Citations

High doses of adenosine (ADO) infused over a prolonged period (3 hours) produces marked vasodilation. However, this vasodilation wanes over time in an isolated canine skeletal muscle preparation. To our knowledge the waning of vasodilation with high doses of ADO has not been shown in humans. In addition, ADO plays a controversial role in exercise hyperemia. Therefore, our primary hypothesis was that in humans during a continuous three‐hour intra‐arterial infusion of a high dose of ADO (23 μgrams ·100 mL forearm vol −1 ·min −1 ) skeletal muscle blood flow would wane over time. Our secondary hypothesis was exercise hyperemic responses during ADO administration would be similar compared to control. Nine healthy young participants (29 ± 2 years) performed rhythmic handgrip (HG) exercise bouts (20% of maximum) during saline and 180 minutes of continuous ADO infusion. Forearm vascular conductance (FVC; ml min −1 (100 mmHg) −1 ) was calculated from measured forearm blood flow and mean blood pressure. FVC during minute 5 versus minute 175 during ADO infusion were 24.22 ± 3.2 versus 23.24 ± 1.95 ml min −1 (100 mmHg) −1 respectively (p>0.05). FVC during exercise increased significantly from baseline (4.78 ± 0.63 to 31.13 ± 4.17 ml min −1 *100 mmHg −1 ) during saline infusion and from (24.22 ± 3.23 to 56.33 ± 4.82 ml min −1 *100 mmHg −1 ) at five minutes and 21.99 ± 1.25 to 59.16 ± 5.21 seen at 150 minutes during ADO infusion. There was no significant difference in the delta FVC (exercise FVC – baseline FVC) between saline exercise bout versus ADO exercise bout (26.34 ± 3.75 versus 32.11 ± 5.55 ml min −1 *100 mmHg −1 ). Our findings show that contrary to the animal model, ADO vasodilatation does not wane in humans. Also, the vasodilator responses to exercise are similar during saline and prolonged ADO infusion compared to control condition. These findings suggest a minimal role of ADO in exercise hyperemia. Support or Funding Information Supported by National Institutes of Health Research Grants HL‐119337 (to M.J. Joyner and F.A. Dinenno) and by CTSA UL1 TR000135. The Caywood Professorship (M.J. Joyner) via the Mayo Foundation also supported this research.

During muscle contractions, an increase in interstitial potassium (K + ) and subsequent activation of inwardly‐rectifying K + (K IR ) channels is thought to be a functional vasodilatory signal contributing to exercise hyperemia. Indeed, our laboratory has recently demonstrated that inhibition of K IR channels blunts exercise hyperemia ~30% in humans. In theory, the vasodilatory response to changes in interstitial K + during muscle contraction could be minimized by artificially elevating the concentration of K + perfusing the vasculature prior to the onset of muscle contractions, thus limiting further recruitment of this pathway and blunting the dilatory response to exercise. We tested the hypothesis that infusion of potassium chloride (KCl) to elevate extracellular potassium concentrations and induce hyperpolarization prior to the onset of muscle contractions would reduce the contribution of this mechanism to the dilatory response to exercise. In young adults (n = 10), we measured forearm blood flow (FBF; Doppler ultrasound) and calculated changes in vascular conductance (FVC; FBF/intra‐arterial blood pressure) during 4 minutes of rhythmic handgrip exercise at 10% of maximal voluntary contraction during 1) control conditions (CTRL), 2) local, intra‐arterial infusion of KCl to elevate the concentration of K + prior to the initiation of exercise, and 3) intra‐arterial infusion of sodium nitroprusside (SNP) as a high flow control to match the K + ‐induced vasodilation observed at rest. As intended, infusion of KCl or SNP elevated resting FVC similarly prior to the onset of exercise (CTRL: 40 ± 7 vs. KCl: 84 ± 13 and SNP: 85 ± 14 ml·min −1 ·100 mmHg −1 ; both P < 0.01 vs. CTRL). Infusion of KCl diminished the subsequent vasodilatory response to handgrip exercise by ~31% when compared to control conditions (ΔFVC: KCl: 92 ± 14 vs. CTRL: 133 ± 16 ml·min −1 ·100 mmHg −1 ; P < 0.01), whereas elevation of resting FVC with SNP prior to contractions did not change the vasodilatory response to exercise (ΔFVC, SNP: 117 ± 15 vs . CTRL: 133 ± 16 ml·min −1 ·100 mmHg −1 ; P = NS). These findings indicate that K + ‐mediated hyperpolarization prior to the onset of muscle contractions blunts the normal vasodilatory response to exercise, which supports a role for elevated K + signaling in the hyperemic response to exercise in humans. Support or Funding Information HL119337, HL095573

Exercise‐induced rapid onset vasodilation produces a temporary rise in skeletal muscle blood flow in humans. The underlying mechanisms for this rapid hyperemia and regulation of muscle blood flow with single contraction are still not well understood. We tested the hypothesis that during a prolonged high dose intrarterial infusion of adenosine triphosphate (ATP; 20 μgrams ·100 mL forearm vol −1 ·min −1 ), the rapid onset vasodilatory response post‐single exercise contraction would be reduced compared to that of adenosine (ADO; 23 μgrams ·100 mL forearm vol −1 ·min −1 ). 14 healthy young participants (30 ± 1 years) performed a forearm single contraction exercise (20% of maximum) at minute 75 of either ATP (N=8) or ADO (N=6) continuous infusion. Artery diameter and blood velocity were measured using Doppler ultrasound. Forearm vascular conductance (FVC; ml·min −1 ·100 mmHg −1 ), time to peak, and area under curve were calculated. The mean resting FVC was higher during ATP (307.15 ± 50.56) and ADO (353.23 ± 43.98) infusion compared to saline infusion (45.17 ± 5.61 and 47.14 ± 5.57) (p<0.05). Compared to ADO, ATP infusion had a lower delta FVC from post‐ to pre‐contraction (174.06 ± 30.75 vs. 276.53 ± 24.90, p<0.05). Area under the curve and time of peak FVC did not significantly differ between drugs. These results suggest a prolonged ATP infusion may have blunted the rapid vasodilatory response more‐so than ADO. Thus, our findings indicate that ATP may be necessary but not obligatory in the rapid onset vasodilation response post‐single muscle contraction in humans. Support or Funding Information Supported by National Institutes of Health Research Grants HL‐119337 (to M.J. Joyner and F.A. Dinenno) and by CTSA UL1 TR000135. The Caywood Professorship (M.J. Joyner) via the Mayo Foundation also supported this research.

BACKGROUND Leptin is an adipokine with a central role in energy homeostasis and appetite regulation. Increases in leptin suppress appetite, while decreases in leptin increase appetite and may consequently cause weight gain. Patients with established obstructive sleep apnea (OSA) have been shown to have elevated leptin. Also, newly diagnosed OSA patients have a history of significant recent weight gain. Repetitive nocturnal intermittent hypoxia has been implicated in OSA pathophysiology and in increases in leptin. However, the factors contributing to weight gain in OSA are not completely understood. In contrast to the history of weight gain in OSA subjects, animal models of intermittent hypoxia (IH) are associated with weight loss along with increases in leptin. Furthermore, acute altitude dependent hypoxia in humans has also been associated with loss of appetite and increases in leptin. We postulated that the duration of IH may be important in defining the nature of effects on leptin. In this study we sought to examine the interactions between IH and leptin secretion, comparing acute versus chronic exposure to IH. We tested the hypothesis that although acute IH increases leptin secretion and likely suppresses appetite, repeated chronic exposure to IH will decrease leptin secretion, and thereby may increase appetite and predispose individuals such as patients with OSA to weight gain. METHODS We used an in‐vitro approach to examine the acute effects of IH (24 cycles of 30 min 0.1% O 2 followed by 30 min exposure to 21% O 2 ) in differentiated human white preadipocytes. Cells grown in normoxic conditions and continuous hypoxia were controls for these experiments. For chronic treatment, cells were exposed to 9 cycles of IH each day for 7 days. Leptin secretion was measured in 48 h conditioned media by ELISA. RESULTS Acute 24 hour exposure to IH or sustained hypoxia increased leptin secretion (p = 0.013). However, chronic 7 day exposure of differentiated human white preadipocytes caused decreases in leptin secretion (p = 0.02). CONCLUSION Acute intermittent hypoxia elicits increases in leptin secretion. Conversely, chronic IH decreases leptin secretion, suggesting a mechanism for development of obesity in the early stages of OSA. We speculate that decreased leptin, as a consequence of chronic IH, would result in increased appetite and hence weight gain. As obesity progresses, leptin levels likely rise secondary to the increase in body fat. Elevated leptin levels in patients with longstanding OSA may be indicative of increased fat mass and not a consequence of IH‐mediated effects on adipocytes. Support or Funding Information Christiane Becari was supported by a grant from CNPq Brazil (# 203802/2014‐4); Prachi Singh was supported by AHA scientist development grant (11SDG7260046).

Background Obstructive sleep apnea (OSA) patients have elevated levels of plasma leptin and are exposed to repeated nocturnal intermittent hypoxia (IH). Leptin induces increased adiponectin expression, suggesting that adiponectin should be increased in OSA. However, levels of plasma adiponectin, an anti‐atherogenic and insulin‐sensitizing adipokine, are inconsistent in OSA, and adiponectin levels are often low. Both increases in leptin and decreases in adiponectin may contribute to increased cardiometabolic risk in OSA. In this study we sought to examine the interactions between IH, leptin, and adiponectin. Using an in‐vitro approach in differentiated human white preadipocytes (dHWP), we tested the hypothesis that adiponectin expression in response to acute IH‐mediated increases in leptin may be modulated in the presence of hypoxia. Methods dHWP were exposed to 24 cycles of IH (30 min 0.1% O2 followed by 30 min exposure to 21% O2), continuous hypoxia (0.1% O2) or continuous normoxia (21% O2, control). Secretion of leptin and adiponectin in the 24h conditioned media was determined by ELISA and intracellular changes in adiponectin were determined by Western blot analysis. Results We show that acute 24 hour exposure to IH or sustained hypoxia increased leptin secretion (p = 0.013) but did not alter adiponectin secretion (p = 0.48). In fact, the intracellular expression of adiponectin actually decreased during the acute exposure to both IH and to continuous hypoxia (p = 0.02). Also, while leptin increased intracellular adiponectin expression in normoxic conditions (p = 0.03), it was unable to do so in the presence of IH and hypoxia (p = 0.36). Conclusions Acute IH and continuous hypoxia mediate increases in leptin, but not in adiponectin secretion. Furthermore, although leptin increases adiponectin in normoxic conditions, IH and continuous hypoxia cause decreases in intra‐cellular adiponectin expression and also attenuate leptin‐mediated increases in adiponectin expression. These hypoxia‐effects on adiponectin expression may help explain the low adiponectin levels in OSA, which may contribute to increased cardiovascular and diabetes risk. Support or Funding Information Christiane Becari was supported by a grant from CNPq Brazil (# 203802/2014‐4); Michaela Pfeifer was supported by the APS STRIDE summer undergraduate research fellowship (NHLBI 1R25HL115473‐01); Prachi Singh was supported by AHA scientist development grant 11SDG7260046.

Background Patients with sleep apnea experience repeated, intermittent hypoxic exposures during sleep and are at increased risk of developing hypertension. The mechanisms that contribute to high blood pressure in these patients are not completely understood. In this context, we sought to examine the effect of intermittent hypoxic exposures on baroreflex sensitivity and blood pressure. We hypothesized 3‐hours of intermittent hypoxia would result in a decrease in baroreflex sensitivity in healthy adults. We further hypothesized any decrease baroreflex sensitivity would be related to an increase in blood pressure following intermittent hypoxia. Methods Six (5M/1F) young healthy, non‐obese, subjects without sleep apnea completed 3‐hours of intermittent hypoxia. Intermittent hypoxia consisted of 25‐second exposures to hypoxic gas (0.05 F i O 2 ) separated by 2 minutes of room air breathing, resulting in 25‐hypoxic events per hour in which oxygen saturation (S p O 2 ) was significantly reduced (Normoxia: 97%, Hypoxia: 89%, p<0.01). Beat‐to‐beat heart rate (electrocardiogram) and blood pressure (finger photoplethysmography) were measured continuously. Spontaneous cardiac baroreflex sensitivity was determined using the sequence method (WinCPRS) and data are reported from 10‐minutes of quiet, normoxic rest before and after the intermittent hypoxia protocol. Results We observed a significant reduction in spontaneous cardiac baroreflex sensitivity to falling pressures (“down‐down”) following intermittent hypoxia (Control=22±2, IH=15±2 ms/mmHg; p=0.02). No change in baroreflex sensitivity to rising pressures (“up‐up”) was observed (Control=25±2, IH=20±6 ms/mmHg; p=0.41). A relationship between baroreflex sensitivity and mean arterial blood pressure was observed, such that a reduction in baroreflex sensitivity tended to be related to an increase in blood pressure (“down‐down”: R=−0.45, p=0.14; “up‐up”: R=−0.54, p=0.07). Conclusions Intermittent hypoxia results in a decrease in baroreflex sensitivity, which is related to an increase in blood pressure. These data may have important implications for mechanisms contributing to the risk of hypertension in individuals with sleep apnea. Support or Funding Information DK90541 (MJJ), HL120570 (JKL)

Previous studies have indicated that acute slow‐paced breathing (6 breaths/min) significantly decreases arterial blood pressure (BP) in both normotensive and hypertensive individuals. More recent work suggested that muscle sympathetic nerve activity (MSNA) was unchanged during acute slow‐paced breathing, however, the pattern of MSNA firing was altered in that more periods of MSNA quiescence were present. Herein, we examined the ability of MSNA to modulate BP on a beat‐to‐beat basis. We hypothesized that the magnitude of the increase in BP following a burst of MSNA is attenuated during acute slow‐paced breathing contributing to the reduction in BP previously reported. BP (brachial catheter) and MSNA (peroneal microneurography) were continuously measured in 10 young (27 ± 1 years), healthy men during 10 minutes of normal spontaneous breathing (baseline) or paced breathing at six breaths per minute. Signal averaging was used to characterize changes in mean BP for 15 cardiac cycles following heartbeats associated with and without MSNA bursts. MSNA bursts were segregated as either single bursts (directly bordered by ≥1 heartbeat lacking MSNA) or multiple bursts (any burst positioned directly adjacent to another MSNA burst). Respiratory rate was 12 ± 1.1 breaths/min during baseline and 6 ± 0.7 breaths/min during slow‐paced breathing. Compared to baseline, mean BP was decreased during slow‐paced breathing (88±2 vs. 85±2 mmHg, P<0.05), whereas MSNA burst frequency was unchanged (13±2.2 vs. 13±2.4 bursts/min, P>0.05). At baseline, mean BP significantly increased following MSNA bursts (+3.2± 0.2 mmHg, P<0.05) and this increase was augmented during slow‐paced breathing (+5.0± 0.5 mmHg, P<0.05). There was a greater increase in mean BP following single as well as multiple MSNA bursts during slow‐paced breathing. In contrast, cardiac cycles without MSNA bursts (non‐bursts) were associated with a significant decrease in mean BP (−0.7± 0.1 mmHg, P<0.05), which was also augmented during slow‐paced breathing (−1.3± 0.3 mmHg, P<0.05). In summary, the pressor and depressor responses coupled to spontaneous oscillations in MSNA are augmented during slow paced breathing with a net effect of reduced BP. Thus, changes in MSNA firing pattern, without changes in burst frequency, can modulate resting BP. Support or Funding Information Funding: NIH HL‐083947 (MJJ)

The vasoactive properties of exogenous adenosine triphosphate (ATP) are unique because it has the dual ability to induce robust vasodilation as well as blunt sympathetic vasoconstriction, similar to what is observed in contracting human skeletal muscle. ATP‐mediated vasodilation and exercise hyperemia are dependent, in part, on activation of inwardly rectifying potassium (K IR ) channels and subsequent smooth muscle cell hyperpolarization, with a modest reliance on the production of nitric oxide (NO) and prostaglandins (PGs). However, recent studies from our laboratory indicate that these pathways are not obligatory for contracting skeletal muscle to blunt α 1 ‐adrenergic vasoconstriction. Given the specific role of K IR channels in ATP‐mediated vasodilation, the primary purpose of this study was to determine whether the ability of ATP to blunt α 1 ‐adrenergic vasoconstriction is dependent upon K IR channel activity in humans. A secondary purpose was to determine whether the ability of ATP to blunt α 1 ‐adrenergic vasoconstriction is dependent upon activation of K IR channels and Na + /K + ‐ATPase, as well as the synthesis of NO and PGs. In a total of 13 young subjects we measured forearm blood flow (Doppler ultrasound) and calculated changes in vascular conductance (FVC) to local intra‐arterial infusion of phenylephrine (PE; α 1 ‐agonist) during infusion of ATP or a control vasodilator(s) (Protocol 1 (n=7): adenosine (ADO), and sodium nitroprusside (SNP); Protocol 2 (n=6): ADO). Vasomotor responses were assessed before and after blockade of signaling pathways that contribute to ATP‐mediated vasodilation (Protocol 1: K IR channels (barium chloride); Protocol 2: combined blockade of K IR channels, Na + /K + ‐ATPase (ouabain), NO (L‐NMMA), PGs (Ketorolac)). Protocol 1: ATP significantly attenuated PE‐mediated vasoconstriction relative to both SNP and ADO (Δ FVC: ATP: −16 ± 2; ADO: −38 ± 6; SNP: −59 ± 6%; P<0.05 vs ADO and SNP). Blockade of K IR channels significantly attenuated ATP mediated vasodilation ~40% (P<0.05 vs. control), but had no effect on the ability of ATP to blunt PE‐ mediated vasoconstriction (Δ FVC: −6 ± 5%; P>0.05 vs. control). Protocol 2: Despite combined blockade of K IR , Na + /K + ‐ATPase, NO, and PGs resulting in profound attenuation of ATP‐mediated vasodilation (−~60%; P<0.01 vs. control), ATP was still able to blunt PE‐mediated vasoconstriction (Δ FVC: −4 ± 14%; P=0.50 vs. control). Together these studies highlight that while NO, PGs, and vascular smooth muscle hyperpolarization (via K IR channel activation) are contributors to the normal vasodilatory response to exercise and exogenous ATP, these pathways are not obligatory for exercise or ATP to blunt α 1 ‐adrenergic vasoconstriction in humans. Support or Funding Information Support: HL119337, HL095573

Activation of group III/IV skeletal muscle afferents (ergoreflex) and hypoxia (chemoreflex) are two potent stimuli of the sympathetic nervous system during exercise. Venous distention by graded sub‐systolic occlusion contributes to the ergoreflex during exercise. Although the ergoreflex causes greater increases in blood pressure (BP) compared with the chemoreflex, the interaction of the ergoreflex and hypoxia during lower extremity dynamic exercise is unknown. Objective To determine the effect of hypoxia and graded cuff occlusion on hemodynamic responses during low‐intensity cycling exercise. Methods 19 healthy adults (9 men, 25±5 yr) completed two sessions. Session 1 included a maximal cycle ergometry exercise test. Session 2 included two randomized bouts (hypoxia: 12.5% FiO 2 and normoxia: 21% FiO 2 ) of constant‐load cycling at 30% peak workload for 3 min (control, CTL) followed by randomized inflation of bilateral thigh pressure tourniquets to 20, 40, 60, 80, 100 mmHg for 2 min with 2 min of deflation between occlusions. The bouts were separated by 30 minutes of rest. Blood pressure, heart rate and cardiac output (CO) were measured continuously. Results During hypoxia, the average SPO 2 was 80%. Venous occlusion and hypoxia had differential effects on mean arterial pressure (MAP, occlusion × hypoxia, p=0.02). Significant increases of MAP from CTL occurred at 60 (Δ9 ± 6mmHg), 80 (Δ15 ± 8mmHg), and 100mmHg (Δ17 ± 9mmHg) under normoxia (p<0.05). MAP only increased at 100mmHg (Δ14 ± 9mmHg) under hypoxia (p<0.05). Under hypoxia, MAP was much greater at CTL and 20mmHg as compared to normoxia (p<0.05). SBP exhibited responses similar to MAP under hypoxia (100mmHg increase, p<0.001, occlusion × hypoxia interaction, p=0.02). CO was greater with venous occlusion as mediated by increases in HR (p<0.001). In addition, CO increased under hypoxia (p=0.007) but had no interaction with venous occlusion. Conclusion During low pressure venous occlusion, hypoxia had a greater effect on BP, but as pressure increased (60–100mmHg), there were no significant differences in BP between normoxia and hypoxia. Therefore, sympathetically mediated increases of BP during exercise attributed to hypoxia may negate any responses mediated by the ergoreflex when elicited by venous occlusion at low pressures.

Excess mitochondrial reactive oxygen species can drive endothelial dysfunction and vascular calcification in atherosclerosis. Recently, SIRT3 has been linked to the activity of the ROS‐degrading enzyme MnSOD through regulation of protein acetylation levels. It remains unclear, however, whether SIRT3 protects against endothelial dysfunction and osteogenic signaling in hypercholesterolemia. We hypothesized, that reduction of SIRT3 will impair endothelial function, decrease mitochondrial antioxidant capacity, and increase osteogenic signaling in hypercholesterolemic mice. We used Ldlr −/− /ApoB 100/100 (LA) mice that were either wild‐type (LA‐SIRT3 +/+ ) or null for SIRT3 (LA‐SIRT3 −/− ). After 12 months of western diet feeding, we assessed endothelial function in aorta using isolated organ chamber baths and measured changes in gene expression using qRT‐PCR. As expected, SIRT3 gene expression was significantly reduced in LA‐SIRT3 +/− mice compared to LA‐SIRT3 +/+ mice, and undetectable in LA‐SIRT3 −/− . Surprisingly, endothelial‐dependent relaxation to acetylcholine (MR Ach ) was unchanged between LA‐SIRT3 +/+ and LA‐SIRT3 −/− (37.0±4.4%, 38.2±4.4%). Endothelial‐independent relaxation (sodium nitroprusside) was also unchanged. MnSOD expression was reduced in LA‐SIRT3 −/− compared to LA‐SIRT3 +/+ (0.80 ± 0.02 vs 1.0±0.08, p < 0.05). In contrast, the late‐stage osteoblast marker, osterix, tended to increase in LA‐SIRT3 −/− (1.3 fold) but failed to reach statistical significance. Collectively, these results suggest that losses of SIRT3 do not impact endothelial function in advanced atherosclerosis, but may contribute to augmentation of osteogenic signaling and accelerated progression of vascular calcification. Support or Funding Information NHLBI