channels with HC-067047 (1 µM) nearly abolished SMC death in PCAs from mice fed CD without affecting SMC death in PCAs from WD mice, which were resilient to H2O2 (Figure 5a). EC death was similarly attenuated in PCAs from CD mice, but not from WD mice (Figure 5b). TRPV4 channel inhibition reduced Ca 2+ entry in vessels from CD mice ( Figure 5c) and eliminated differences in cell death between dietary groups. This effect of HC-067047 was not observed in PCAs from WD mice, consistent with their attenuated Ca 2+ entry.

WD Diminishes Ca 2+ Influx through TRP Channels Induced by H 2 O 2
TRPC3 and TRPV4 channels are integral to Ca 2+ entry and cell death in response to H 2 O 2 exposure [14]. Inhibition of TRPV4 channels with HC-067047 (1 µM) nearly abolished SMC death in PCAs from mice fed CD without affecting SMC death in PCAs from WD mice, which were resilient to H 2 O 2 (Figure 5a). EC death was similarly attenuated in PCAs from CD mice, but not from WD mice (Figure 5b). TRPV4 channel inhibition reduced Ca 2+ entry in vessels from CD mice (Figure 5c) and eliminated differences in cell death between dietary groups. This effect of HC-067047 was not observed in PCAs from WD mice, consistent with their attenuated Ca 2+ entry.

WD Diminishes Ca 2+ Influx through TRP Channels Induced by H2O2
TRPC3 and TRPV4 channels are integral to Ca 2+ entry and cell death in response to H2O2 exposure [14]. Inhibition of TRPV4 channels with HC-067047 (1 µM) nearly abolished SMC death in PCAs from mice fed CD without affecting SMC death in PCAs from WD mice, which were resilient to H2O2 (Figure 5a). EC death was similarly attenuated in PCAs from CD mice, but not from WD mice (Figure 5b). TRPV4 channel inhibition reduced Ca 2+ entry in vessels from CD mice (Figure 5c) and eliminated differences in cell death between dietary groups. This effect of HC-067047 was not observed in PCAs from WD mice, consistent with their attenuated Ca 2+ entry. To test the effects of TRPC3 channel inhibition on cellular responses to H 2 O 2 , PCAs were treated with Pyr3 (1 µM), which reduced SMC death (Figure 6a) and EC death (Figure 6b) in vessels from CD, but not WD, mice. Similar to the effects of TRPV4 inhibition ( Figure 5), differences in both EC and SMC death in PCAs were eliminated by Pyr3. Inhibition of TRPC3 channels also reduced the [Ca 2+ ] i response to H 2 O 2 in vessels from CD mice (Figure 6c). The increase in [Ca 2+ ] i was also attenuated by Pyr3 in vessels from WD mice, albeit to a lesser extent than in vessels from CD mice. Furthermore, Pyr3 eliminated differences in [Ca 2+ ] i responses of PCAs to H 2 O 2 between CD and WD mice (Figure 6c). To test the effects of TRPC3 channel inhibition on cellular responses to H2O2, PCAs were treated with Pyr3 (1 µM), which reduced SMC death (Figure 6a) and EC death (Figure 6b) in vessels from CD, but not WD, mice. Similar to the effects of TRPV4 inhibition ( Figure 5), differences in both EC and SMC death in PCAs were eliminated by Pyr3. Inhibition of TRPC3 channels also reduced the [Ca 2+ ]i response to H2O2 in vessels from CD mice (Figure 6c). The increase in [Ca 2+ ]i was also attenuated by Pyr3 in vessels from WD mice, albeit to a lesser extent than in vessels from CD mice. Furthermore, Pyr3 eliminated differences in [Ca 2+ ]i responses of PCAs to H2O2 between CD and WD mice (Figure 6c).

Src Kinases Contribute to Cell Death during H2O2 Exposure
Oxidative stress can activate Src family kinases to enhance TRP channel activity [32]. In PCAs from CD mice, the Src kinase antagonist SU6656 (10 µM) reduced SMC ( Figure  7a), but not EC death (Figure 7b) in response to H2O2. In PCAs from WD mice, SU6656 had no further effect. SU6656 reduced the [Ca 2+ ]i response to H2O2 in PCAs from CD, but not WD, mice (Figure 7c).

Src Kinases Contribute to Cell Death during H 2 O 2 Exposure
Oxidative stress can activate Src family kinases to enhance TRP channel activity [32]. In PCAs from CD mice, the Src kinase antagonist SU6656 (10 µM) reduced SMC (Figure 7a), but not EC death (Figure 7b) in response to H 2 O 2 . In PCAs from WD mice, SU6656 had no further effect. SU6656 reduced the [Ca 2+ ] i response to H 2 O 2 in PCAs from CD, but not WD, mice (Figure 7c).

Discussion
We evaluated the resilience of cerebral arteries from 22 wk old male mice fed westernstyle (WD) and high fat (DIO) diets during 50 min exposure to H2O2 (200 µM). Our key findings are that following ~16 wk of feeding: (1) WD, but not DIO, enhanced basal ROS production in PCAs compared to the respective control diets; (2) WD, but not DIO, attenuated SMC and EC death; (3) WD, but not DIO, attenuated ΔΨm depolarization; (4) WD attenuated Ca 2+ entry through the TRPV4 and TRPC3 channels; and (5) Src kinases contributed to SMC, but not EC, death in PCAs from mice fed the control diet (CD) vs. WD. We propose that cerebral arteries develop resilience to oxidative stress during prolonged consumption of a western-style diet that is high in processed carbohydrates as well as fats. Remarkably, this adaptation preserves vascular cell integrity during acute oxidative stress imposed by H2O2.

Effects of High Fat Diet on Oxidative Stress and Vascular Cell Death
Oxidative stress and cell death are integral to the pathogenesis of vascular disease, stroke, and traumatic brain injury [10,[33][34][35]. Obesity is an independent risk factor that may augment the adverse effects of hypertension, diabetes, and hyperlipidemia on the vasculature [5]. Basal ROS production was greater in PCAs from WD mice vs. CD mice (Figure 1), consistent with the effects of WD in the aorta [36] and skeletal muscle resistance

Discussion
We evaluated the resilience of cerebral arteries from 22 wk old male mice fed westernstyle (WD) and high fat (DIO) diets during 50 min exposure to H 2 O 2 (200 µM). Our key findings are that following~16 wk of feeding: (1) WD, but not DIO, enhanced basal ROS production in PCAs compared to the respective control diets; (2) WD, but not DIO, attenuated SMC and EC death; (3) WD, but not DIO, attenuated ∆Ψ m depolarization; (4) WD attenuated Ca 2+ entry through the TRPV4 and TRPC3 channels; and (5) Src kinases contributed to SMC, but not EC, death in PCAs from mice fed the control diet (CD) vs. WD. We propose that cerebral arteries develop resilience to oxidative stress during prolonged consumption of a western-style diet that is high in processed carbohydrates as well as fats. Remarkably, this adaptation preserves vascular cell integrity during acute oxidative stress imposed by H 2 O 2 .

Effects of High Fat Diet on Oxidative Stress and Vascular Cell Death
Oxidative stress and cell death are integral to the pathogenesis of vascular disease, stroke, and traumatic brain injury [10,[33][34][35]. Obesity is an independent risk factor that may augment the adverse effects of hypertension, diabetes, and hyperlipidemia on the vasculature [5]. Basal ROS production was greater in PCAs from WD mice vs. CD mice (Figure 1), consistent with the effects of WD in the aorta [36] and skeletal muscle resistance arteries [15]. Although there was a similar trend for PCAs from DIO mice, ROS production was not statistically different from SD mice despite similar increases in BW for WD and DIO mice (Table 1). Finding here that WD augments ROS production in mouse cerebral arteries is consistent with the increased oxidative stress in obese humans consuming processed carbohydrate [4]. Because our experimental design exposes a vessel to constant oxidative stress (200 µM H 2 O 2 in the superfusion solution), it seems unlikely that differences in antioxidant capacity are responsible for the differences in cell death associated with WD. However, this possibility cannot be excluded. Chronic oxidative stress can upregulate antioxidant defenses including the transcription factor Nrf2 and thereby protect mitochondria [37,38]. To resolve the question of how WD and DIO may differentially modify the antioxidant response in the cerebral vessel wall will require further study, as will identifying the source(s) of ROS production. Whether or not the protection from H 2 O 2 includes an antioxidant response (or other adaptations), chronic elevation of oxidative stress appears to be integral to greater resilience of SMCs and ECs in the arterial wall during acute exposure to H 2 O 2 .
Vessels from males were studied for the present experiments because those from females are intrinsically protected during H 2 O 2 exposure [12,15]. Finding that H 2 O 2 elicited similar levels of death in SMCs and ECs of PCAs ( Figure 2) differs from our previous observations that ECs are more resilient than SMCs to H 2 O 2 [14]. In agreement with reports of chronic oxidative stress promoting vascular resilience [12,15,19], WD (but not DIO) increased SMC and EC survival during H 2 O 2 exposure, indicating a distinct effect of consuming processed carbohydrates (high fructose corn syrup). Furthermore, depolarization of ∆Ψ m was attenuated in PCAs from WD mice, whereas ∆Ψ m depolarization to H 2 O 2 prevailed in DIO mice (Figure 3). In skeletal muscle, high levels of fructose lead to mitochondrial dysfunctions including decreases in DNA content, impaired energy metabolism, and decreased activity of respiratory complexes [39]. These reductions in functional capacity of mitochondria would be maladaptive to vascular cells yet may be beneficial in the context of limiting ∆Ψ m depolarization to H 2 O 2 . However, as shown in pancreatic β cells, nutrient excess can augment ∆Ψ m [40], which may limit depolarization to H 2 O 2 . Further study is required to identify the specific effects of high fat diets on mitochondrial function in the cerebral vasculature.
A rise in [Ca 2+ ] i can elicit apoptosis through elevating mitochondrial Ca 2+ content and depolarizing ∆Ψ m , resulting in the release of cytochrome C and activation of caspases [13]. In the present experiments, the extent of cell death was related to the progressive rise in [Ca 2+ ] i during H 2 O 2 exposure. That this [Ca 2+ ] i response was reduced in WD vs. CD mice ( Figure 4) but not in DIO vs. SD mice, again points to a role for processed carbohydrates in vascular adaptation to H 2 O 2 exposure.

Changes in TRPV4 and TRPC3 Channel-Mediated Ca 2+ Entry Contribute to Differences in H 2 O 2 -Induced Cell Death
The integral role of TRPV4 and TRPC3 channels as routes of Ca 2+ influx leading to apoptosis in the vascular wall during H 2 O 2 exposure [14] is confirmed by the present experiments. Inhibition of either TRPV4 ( Figure 5) or TRPC3 ( Figure 6) channels limited SMC and EC death in CD, but not WD mice that had adapted by reducing Ca 2+ influx. Finding that either TRP channel inhibitor was able to prevent cell death suggests an interaction between the respective channel subunits. Functional TRP channels are composed of tetramers and both TRPC3 and TRPV4 subunits are capable of forming functional heteromeric channels [41]. While a distinct heteromer remains to be identified in the context of this study, both TRPC3 and TRPV4 form functional tetramers with TRPC1 [42,43]. A key question for future studies is whether (and if so, how) TRP channel expression is affected by WD in vascular cells.
Src family kinases can be activated by oxidative stress [32,44] which has been linked to apoptosis in epithelial cells [45]. These kinases can activate both TRPV4 through phosphorylation of Tyr 110 [32] and TRPC3 channels though phosphorylation of Tyr 226 [46,47]. Unlike the protective effect of TRP channel inhibition on both cell layers of PCAs ( Figures 5 and 6), Src kinase inhibition reduced cell death in SMCs, but not ECs (Figure 7). Given the role of TRPV4 and TRPC3 channels in mediating death of respective cell types during H 2 O 2 exposure, this differential outcome suggests that alternative mechanisms of channel activation are involved in ECs vs. SMCs. Although H 2 O 2 can activate different TRP channel isoforms through the oxidation of cysteine residues [48,49], it remains to be determined whether such activation occurs in ECs. Other signaling events may include the activation of Src kinase by H 2 O 2 or oxidation of Ca 2+ /calmodulin-dependent protein kinase, which can be stimulated by oxidative stress and thereby activate Src kinase [50,51]. Nevertheless, our finding that Src kinase inhibition limited SMC death and attenuated the rise in [Ca 2+ ] i supports a role for Src kinase activity in transducing the signal from H 2 O 2 to SMCs of mouse PCAs.
The mechanism(s) by which WD and chronic oxidative stress reduce Ca 2+ influx and thereby enhance resilience to H 2 O 2 remain(s) to be fully defined. Obesity has been linked to reduced TRPV4-dependent dilation in mesenteric arteries resulting from peroxynitritedependent inactivation of AKAP 150 [52]. It is also possible that adaptations of the plasma membrane facilitate the greater cell survival in PCAs from WD mice vs. CD mice. Both TRP channels [53] and Src kinases [54] can be regulated by local lipid domains. Recent findings show that oxidized phospholipids increase stress tolerance in ECs, thereby limiting cell death [55]. Whether the protective effect of WD on Ca 2+ influx and vascular cell death can be explained by changes in membrane lipids that affect TRP channels and/or Src kinases during acute oxidative stress remains to be addressed.

Conclusions
In cerebral arteries from adult mice, different high fat diets dissimilarly alter the resilience to acute oxidative stress. Whereas a high-fat-diet-induced obesity model did not affect susceptibility to H 2 O 2 , western-style diet, which is high in processed carbohydrates as well as fat, increased vascular ROS production and protected SMCs and ECs during acute H 2 O 2 exposure. This enhanced vascular resilience to oxidative stress is mediated by limiting Ca 2+ entry through TRP channels, with Src kinase having an integral role in SMCs. Although the incidence of ischemic stroke has been reported to be lower in men consuming a high fat diet [56], not all studies agree [57]. The present data suggest that such inconsistencies between studies may reflect the influence of processed carbohydrates in addition to elevated fat consumption.

Conflicts of Interest:
The authors declare no conflict of interest.