Apoptosis in resistance arteries induced by hydrogen peroxide: greater resilience of endothelium versus smooth muscle

Abstract

Reactive oxygen species (ROS) are implicated in cardiovascular and neurologic disorders including atherosclerosis, heart attack, stroke, and traumatic brain injury. Although oxidative stress can lead to apoptosis of vascular cells, such findings are largely based upon isolated vascular smooth muscle cells (SMCs) and endothelial cells (ECs) studied in culture. Studying intact resistance arteries, we have focused on understanding how SMCs and ECs in the blood vessel wall respond to acute oxidative stress induced by hydrogen peroxide, a ubiquitous, membrane-permeant ROS. We find that apoptosis induced by H₂O₂ is far greater in SMCs compared to ECs. For both cell types, apoptosis is associated with a rise in intracellular calcium concentration ([Ca²⁺]_i) during H₂O₂ exposure. Consistent with their greater death, the rise in [Ca²⁺]_i for SMCs exceeds that in ECs. Finding that disruption of the endothelium increases SMC death, we address how myoendothelial coupling and paracrine signaling attenuate apoptosis. Remarkably, conditions associated with chronic oxidative stress (advanced age, Western-style diet) protect SMCs during H₂O₂ exposure, as does female sex. In light of intracellular Ca²⁺ handling, we consider how glycolytic versus oxidative pathways for ATP production and changes in mitochondrial structure and function impact cellular resilience to H₂O₂-induced apoptosis. Gaining new insight into protective signaling within and between SMCs and ECs of the arterial wall can be applied to promote vascular cell survival (and recovery of blood flow) in tissues subjected to acute oxidative stress as occurs during reperfusion following myocardial infarction and thrombotic stroke.

INTRODUCTION

Apoptosis is essential to maintaining the balance between survival of cells that are functioning properly and those that have been damaged or are no longer needed. Under adverse conditions, the induction of apoptosis can also contribute to cardiovascular diseases including atherosclerosis, hypertension, heart disease, and stroke (1, 2). In response to injury, apoptosis is initiated through the intrinsic pathway, characterized by permeabilization of the mitochondrial membrane upon triggering oligomerization of the Bcl-2 family of proteins Bax and Bak (3, 4). Formation of a pore spanning mitochondrial membranes leads to Ca²⁺ overload, loss of mitochondrial membrane potential (ΔΨ_m), and release of cytochrome C from the intermembrane space into the cytosol, thereby activating the initiator caspase-9 and executioner caspase-3.

Reactive oxygen species (ROS) in the vasculature include superoxide (O₂^•−), hydrogen peroxide (H₂O₂), and hydroxyl radical (^•OH) produced by NADPH oxidase, mitochondria, and xanthine oxidase; ROS are scavenged by superoxide dismutases, catalase, and peroxidases (5). Our focus on H₂O₂ is based upon it being integral to multiple pathways of ROS production (6). Among ROS, H₂O₂ is nonpolar and able to diffuse across biological membranes and through aquaporin channels (7). Compared to other ROS, its longer half-life enables H₂O₂ to act as a second messenger, mediating vascular cell differentiation, proliferation, metabolism, and vasodilation (8, 9). Low levels of H₂O₂ can protect vascular smooth muscle cells (SMCs) through Akt signaling (10), while excessive levels of H₂O₂ can oxidize proteins and lipids, elicit DNA strand breaks, and thereby act as a pro-apoptotic agent (11), particularly during ischemia-reperfusion (I/R) injuries (12). The ROS produced during I/R provoke mitochondrial fission and alter signaling at endothelial cell (EC) membranes, leading to dysregulated function throughout the microcirculation (13). As shown in the cardiac microvasculature, I/R injury is associated with opening of the mitochondrial permeability transition pore (mPTP) during ischemia (14), with disrupted mitochondria and NADPH oxidases generating excessive ROS upon reperfusion, leading to intrinsic apoptosis.

Studies of vascular cell death have primarily been performed in culture using either ECs or SMCs. Remarkably, little is known of whether SMCs and ECs comprising the wall of intact blood vessels differ in sensitivity to ROS-induced apoptosis. Furthermore, interactions between native SMCs and ECs that may affect resilience to oxidative stress cannot be evaluated in isolated cells. To address these limitations, we have studied the resilience of respective cell types in pressurized mouse resistance arteries to determine: 1) how the response to acute H₂O₂ exposure differs between native ECs and SMCs; and 2) how interrupting the physiological integrity of ECs and SMCs affects apoptosis. We summarize key findings and consider potential mechanisms with an emphasis on mitochondrial size, shape, and function as variables affecting differential susceptibility to ROS injury between SMCs and ECs in vessels integral to blood flow control.

GREATER RESILIENCE OF ECs VERSUS SMCs IN RESISTANCE ARTERIES DURING H₂O₂ EXPOSURE

Nuclear staining of resistance arteries (100–150 μm in diameter) comprised of a monolayer of SMCs circumscribing the endothelium defines apoptosis in respective cell layers of the vessel wall (Fig. 1). During acute exposure to H₂O₂ (200 μm for 50 min at 37°C), ECs demonstrate far greater resilience compared with SMCs; whereas ∼40% of SMCs undergo apoptosis, EC death is <10% (15, 16). Despite relaxing SMCs and protecting ECs from tumor necrosis factor α (17), nitric oxide (NO) exerts a detrimental effect on SMCs in the presence of H₂O₂; inhibiting NO synthase reduced SMC death by more than half through attenuating the production of peroxynitrite (ONOO⁻) (16). Thus, by generating ONOO⁻ upon reacting with superoxide (O₂^•−) in the arterial wall (18), NO is detrimental during acute oxidative stress. Whereas scavenging ONOO⁻ reduced EC death, NO synthase inhibition had no impact on EC survival. Remarkably, intraluminal flow increased the resilience of ECs (but not SMCs) to H₂O₂ (16). This outcome is consistent with findings from human ECs studied in culture, where shear stress inhibited apoptosis (19). It is, therefore, important to consider mechanotransduction pathways other than NO production that could modulate EC resilience in vivo. An apparent candidate is platelet endothelial cell adhesion molecule (PECAM)-1 (CD31), which is also activated by shear stress and inhibits EC apoptosis (20).

Figure 1.

Quantification of vascular cell death. A: propidium iodide staining nuclei of dead cells. B: Hoechst 33342 dye staining nuclei of all cells. C: merged image of A and B. Images acquired with DS-Qi2 camera and Elements software (Nikon) in a pressurized (100 cm H₂O) mouse superior epigastric artery (internal diameter, 150 μm) from a male C57BL/6 mouse following 50 min exposure to H₂O₂ (200 μm) at 37 °C. Endothelial cell nuclei are oval and oriented along vessel axis; smooth muscle cell nuclei are narrow and oriented perpendicular to vessel axis. Scale bars = 50 μm.

Additional signals originating in the endothelium may contribute to greater resilience of ECs compared to SMCs (Fig. 2). Angiopoietins, which are established growth factors in angiogenesis, can protect ECs from H₂O₂-induced apoptosis by signaling through the phosphatidylinositol 3′ (PI3)-kinase/Akt pathway to attenuate Jun N-terminal kinase (JNK) phosphorylation, which otherwise promotes apoptosis (21, 22). In rat cardiomyocytes subjected to hypoxic preconditioning, PI3-kinase/Akt signaling maintained levels of the mitochondrial stabilizing protein Bcl-2, thereby antagonizing intrinsic apoptosis (23). Whether greater maintenance of Bcl-2 in ECs versus SMCs accounts for greater EC resilience to H₂O₂ remains to be determined.

Figure 2.

Endothelial cells (ECs) are more resilient to apoptosis and protect smooth muscle cells (SMCs). Acute oxidative stress results in aberrant Ca²⁺ entry through transient receptor potential (TRP) channels and mitochondrial dysfunction. Mitochondrial Ca²⁺ accumulation and ΔΨ_m depolarization activate the voltage-dependent anion channel (VDAC), thereby contributing to the release of cytochrome C (Cyt C) into the cytosol, which activates caspases. Greater resilience to apoptosis in ECs vs. SMCs may be mediated by shear stress activating mechanoreceptors such as platelet endothelial cell adhesion molecule (PECAM)-1, factors acting on the endothelium such as angiopoietins (ANGPT), or local production of protective factors such as humanin, which inhibits proapoptotic Bax. The endothelium can protect SMCs from cell death. Although nitric oxide (NO) is beneficial under some circumstances, during conditions of oxidative stress it reacts with superoxide (O₂^•−) to form peroxynitrite (ONOO⁻), thereby potentiating Ca²⁺ entry and apoptosis. Other endothelium-derived substances which may mediate protection against reactive oxygen species (ROS) include arachidonic acid (AA) metabolites including epoxyeicosatrienoic acids (EETs) produced by cytochrome P (CYP)-450 and prostaglandins (PGs) generated by cyclooxygenase (COX), and carbon monoxide (CO) produced by heme oxygenase-1 (HO-1). Myoendothelial gap junctions (MEGJs) can also increase resilience to cell death through cell-cell coupling and connexin-mediated intracellular signaling.

Changes in miRNA expression in response to acute oxidative stress may also differ between SMCs and ECs. For example, among the 22 miRNAs which regulate protein phosphatase 2 A (PP2A), miR-17, miR-20, and miR-31 can reduce PP2A activity during oxidative stress (24). Because PP2A signaling promotes EC survival during I/R (25), reducing its activity will promote EC apoptosis. During acute lung injury, miR-34a promotes intrinsic apoptosis in ECs from mice (26). At present, the effects of ROS on miRNA expression in SMCs are not well defined, although miR-25 and miR-138 can reduce expression of the mitochondrial Ca²⁺ uniporter (27). Nevertheless, an array of antiapoptotic miRNAs target proapoptotic mRNAs or are positive regulators of antiapoptotic mRNAs (28). Furthermore, there are proapoptotic miRNAs which target antiapoptotic mRNAs that block proapoptotic RNAs. Identifying which miRNAs are activated by oxidative stress in SMCs and ECs and the full range of ensuing regulatory pathways promoting EC versus SMC resilience to ROS require further study.

DIFFERENCES IN CA²⁺ HANDLING CONTRIBUTE TO GREATER DEATH OF SMCs VERSUS ECs

The regulation of intracellular Ca²⁺ concentration ([Ca²⁺]_i) is integral to homeostasis. Aberrant increases in [Ca²⁺]_i contribute to cell death by disrupting membranes, activating proteolysis, decreasing ΔΨ_m, and initiating apoptosis (3, 16, 29). Exposure to H₂O₂ initiates a progressive rise in [Ca²⁺]_i in both ECs and SMCs, with cell death increasing in parallel with the rise in [Ca²⁺]_i. Thus, greater death of SMCs versus ECs is associated with greater increases in [Ca²⁺]_i (15). Because removal of extracellular Ca²⁺ prevents the rise in [Ca²⁺]_i and abrogates cell death (16), the lethal effect of H₂O₂ occurs secondary to [Ca²⁺]_i influx (Fig. 2). Under conditions where vessels have adapted to chronic oxidative stress (e.g., advanced age, Western-style diet), greater SMC resilience during H₂O₂ exposure is associated with a significant reduction in Ca²⁺ influx and attenuated rise in [Ca²⁺]_i (15, 16). Transient receptor potential (TRP) channels, particularly the TRPV4 isoform, appear to be the predominant pathway for Ca²⁺ influx, contributing to cell death during H₂O₂ exposure, particularly in ECs (15, 30). These findings are consistent with the ability of H₂O₂ to activate and increase sensitivity of the TRPV4 channel in pulmonary and coronary ECs (31, 32). How this signaling pathway adapts during conditions of chronic oxidative stress to attenuate Ca²⁺ influx and cell death remains to be defined.

Mitochondria are integral to intracellular Ca²⁺ signaling. The function and regulation of the mitochondrial Ca²⁺ uniporter and the role of mitochondria as intracellular Ca²⁺ stores are well established (33). Many mitochondrial enzymes are Ca²⁺-dependent and perturbations in mitochondrial Ca²⁺ concentration ([Ca²⁺]_m) lead to dysfunction (34). Accumulation of [Ca²⁺]_m leads to mitochondrial swelling and opening of the mPTP; increased permeability of the inner mitochondrial membrane dissipates ΔΨ_m, which is an established characteristic of intrinsic apoptosis. Mitochondrial depolarization activates the voltage-dependent anion channel (VDAC), thereby contributing to the release of apoptosis-inducing factor (AIF) and cytochrome C into the cytoplasm (35) to activate caspase-9 and intrinsic apoptosis (36). As shown in cerebral vascular SMCs, the activation of transmembrane member 16 A (TMEM16A), a constituent of the Ca²⁺-activated Cl⁻ channel, also increases opening of the mPTP (37). Although these findings provide mechanistic insight into regulation of mitochondrial [Ca²⁺] and membrane permeability in SMCs, the repertoire of Ca²⁺ channels in mitochondria, and possible differences in their expression between SMCs and ECs, remain to be identified. The greater resilience to H₂O₂ and blunted [Ca²⁺]_i response in ECs versus SMCs suggests that mitochondria in ECs are less susceptible to Ca²⁺ overload and depolarization of ΔΨ_m when compared to mitochondria in SMCs. Whether such differences are intrinsic to respective cell types or how they regulate their intracellular milieu are questions for future study.

DIFFERENTIAL ENERGY METABOLISM IN SMCs VERSUS ECs

Energy metabolism is integral to the regulation of ΔΨ_m and apoptosis. As shown in culture, human umbilical vein ECs (HUVECs) have low metabolic activity at rest. When activated during sprouting angiogenesis they rely on anaerobic glycolysis to generate ∼85% of their ATP production (38). Such reliance of ECs on glycolysis promotes the availability of oxygen carried in the bloodstream for diffusion to surrounding parenchymal cells while minimizing endothelial production of ROS generated during oxidative phosphorylation (OXPHOS). In contrast, human coronary artery SMCs derive ∼50% of their ATP from glycolysis both at rest and during cell proliferation (39), with the remainder coming from OXPHOS. Tissue samples isolated from human subjects show the capacity for generating ATP through OXPHOS in SMCs to be ∼1/3 of that in cardiac or skeletal muscle cells, corresponding to respective differences in mitochondrial content (40). Expressed relative to cytoplasmic volume, mitochondrial content is low in ECs (2%–6%) and SMCs (3%–8%) when compared with that in rat and human cardiomyocytes (∼32%) (40, 41). In human cancer cells, repression of OXPHOS with greater reliance on glycolysis is a key component of resilience to apoptosis (42).

The key enzyme in OXPHOS, ATP synthase, is critical for Bax-dependent proapoptotic function (43). In addition, hexokinase (the enzyme initiating glycolysis from glucose) binds to VDAC, competing with Bcl2 family proteins to reduce mitochondrial susceptibility to proapoptotic signals, which thereby limits permeabilization of the outer mitochondrial membrane (44). This interaction can be augmented by Akt phosphorylation which reduces the ability for oxidative stress or aberrant increases in [Ca²⁺]_m to elicit cytochrome C release (45). When compared with parenchymal cells (e.g., cardiomyocytes and neurons) during ROS exposure, greater reliance on glycolysis for ATP production in vascular cells may contribute to maintaining integrity of the blood vessel wall. By preserving vascular function during ROS exposure [per adaptations to advanced age (16) and Western-style diet (15)], the maintenance of blood flow regulation may serve to limit tissue damage otherwise resulting from hypoxia, nutrient deprivation, and metabolic acidosis.

Under physiological conditions, ΔΨ_m drives proton flux from the intermembrane space through ATP synthase into the mitochondrial matrix to produce ATP. If ΔΨ_m depolarizes to a critical level, ATP synthase can reverse direction to become an ATP-dependent proton pump (46). This reversal preserves ΔΨ_m and prevents the activation of VDAC and opening of the mPTP, thereby preventing apoptosis. When OXPHOS is compromised by diminished ΔΨ_m, glycolysis continues to supply the ATP needed to preserve ΔΨ_m. Thus, the greater reliance of ECs versus SMCs on glycolysis (38, 39) may explain their differential susceptibility to acute oxidative stress observed in resistance arteries (15, 16). Mitochondrial respiration changes when cells adapt to being cultured (42). Because current knowledge concerning vascular ROS is largely from cells studied in culture, rigorous examination of ATP-generating pathways used by ECs and SMCs of intact vessels is vital to advancing our understanding of whether and how differences in energy metabolism and regulation of ΔΨ_m may contribute to differential susceptibility to apoptosis during oxidative stress.

ROLE OF MITOCHONDRIAL MORPHOLOGY IN CELL DEATH

Mitochondria undergo continuous fission and fusion, which controls their morphology and number as well as their function and distribution within cells (47). Ultrastructural studies of rat cerebral arteries show that EC mitochondria are small and rounded (diameter, ∼1 µm) whereas quiescent SMC mitochondria are larger and longer (length, ∼13 µm) (48, 49). Although SMCs and ECs have similar mitochondrial content (40, 41), differential responses to oxidative stress may reflect corresponding differences in mitochondrial morphology, metabolic pathways for energy production, and Ca²⁺ handling (15, 16) (Fig. 3). In mouse cortical neurons, larger and longer mitochondria exhibit greater Ca²⁺ uptake capacity when compared with smaller, rounded mitochondria (50). Because increasing [Ca²⁺]_m depolarizes ΔΨ_m (34), the larger and longer mitochondria (with greater Ca²⁺ uptake) in SMCs versus ECs may promote the dissipation of ΔΨ_m during H₂O₂ exposure, corresponding to greater susceptibility of SMCs to apoptosis.

Figure 3.

Mitochondrial metabolism and shape affect apoptosis. During H₂O₂ exposure, greater resilience of endothelial cells (ECs) vs. smooth muscle cells (SMCs) may reflect greater reliance of ECs on glycolysis as a source of energy to preserve ΔΨ_m and prevent the activation of the voltage-dependent anion channel (VDAC). In SMCs having greater reliance on oxidative phosphorylation (OXPHOS), loss of ΔΨ_m activates VDAC, promoting apoptosis. Larger, longer mitochondria are more susceptible to Ca²⁺ overload. Mitochondrial dynamics affecting apoptosis include fusion and fission to affect their size and shape (i.e., surface area and volume) and interactions with the sarcoendoplasmic reticulum (SER) at mitochondrially associated membranes (MAMs).

Irrespective of fission or fusion, depolarization of ΔΨ_m leads to increased ROS generation preceding collapse of mitochondrial ultrastructure (3, 14). Using a novel imaging approach to study mitochondria in mouse embryonic fibroblasts and HeLa cells exposed to the uncoupler CCCP, ΔΨ_m depolarized with a change from tubular to globular shape without undergoing either fission or fusion (51). Tubular networks can protect mitochondria from autosomal degradation during nutrient deprivation (52). Thus, globular structure (independent of changes in organelle volume) may promote apoptosis during acute oxidative stress. Changes in mitochondrial shape can also affect ΔΨ_m as observed in C2C12 myocytes, where inhibition of myoblast differentiation was accompanied by mitochondrial elongation and ΔΨ_m depolarization (53).

Activation of mitochondrial fission was thought to be integral to mitochondrial membrane permeabilization and the induction of apoptosis (54). Nevertheless, fission can reduce the susceptibility to oxidative stress-induced cell death (54). Findings in HeLa cells indicate that, for Ca²⁺-dependent apoptotic pathways, mitochondrial fission reduces Ca²⁺ uptake, prevents mitochondrial Ca²⁺ waves, and increases resilience to apoptosis (55). This protective effect can be explained by disruption of the mitochondrial reticulum into dispersed fragments, which may lack a Ca²⁺ source to initiate apoptosis. As shown in cardiac myotubes, in contrast to the effects of mitochondrial dispersion, apoptotic signaling can be propagated across the mitochondrial reticulum throughout the cell via Ca²⁺ waves, which depolarize ΔΨ_m and permeabilize the mitochondrial membrane to release cytochrome C and activate caspase-9 (56). Studies of cancer cells illustrate that mitochondria can also propagate ROS signaling (57). Thus, mitochondrial coordination of Ca²⁺ and ROS signaling, coupled with dynamic changes in their morphology and function, may become recognized as integral to regulating cell death.

ROLE OF THE ENDOPLASMIC RETICULUM IN CELL SURVIVAL

Coordinated signaling with the endoplasmic reticulum (ER) may also regulate dynamic changes in the motility, structure, and function of mitochondria. The existence of Ca²⁺-rich microdomains at sites of contact between the ER and mitochondria, known as mitochondrial-associated membranes (MAMs), are integral to Ca²⁺ transfer between respective organelles (36) (Fig. 3). In both ECs and SMCs, MAMs are located adjacent to ER Ca²⁺ channels, including inositol 1,4,5-triphosphate receptors (IP3Rs) and VDAC, suggesting that MAMs may be critical for maintenance of Ca²⁺ homeostasis and the regulation of cell death. During IP₃-induced Ca²⁺ release from the ER, these microdomains regulate Ca²⁺ transfer from the mitochondria to the ER (58). However, because Ca²⁺ transfer is bidirectional between organelles, MAMs also facilitate mitochondrial Ca²⁺ overload via the ER. Thus, ER stress can increase mitochondrial ROS and activate downstream cell death pathways (59). This role is supported by altered molecular composition of MAMs in neuronal cells during disease, where there is an increase in the ganglioside GM1 in the ER membrane. GM1 binds Ca²⁺ and modulates Ca²⁺ flux across membranes while facilitating the formation and activation of a Ca²⁺ pore composed of the IP3R1, VDAC1, and Grp75 (60). Hypoxia/reoxygenation injury associated with ROS production in HUVECs augments [Ca²⁺]_m through greater interaction among IP3R1, VDAC1, and Grp75 (61), which lead to mitochondrial Ca²⁺ overload and cell death. Consistently, downregulation of VDAC1 disrupts the IP3R1/VDAC1/Grp75 complex and prevents Ca²⁺ transfer from the ER to mitochondria, thereby attenuating Ca²⁺ overload and cell death (61). Conversely, an increase in IP3R/VDAC association in the ER membrane at MAMs can promote mitochondrial Ca²⁺ overload and cell death. How coupling between mitochondria and the ER may differ between ECs and SMCs remains to be investigated. Nevertheless, because SMC mitochondria favor interaction with the ER (48), having a lower capacity for Ca²⁺ exchange through MAMs of ECs would promote their having greater resilience to oxidative stress. How mitochondria interact with other organelles, such as lysosomes (62), to regulate [Ca²⁺]_m is an emerging topic and may lead to further understanding of apoptotic signaling.

INFLUENCE OF SEX AND PULMONARY HYPERTENSION

Sex and disease can also regulate mitochondrial structure and dynamics. During exposure to H₂O₂, resistance arteries of female mice are more resilient to cell death when compared with those of males (15, 16). Following deprivation of oxygen and glucose, neuronal mitochondria from males have increased expression of the fusion protein, mitofusin 1 (Mfn1) when compared with those from females (63). Thus, as discussed above, mitochondrial fusion may augment sensitivity to acute oxidative stress-induced death by promoting Ca²⁺ overload in cells of males more readily than those in females. Mitochondria isolated from brains and hearts of female rats exhibited lower Ca²⁺ uptake compared with those from males (64, 65). Although estrogen can reduce Ca²⁺ retention in brain mitochondria from both sexes (65), the prosurvival effects of estrogen are more readily apparent in females (66). Such differences between sexes could attenuate Ca²⁺ overload and ΔΨ_m depolarization and thereby reduce vascular cell death in females compared with males.

In humans with pulmonary hypertension, pulmonary artery SMCs have smaller mitochondria compared with SMCs from healthy subjects, attributable to increased dynamin-related protein-1 (DRP1) activity mediated by hypoxia-inducible factor (HIF)-1α eliciting mitochondrial fission (49). Pulmonary hypertension also reduces the expression of mitochondrial Ca²⁺ uniporter channels, leading to reduced levels of [Ca²⁺]_m (27). These changes to SMC mitochondria contribute to proliferation and reduced apoptosis of pulmonary artery SMCs. However, the effects of sex and cardiovascular disorders on mitochondrial morphology, motility, and Ca²⁺ dynamics on vascular cell susceptibility to death in response to oxidative stress await further study.

CELL-CELL INTERACTION PROMOTES SMC SURVIVAL DURING ACUTE OXIDATIVE STRESS

Until recently, the effect of acute oxidative stress on vascular cell death had not been addressed in intact blood vessels. To address how SMCs and ECs may interact in the intact vessel wall, we studied small resistance arteries isolated from mouse skeletal muscle having a single layer of each cell type (Fig. 1). During exposure to H₂O₂, disrupting the endothelium significantly increased SMC death (15, 16), while removal of SMCs significantly increased EC death (16, 30). Per earlier discussion, the increase in cell death was consistently accompanied by corresponding increases in [Ca²⁺]_i. Control experiments confirmed that, in the absence of H₂O₂, selective removal of one cell layer did not damage the remaining cell layer or affect its resting level of [Ca²⁺]_i.

In resistance vessels, SMCs and ECs signal reciprocally by direct coupling through gap junctions and the release of paracrine substances (Fig. 2). Gap junctions are composed of connexons (each comprised of six connexin protein subunits) in the plasma membrane of opposing cells. The docking of connexons between ECs and SMCs forms myoendothelial gap junctions (MEGJs) (67) that facilitate the diffusion of ions and small molecules (<1 kDa) between respective cell layers. Inhibiting gap junction channels with carbenoxolone increased SMC death to H₂O₂ (16), suggesting that myoendothelial coupling promotes SMC resilience in resistance arteries. In contrast, carbenoxolone had no effect on EC death, which was several-fold less than SMC death under identical conditions. Because of their greater resilience to H₂O₂, coupled with the ability of Ca²⁺ to diffuse through MEGJs and between adjacent ECs (68), the endothelium may serve as a Ca²⁺ sink to buffer the aberrant elevation of [Ca²⁺]_i in SMCs induced by H₂O₂.

An alternative explanation is that connexons may affect cell death independent of cell-cell coupling. When their pore is open, connexons form a low-resistance pathway that leads to intracellular depletion of small molecules such as ATP, glutamate, and NAD⁺, which are all essential to cell survival (69). In neurons, ischemia can open connexons (70), which may thereby potentiate cell death. Independent of pore formation, the phosphorylation state of connexin subunits has also been implicated in cell survival (71). Whether either of these regulatory events happen in vascular cells during oxidative stress remains to be determined, as does a role for other junctional proteins integral to cell signaling, including tight junctions, anchoring junctions, and desmosomes. Also to be determined is whether SMC and EC interactions with the extracellular matrix and/or internal elastic lamina exert protection from ROS.

Multiple autacoids are produced by ECs that can promote their resilience to H₂O₂. Through paracrine signaling, these same signaling molecules can promote survival of surrounding SMCs. Humanin is a mitochondrial-derived peptide produced by ECs (but not by SMCs) (72). Across species, levels of humanin are linked to increased lifespan (73). Humanin acts on G-protein coupled receptors to activate cytoprotective signaling pathways such as Akt (74, 75) while interfering with proapoptotic proteins Bid and Bax (76). By attenuating ROS production, humanin protected human aortic ECs from oxidized LDL-induced apoptosis (72). Studies of myocardial I/R injury indicate that humanin exerts protection through reducing ROS production at Complex I of the electron transport chain (77). Humanin also promotes SMC survival. Thus, treating cultured human cerebrovascular SMCs with humanin rescued them from amyloid β protein-mediated toxicity, a hallmark of degeneration during Alzheimer’s disease (78). Furthermore, pretreating mice with humanin protected against cerebral I/R injury (79). Another endothelium-derived autacoid is the signaling gas carbon monoxide produced by heme oxygenase-1, which can promote EC survival through the Nrf2 pathway by blocking the release of cytochrome C, inhibiting apoptosis (80). Carbon monoxide also promotes cell survival by stimulating the Akt pathway (10) discussed earlier. In addition, prostaglandins and epoxyeicosatrienoic acids (EETs), which derive from arachidonic acid in ECs, exert anti-inflammatory, cytoprotective effects via activation of peroxisome proliferator-activated receptor γ (PPARγ) (81, 82).

CONCLUSIONS

How oxidative stress affects individual SMCs and ECs and their organelles has been well studied in cell culture. However, much less is known of how respective cell layers of the intact vessel wall respond to oxidative stress. In mouse resistance arteries acutely exposed to H₂O₂, we consistently find greater resilience to ROS-induced cell death in ECs versus SMCs. Potential mechanisms contributing to differential susceptibility of respective cell types include reliance on glycolysis versus oxidative phosphorylation, mitochondrial structure and function, coupling between intracellular organelles, intracellular Ca²⁺ handling, and their respective roles in modulating apoptosis. These variables may also contribute to greater cellular resilience in vessels of females versus males. Each of these relationships require further study to understand their role in regulating vascular cell death. Apparent mechanisms contributing to the protection of SMCs by ECs in the vascular wall include myoendothelial gap junctions and local diffusion of paracrine factors. Due to the prevalence of ROS-associated cell death in ischemic events such as myocardial infarction and ischemic stroke (2, 13), understanding mechanisms that protect the vasculature from apoptosis is vital to limiting damage from these catastrophic events while promoting the integrity of vascular perfusion during recovery.

GRANTS

This work was supported by American Heart Association Grant 19TPA34850102 and National Heart, Lung, and Blood Institute Grant R37-HL-041026.

DISCLAIMERS

The content of this article is solely the responsibility of the authors and does not necessarily represent the official views of the American Heart Association or the National Institutes of Health.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

R.L.S. and C.E.N. prepared figures; R.L.S. and C.E.N. drafted manuscript; R.L.S., C.E.N., and S.S.S. edited and revised manuscript; R.L.S., C.E.N., and S.S.S. approved final version of manuscript.