CrossTalk opposing view: Acute exercise does not elicit damage to the endothelial layer of systemic blood vessels in healthy individuals

The integrity of the arterial endothelial monolayer is critically linked to its ability to properly maintain vascular health and function. Therefore, the endothelium of healthy individuals is thought to remain in a stable, quiescent state, with a low rate of cell turnover (Woywodt et al. 2002). Endothelial cell (EC) injury in response to prolonged/repeated exposure to cardiovascular risk factors causes endothelial dysfunction, and increased EC turnover due to detachment of damaged ECs from the vessel wall is recognized as a hallmark of atherogenesis (Versari et al. 2007; Pober et al. 2009). Exercise training combats the development of cardiovascular diseases through direct and indirect beneficial effects on the endothelium. It is generally accepted that the accumulated response to episodic increases in laminar shear stress along the endothelium is largely responsible for the functional and structural vascular adaptations to training (Green et al. 2017). In vitro studies additionally show that laminar shear stress induces a healthy, quiescent phenotype in cultured ECs (Chistiakov et al. 2017). In contrast, a handful of papers have postulated that aerobic exercise causes endothelial damage, potentially mediated by elevations in shear stress (O'Sullivan, 2003; Mӧbius-Winkler et al. 2009; Uhlemann et al. 2014). This hypothesis is controversial, considering that the literature overwhelmingly supports protective effects of elevated laminar shear stress, while vascular regions exposed to disturbed flow/low shear exhibit the highest rates of EC turnover and atherosclerotic lesions. Four decades of research has validated the detrimental role of low shear after a debate emerged around 1970, when two competing theories arose to explain the localization of atherosclerosis. One contended direct endothelial damage and denudation by high shear stress, shown experimentally with supra-physiological levels (∼400 dynes cm−2) in dog thoracic arteries, and the other implicated low shear stress (Malek et al. 1999). The high levels of shear stress needed to cause endothelial damage do not appear to occur in healthy humans, with normal ranges of 10–70 dynes cm−2 in arteries and 1–6 dynes cm−2 in veins (Malek et al. 1999). Shear stress is actually a weak force as compared to the stretching tension caused by blood pressure and, as such, ECs require highly sensitive mechanisms to sense it (Green et al. 2017). Moreover, ECs that have replaced injured and detached cells may be dysfunctional (Vanhoutte, 2010). These observations do not suggest the potential for a beneficial, adaptive role of exercise-induced endothelial damage (by elevated shear stress or otherwise) and subsequent repair similar to that observed in skeletal muscle (i.e. a repeated bout effect). Endothelial damage in humans is assessed by the quantification of biomarkers in circulation that are supported by research on clinical populations with vascular diseases. There are inherent issues regarding the specificity of these markers in healthy individuals, however, and the few studies on exercise do not collectively provide convincing evidence of endothelial damage. Though these biomarkers are not specific to areas of the vascular tree (i.e. conduit vs. resistance vs. microvessels), in the context of this discussion it is presumed that they primarily arise throughout the arterial vasculature, where exercise-induced increases in blood pressure and shear rate are much greater than in the venous system. In this article acute exercise refers to any single bout of exercise, as studies have examined a wide range of intensities and durations (maximal exercise test to 246 km run). The most direct circulating biomarker of disrupted endothelial integrity is circulating endothelial cells (CECs), which should increase in response to mature EC detachment from the vessel wall. To date, there are four reports on CEC count following acute exercise in healthy individuals, of which one has observed increases. In that study, CECs identified as CD146+ cells via flow cytometry increased in trained cyclists after 2 h and peaked 3.5 h into a 4 h cycling bout (Mӧbius-Winkler et al. 2009). While CD146 is a marker of mature ECs, it is also present on activated T-lymphocytes, trophoblasts and mesenchymal stem cells (Erdbruegger et al. 2010), which potentially confounds the interpretation that shedding of ECs occurred. Thus, in order to accurately identify CECs it is recommended to use a combination of cell markers (Erdbruegger et al. 2010). All other authors, using diverse markers for CECs, have reported no changes immediately following intense cycling or running (O'Sullivan, 2003; Obeid et al. 2015; West et al. 2015). Endothelial microparticles (EMPs) are small (≤1 μm), membrane-shed fragments released from activated (CD62E+) or apoptotic (CD31+/CD41− or CD42−) ECs. Studies generally find no change in CD62E+ EMPs after acute exercise (Mӧbius-Winkler et al. 2009; Wilhelm et al. 2016), though two described training-specific (Sossdorf et al. 2011) and sex-specific increases (Lansford et al. 2016). The role of CD62E+ EMPs in healthy individuals is unclear. CD62E (E-selectin) is induced in ECs during activation, so increased CD62+ EMPs may reflect changes related to an inflammatory response, and not EC injury (Pober et al. 2009). Investigations of CD31+/CD42b− EMP count in young, healthy individuals have found no change (Lansford et al. 2016) or decreases (Wahl et al. 2014) after exercise. While exercise studies in humans are lacking, it is important to note that laminar shear stress on cultured ECs suppresses apoptosis (Dimmeler et al. 1999) and reduces both CD62E+ and CD31+/CD42a− EMPs (Kim et al. 2015). Conversely, disturbed blood flow by cuff occlusion increased both EMP types (Jenkins et al. 2013). Thrombomodulin (TM) and von Willebrand Factor (vWF) are membrane glycoproteins expressed on the luminal surface of ECs. Damage to the endothelial plasma membrane should result in leakage of these proteins into the circulation. Additionally, vWF is stored in Weibel–Palade bodies and is released in response to endothelial injury (Chong et al. 2003). Intense exercise, such as the Boston Marathon, has been shown to significantly increase plasma vWF levels (Siegel et al. 2001). The effects of shorter duration exercise on vWF levels in circulation are equivocal, inducing increases or no change (O'Sullivan, 2003; Kanazawa et al. 2006). These contradictions and the fact that circulating vWF is an acute phase reactant, is released by platelets and does not consistently correlate with other markers of endothelial damage (Chong et al. 2003) limit any conclusions. Other mechanisms have also been proposed to mediate exercise-induced increases in vWF, including adrenergic receptor-mediated stimulation and shear stress independent of endothelial damage (Chong et al. 2003; Gonzales et al. 2009). TM is present in circulation in its soluble form (s-TM) in very low concentrations (ng mL−1). Plasma s-TM was increased following a maximal exercise test in one investigation (O'Sullivan, 2003), but was not altered in others (Kanazawa et al. 2006, 2007). Strikingly, immediately after an ultra-endurance 246 km foot race, s-TM was increased from resting levels of 4.0 ± 0.3 ng mL−1 to only 6.9 ± 0.7 ng mL−1 (Bartzeliotou et al. 2007). Thus, the influence of exercise on TM release from the endothelium seems minimal and may not result from lysis, given increases in circulating s-TM can also reflect increased endothelial membrane expression (Bartzeliotou et al. 2007) or may derive from other cell types that express TM, such as platelets and monocytes. A number of endothelium-specific/enriched microRNAs (miRs) have been quantified in circulation in response to exercise (Sapp et al. 2017). Uhlemann et al. (2014) found that the pro-angiogenic circulating (ci)-miR-126 increased in response to a cycling test, 4 h cycling and a marathon. The authors argue that this provides evidence of EC lysis during exercise and propose ci-miR-126 as a novel marker of endothelial damage. However, miRs are also known to be actively secreted from ECs in vitro, and the authors provide no corroborating evidence that endothelial damage occurred. Indeed, the mechanisms responsible for endothelial miR secretion in vivo are largely unexplored. Wahl et al. (2016) recently showed that the endothelial-enriched miR-21 and -126 increase in circulation after exercise, and in EMPs of apoptotic ECs in culture. These findings presumably represent separate mechanisms of secretion, as the same authors previously presented data suggesting EMPs are taken up by ECs (decrease in circulation) during exercise and prevent apoptosis (Wahl et al. 2014). Potential mechanisms responsible for active endothelial miR secretion include redox signalling and laminar shear stress (Sapp et al. 2017). If endothelial damage does occur with exercise, a relatively sustained post-exercise endothelial dysfunction should be apparent. However, Tinken and colleagues (2009) showed that shear stress is responsible for an increase in endothelial function, as measured by flow-mediated dilatation (FMD), following handgrip exercise, cycling and forearm heating. Although similar studies support immediate exercise-induced increases in FMD (Green et al. 2017), other evidence suggests a biphasic response, in which FMD is decreased acutely after exercise, followed by a rebound above resting levels after ∼1–24 h and return to resting levels by ∼24–48 h post-exercise (Dawson et al. 2013). This may be explained by differences in exercise intensity and shear stress patterns. Decreased FMD is usually reported immediately after higher intensity exercise (>80% ), while increases are typically reported following low to moderate intensity exercise (50–80% ) (Dawson et al. 2013). Though shear stress increases in a step-wise manner with exercise intensity, there is no evidence that shear reaches some threshold during intense exercise where it switches from a beneficial to endothelium-damaging stimulus and is thus responsible for reduced endothelial function. Reactive oxygen species (ROS) production and sympathetic nervous system activation both also increase substantially during strenuous exercise, which may cancel out immediate beneficial effects of elevated shear stress and may lead to the observed reductions in FMD (Dawson et al. 2013; Green et al. 2017). For example, the immediate decrease and delayed increase in endothelium-dependent dilatation following intense exercise (85–90% ) was mirrored in rat aortas in response to acetylcholine (Haram et al. 2006), but the transient reduction in dilatation was abolished when arteries were incubated with the ROS scavenger superoxide dismutase (Haram et al. 2006). Further, human investigations reporting a post-exercise decrease in FMD primarily utilize high-intensity cycling, which induces retrograde shear stress in the upper limb vasculature, where FMD is conventionally assessed (Green et al. 2017). Retrograde shear has been shown to reduce FMD in a stepwise fashion (Thijssen et al. 2009), potentially mediated by increased oxidative stress, while antegrade shear, which stimulates release of vasoactive molecules such as nitric oxide (NO), exerts the opposite, beneficial effect (Dawson et al. 2013). Given endothelium-dependent dilatation often returns to or rises above pre-exercise levels by ∼1 h even after strenuous exercise, it is unlikely that damage and repair mechanisms occur in such a short time. Intraluminal pressure is another important haemodynamic factor that increases with exercise and causes circumferential strain to ECs. Specifically during resistance exercise, arterial blood pressure can reach 200–400 mmHg, which directly causes a transient impairment in endothelium-dependent dilatation immediately after exercise (Robinson et al. 2017). This has been explained mechanistically to occur through increased ROS production and subsequent decreased availability of NO. Further, no decrease in endothelial function is apparent in trained individuals following acute resistance exercise, due to adaptations (increased oxidative enzymes) facilitating a switch in vasodilating agent from NO to H2O2 (Robinson et al. 2017). In conclusion, though mechanisms such as oxidative stress are implicated in transient depression of endothelial function (≤1 h) following intense exercise, the available data do not support a damage-inducing effect of exercise on the endothelium or injurious effects of shear stress that may mediate this process. Readers are invited to give their views on this and the accompanying CrossTalk articles in this issue by submitting a brief (250 word) comment. Comments may be submitted up to 6 weeks after publication of the article, at which point the discussion will close and the CrossTalk authors will be invited to submit a ‘LastWord’ Please email your comment, including a title and a declaration of interest, to jphysiol@physoc.org. Comments will be moderated and accepted comments will be published online only as ‘supporting information’ to the original debate articles once discussion has closed. Ryan Sapp is currently a PhD student in the Department of Kinesiology at the University of Maryland. His doctoral research under the guidance of James Hagberg is focused on the mechanisms by which acute and chronic exercise modulate circulating microRNAs and endothelial cell functions. James Hagberg, PhD, is a Professor of Kinesiology and Medicine who has a long history of investigating the effects of acute and chronic exercise on numerous cardiovascular disease risk factors in a wide range of populations. More recently he has focused on the molecular bases for the benefits of exercise on the repair and maintenance of the endothelium. Disclaimer: Supporting information has been peer-reviewed but not copyedited. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article. The authors declare no conflicts of interest. Both authors have approved the final version of the manuscript and agree to be accountable for all aspects of the work. All persons designated as authors qualify for authorship, and all those who qualify for authorship are listed.