Moderate whole body heating attenuates the exercise pressor reflex responses in older humans

Abstract

Prior reports show that whole body heat stress attenuates the pressor response to exercise in young healthy subjects. The effects of moderate whole body heating (WBH; e.g., increase in internal temperature T_core of ∼0.4°C–0.5°C) or limb heating on sympathetic and cardiovascular responses to exercise in older healthy humans remain unclear. We examined the muscle sympathetic nerve activity (MSNA), mean arterial blood pressure (MAP), and heart rate (HR) in 14 older (62 ± 2 yr) healthy subjects during fatiguing isometric handgrip exercise and postexercise circulatory occlusion (PECO). The protocol was performed under normothermic, moderate WBH, and local limb (i.e., forearm) heating conditions during three visits. During the mild WBH stage (increase in T_core of <0.3°C), HR increased, whereas BP and MSNA decreased from baseline. Under the moderate WBH condition (increase in T_core of ∼0.4°C), BP decreased, HR increased, and MSNA was unchanged from baseline. Compared with the normothermic trial, the absolute MAP during fatiguing exercise and PECO was lower during the WBH trial. Moreover, MSNA and MAP responses (i.e., changes) to fatiguing exercise were also less than those seen during the normothermic trial. Limb heating induced a similar increase in forearm muscle temperature to that seen in the WBH trial (∼0.7°C–1.5°C). Limb heating did not alter resting MAP, HR, or MSNA. The MSNA and hemodynamic responses to exercise in the limb heating trial were not different from those in the normothermic trial. These data suggest that moderate WBH attenuates MSNA and BP responses to exercise in older healthy humans.

INTRODUCTION

The effects of heat stress on autonomic and cardiovascular systems have been well studied (1, 2). For example, it is known that whole body heat stress raises skin blood flow (SkBF) (1). To maintain enough blood pressure (BP) when peripheral vasodilation is induced by heating, heart rate (HR), and cardiac output increase (1). Moreover, many reports have shown that whole body heat stress evokes significant increases in muscle sympathetic nerve activity (MSNA) in young healthy (3–7) and older healthy (8–10) individuals. Thus, whole body heat stress is a potent activator of the sympathetic nervous system (2).

It should be noted that those prior studies on effects of heat exposure on cardiovascular system were performed to examine the effects of a “heat stress” that raised internal temperature (T_core) by ∼0.5°C–1.3°C. The rationale for performing these studies was based on the fact that heat stress as seen with heat waves or when exercise is performed on hot days can induce adverse cardiovascular events (e.g., orthostatic syncope). However, humans are much more frequently exposed to less intense periods of heat exposure (e.g., elder individuals walking on a warm day). It is thus surprising that much less is known about the effects of less intense whole body heating (moderate WBH, e.g., an increase in the T_core of <0.5°C). Prior work from this laboratory showed that mild WBH (i.e., increase in T_core < 0.3°C) led to a reduction in MSNA in patients with chronic heart failure (8). Of note, a decrease in MSNA was observed in some but not all of the older healthy control subjects in that study. In that prior study, the period of mild WBH was not controlled.

During exercise, MSNA rises (11–13), which in turn leads to an increase in BP (i.e., exercise pressor reflex) (14). The increased input from mechanosensitive and metabosensitive afferents (i.e., group III and IV afferent fibers) (15–17) and central command (18) play key roles in evoking the increase in BP with exercise. Postexercise circulatory occlusion (PECO) is a well-established approach to activate metabosensitive muscle afferents (19). In animal studies, static muscle stretch is used to stimulate mechanosensitive afferents and evoke increases in BP (20). However, in humans, static passive stretch of forearm muscles under freely perfused conditions does not significantly alter MSNA and BP (21). Instead, when the mechanosensitive afferents are sensitized by metabolites under PECO conditions, static passive stretch of forearm muscles induces significant increases in MSNA and BP (21). Thus, stretch under PECO conditions (PECO + stretch) has been used to examine the effects of stimulation of muscle mechanosensitive afferents under varied conditions in prior human studies (22–24).

It is well known that resting MSNA increases with aging (25). The increase in HR during exercise becomes smaller with aging under normothermic condition (26). Physiological responses to heat stress at rest and during exercise may differ with age (27, 28). For instance, the increase in SkBF induced by passive heating (29) or exercise in hot environments (27) is attenuated with aging. Moreover, heat stress-induced increases in cardiac output are also smaller in older subjects. This effect is due primarily to reductions in stroke volume seen with aging, and the reduced cardiac output is due primarily to an attenuated rise in stroke volume (30). Prior echocardiographic studies suggest that cardiac systolic function “improves” during heat stress in young and middle-aged healthy subjects (31). It is possible that age-associated attenuation in cardiac output increase during heat exposure may be related to age-induced changes in myocardial function. The precise effect of heat exposure on the cardiac function in older individuals is unknown.

Moderate heat exposure rather than heat stress is likely to be more representative of routine heat exposure that is seen in older individuals. However, the sympathetic and cardiovascular (e.g., myocardium function) responses to exercise seen in response to moderate heat exposure in older individuals have not been examined thoroughly. Accordingly, the purpose of the present study was to determine sympathetic and cardiovascular responses to exercise under moderate heat exposure in older adults. A prior report from our laboratory (22) showed that in young healthy subjects, whole body heat stress (ΔT_core of ∼0.6°C via a heating suit) attenuated the pressor response seen during PECO. Moreover, in that prior study, the MSNA response to stimulation of muscle mechanoreceptors during PECO (i.e., PECO + stretch) was attenuated during heat stress conditions. It is unclear if this effect can be observed in older individuals under moderate WBH conditions. A prior study on cats showed that the elevated muscle temperature (T_muscle) attenuated P2X receptor-mediated BP responses to muscle afferent activation induced by muscle stretch (32). However, there are no data regarding the T_muscle in exercising limbs in humans under moderate WBH conditions. Importantly, it is unknown what effect passive moderate heating would have on muscle afferent sensitivity and in turn sympathetic nerve responses to exercise. Thus, it is necessary to examine the effects of both moderate WBH and limb heating on sympathetic nerve responses to exercise to determine whether muscle afferent sensitivity is altered at this heating level. Since the MSNA response to muscle stretch during PECO in young healthy subjects was attenuated in WBH condition (22), we speculate that the sensitivity of muscle mechanoreceptors may be decreased under moderate WBH conditions.

In this study, we examined the MSNA and hemodynamic responses to handgrip exercise during moderate WBH and local limb (i.e., forearm) heating (LH) conditions in older healthy individuals. We hypothesized that the moderate WBH would decrease the sensitivity of muscle mechanoreceptors and in turn attenuate the MSNA and pressor responses seen with exercise.

METHODS

Subjects

Fourteen healthy subjects (9 male, 5 female) participated in this study. The average age was 62 ± 7 yr with a height of 173 ± 9 cm and a weight of 81 ± 17 kg (body mass index = 26.7 ± 3.7). The baseline supine BPs in all subjects were <140/90 mmHg. None was taking medications, and no subject had cardiovascular disease. The experimental protocol was approved by the Institutional Review Board of Penn State College of Medicine and the Milton S. Hershey Medical Center and conformed with the Declaration of Helsinki. Each subject had the purposes and risks of the protocol explained to them before written informed consent was obtained.

Measurements

As described in our previous reports (22, 33), T_core was measured from an ingestible telemetric temperature pill (HQ Inc. Technologies, Palmetto, FL) that was swallowed by volunteers ∼1.5–2 h before data collection. Mean skin temperature (T_sk) was measured via the weighted average of six thermocouples attached to the skin on the chest (22%), abdomen (14%), upper back (19%), lower back (19%), thigh (14%), and calf (11%) (9, 34). To indicate the increase in both skin and internal temperatures, mean body temperature (T_body) was also calculated [0.9 × T_core + 0.1 × T_sk (35)] and reported. SkBF was indexed from dorsal forearm skin (nonexercising arm) using the mean values of two integrating flow probes of laser-Doppler flowmetry (MoorLab, Moor Instruments Ltd., Devon, UK). Cutaneous vascular conductance (CVC) was calculated from the ratio of the SkBF to mean arterial BP (MAP). The final SkBF and CVC were expressed as percentage of the normothermic baseline. Forearm sweat rate was measured via capacitance hygrometry (Vaisala, Woburn, MA) using the ventilated capsule method (surface area = 2.0 cm²) adjacent to the laser-Doppler probe. The reported sweat rates are changes from the normothermic baseline reading. The areas from which SkBF and sweat rate were measured were not covered by the suit, and the local temperature of these areas was not controlled. In the WBH trial and the LH trial, forearm T_muscle was measured using a 26-gauge intramuscular needle microprobe (MT-26/4HT, Type T thermocouple, Physitemp Instruments Inc., Clifton, NJ). The tip of probe was placed 2–3 cm below the skin into the flexor muscles of the forearm (36). To limit the possibility that heating the probe at the surface of the skin altered temperature readings in the muscle, the top of the probe was insulated from direct contact with the water-perfused suit/sleeve. The main purpose of measuring T_muscle in this study was to monitor muscle temperature changes during the WBH and LH procedures. Thus, this measurement was not performed in the normothermic control trial. In addition, a thermocouple sensor was placed on the skin adjacent to the T_muscle probe (or the similar site in the normothermic control trial) to measure the forearm skin temperature (T_arm).

As described in our previous reports, systolic pressure (SBP), diastolic pressure (DBP), and MAP during resting and heating conditions were measured with an automated sphygmomanometer from the brachial artery (SureSigns VS3, Philips, Philip Medical System). Beat-by-beat BP was recorded from a finger of the nonexercising arm (Finometer, Finapres Medical Systems, Amsterdam, The Netherlands) with resting values verified with the upper arm cuff pressure. HR was monitored from the electrocardiogram (Cardicap/5, Datex-Ohmeda, GE Healthcare, NJ). Respiratory frequency was monitored using piezoelectric pneumography. Multifiber recordings of MSNA were obtained with a tungsten microelectrode inserted in the peroneal nerve. A reference electrode was placed subcutaneously 2–3 cm from the recording electrode. The signal was amplified, filtered with a bandwidth of 500–5,000 Hz, and integrated with a time constant of 0.1 s (Iowa Bioengineering, Iowa City, IA). The recording electrode was adjusted until a site was found in which muscle sympathetic bursts were clearly identified using previously established criteria (37). The nerve signal was routed to a computer screen and a loudspeaker for monitoring throughout the study.

Cardiac function was assessed with echocardiography and tissue Doppler. As described in a prior report (38), transthoracic echocardiography was performed using a digital ultrasound system (model iE33; Philips Ultrasound, Bothell, WA) and a 5–1 MHz probe. Left atrial diameter was measured in the parasternal long-axis view, and then, adequate M-mode imaging was obtained from the parasternal view between the mitral valve and the papillary muscle. The ultrasound beam was positioned perpendicular to the intraventricular septum and the left ventricle posterior wall, allowing a clear view of the left ventricle diameter during diastole and systole. Left ventricular internal diameter was measured at the furthest endocardium endpoints (39). Left ventricular internal end-diastolic diameter was measured, and left ventricular end-diastolic volume, stroke volume, and ejection fraction were calculated by the echocardiography device (40). All measurements were averaged over three to five cardiac cycles. Tissue-Doppler imaging analysis was performed on echocardiographic data collected in the apical four-chamber view, while the pulsed Doppler sample volume was placed at the septal and lateral mitral annulus. Three major components of regional myocardial velocities were recorded: 1) the positive systolic velocity when the mitral ring moved toward the cardiac apex (S_m), which is a sensitive marker of regional myocardial systolic function and lower loading dependence (41, 42); 2) the first rapid negative diastolic velocity (E_m) when the mitral annulus moves toward the base away from the apex; and 3) the second negative diastolic velocity (A_m). A_m and the ratio of early diastolic mitral flow (E) over E_m (i.e., E/E_m) were used as an index of diastolic function (43). All tissue-Doppler data were averaged from two different sites (septal and lateral walls) at the mitral annulus over three to five cardiac cycles.

Protocols

Three trials (i.e., WBH trial, normothermic control trial, and LH trial) were performed in three separate visits with an interval greater than 1 wk. The WBH trial and the normothermic control trial were performed in a random order. To match the elevation of the T_muscle during the LH trial with that seen in the WBH trial, the LH trial was performed last. The timelines of the three trials are shown in Fig. 1.

Figure 1.

Timeline of the experimental protocol. LH, limb heating trial; NT, nonheating trial; prior exer, baseline prior to exercise; T_arm, arm skin temperature; T_core, internal temperature; T_muscle, muscle temperature; T_sk, mean skin temperature; WBH, whole body heating trial.

Subjects refrained from caffeine, alcohol, and intense exercise 24 h before the study. All subjects had a light breakfast before the study was performed in the morning. All procedures were conducted with the subject in the supine position in a room with an ambient temperature of ∼23°C. During each visit, subjects were dressed in a tube-lined suit that permitted control of T_sk by changing the temperature of the water-perfused suit. The suit covered the entire body surface with the exception of both hands, the neck and head, and the feet. The skin around the forearm SkBF probes, sweat rate senor, T_muscle probe, T_arm probe on forearms, and around the nerve activity recording site in the lower leg was not covered with the suit. In each of the three visits, maximal force generated from a voluntary handgrip contraction (MVC) was determined upon repeated (i.e., ∼3) isometric contractions from the nondominant forearm using a handgrip dynamometer. To ensure that the strength of the passive stretch was as vigorous as possible without evoking pain, the stretch strength for each subject was tested before the study. As described in our previous reports (21, 44), a specifically designed brace with a joint at the wrist was used to support the subject’s forearm and hand. The hand was flexed in the dorsal direction (stretch) as the force was measured with a digital force gauge (IMADA, DPS-220, Northbrook, IL). During “stretch,” the position of the forearm and wrist remained fixed. The maximal force used to stretch the muscles without inducing pain was used for the stretch protocols. No subjects complained of pain with stretch. A prior study (21) showed that under freely perfused conditions, vigorous passive forearm muscle stretch that does not evoke pain does not alter measured MSNA and BP. Thus, the passive stretch was not performed under freely perfused conditions in the present study.

Whole body heating trial.

Baseline cardiac function indices were assessed before microneurography was performed. After full instrumentation, subjects were asked to remain still and refrain from speaking during an acclimation period (5–10 min) as hemodynamic variables and MSNA stabilized and as the tube-lined suit was perfused with 34°C water. Subsequently, all variables were continuously recorded for 6 min as automated sphygmomanometer BPs were obtained two or three times. Thereafter, T_sk was increased by perfusing the tube-lined suit with 48°C–45°C water. To observe the effects of mild WBH, the water temperature was adjusted to control the increase rate in T_core. After ∼30–40 min, WBH increased T_core by ∼0.3°C. Then heating was continued for an additional 5–15 min (total heating period was 50 min) before the cardiac function was assessed over ∼10 min. As heating continued, measurements of HR, BPs, and MSNA were collected during the 6-min “prior to exercise” period under the moderate WBH condition (Fig. 1). Each subject then performed static isometric handgrip at 30% MVC to fatigue followed by 4-min PECO as shown in Fig. 2. A visual force indicator was used so that subjects could maintain the force necessary to maintain the 30% MVC level during the contraction. During grip, subjects were asked to report their perceived level of effort using the Borg scale of 6–20 (45). The determination of fatigue was based on 1) the inability of the volunteer to maintain the desired force production and 2) the assessment of the volunteers that the work was “extremely hard.” When a Borg scale of ∼19 (extremely hard) was reported, a cuff on the upper arm was inflated to 250 mmHg before the subject stopped grip. Stretch was performed for 1.5 min during the occlusion. To decrease any possible order effect, the PECO + stretch (1.5 min) and the PECO only (1.5 min) periods were performed in a random order in the last 3 min of the 4-min occlusion. The first minute’s data from PECO were not included for any of the subjects. Subjects did not complain of any pain caused by the stretch maneuver performed during PECO.

Figure 2.

Representative tracings of handgrip force, heart rate (HR), muscle sympathetic nerve activity (MSNA), and arterial blood pressure (BP) in normothermic (top) and whole body heating (bottom) trials in one subject. bpm, beats/min.

Normothermic control trial.

In a separate visit, the T_sk was maintained at ∼34°C throughout the study by perfusing 34°C water via the tube-lined suit. Approximately 50 min after the baseline data collection, cardiac function was assessed (∼10 min). Then, the exercise paradigm was performed as that described in Whole body heating trial.

Local forearm heating trial.

In a separate visit, the subjects were dressed in the tube-lined suit while 34°C water was perfused throughout this trial. The exercising forearm (wrist to elbow) was covered with a separate tube-lined sleeve, which was connected with a separate heating unit. During the acclimation period and baseline data collection, 34°C water was perfused in the sleeve. Thereafter, 54°C–50°C water was perfused in the sleeve to raise the forearm T_muscle, while 34°C water was perfused in the suit. To match both elevation in T_muscle and the heating period in the LH trial with those in the WBH trial, higher water temperature was used in the heating sleeve. When the elevation in T_muscle was similar to that seen during the WBH trial (∼0.7°C–1.5°C), the exercise paradigm was performed as that described in Whole body heating trial. Cardiac function was not assessed in the LH trial.

Data analysis.

Data were sampled at 200 Hz via a data acquisition system (MacLab, ADInstruments, Castle Hill, Australia). MSNA bursts were evaluated by a computer program that identified bursts based on fixed criteria, including an appropriate latency following the R-wave of the electrocardiogram (44). Integrated MSNA was normalized by assigning a value of 100 to the mean amplitude of the largest 10% of the bursts during the 6-min normothermic baseline period (46). Normalization of the MSNA signal was performed to reduce variability between subjects attributed to factors including needle placement. Total MSNA was identified from burst area of the integrated neurogram that was evaluated on a beat-by-beat basis.

For the effects of heating procedure, the mean values of the 6-min normothermic baseline, the “mild WBH” (e.g., T_sk of >∼35.5°C and ΔT_core of <0.3°C) period or the time control period, and the 6-min “prior to exercise” period were calculated. For the effects of exercise, values were averaged during the 6-min period of “prior to exercise,” the last minute of exercise, PECO alone (1.5 min), and stretch (1.5 min) during PECO. The change (i.e., Δ) in measured responses between the “prior to exercise” and exercise periods was determined. Statistical analyses were performed using commercially available software (SigmaPlot 14.0, Systat Software Inc.). The effects of the heating (factor 1: WBH, normothermic control, and LH trials) on the responses during the stages of the exercise paradigm (factor 2: resting baseline, handgrip, PECO only, and PECO + stretch) were evaluated via a two-way repeated-measures ANOVA, followed by multiple-comparison Tukey’s post hoc analyses where appropriate. In addition, the data in two of our prior studies (8, 22) were retrospectively analyzed. The change (i.e., Δ) in measured responses between the “mild WBH” and normothermic baseline in the older subjects in the present report is compared with changes evoked by “mild WBH” in the older (62 ± 7 yr, subject number n = 12) and young (24 ± 2 yr, n = 14) subjects from our prior study (8). Results were compared using one-way ANOVA. The change (i.e., Δ) during the exercise trial in this project was compared with respective changes during exercise with the same protocol in young subjects (24 ± 2 yr, n = 14) in another prior study (22) via two-way ANOVA. All values are reported as means ± SD. P values < 0.05 were considered statistically significant.

RESULTS

Effects of “Mild WBH”

In the WBH trial, this “mild heating” period (21.1 ± 1.8 min) was under the condition of T_sk of >∼35.5°C and ΔT_core of <0.3°C. During this period, in some subjects, T_core decreased slightly at first and then increased, and in other subjects, T_core was not altered at first and then gradually increased. The averaged T_core over this period was not different from the baseline (Table 1). The SkBF and CVC in the WBH trial increased significantly from the baseline (Table 1). No sweating was observed. During this period, all subjects reported feeling “warm,” “not hot,” and “comfortable.” During this “mild WBH” period, absolute SBP, DBP (Table 1), MAP, and MSNA were significantly lower than normothermic baseline, whereas the HR was significantly higher than the baseline (Fig. 3). Moreover, compared with the values during the similar time control period in the normothermic trial, the BPs and MSNA values seen during the “mild WBH” period were lower, and HR was higher. During the similar time control period in the normothermic trial, T_sk, T_core, SkBF, CVC, SBP, DBP, MAP, and MSNA were not different from the baseline.

Table 1.

Hemodynamic and thermoregulatory variables during heating and the time control period

	Baseline	Mild WBH Time Control	Prior Exercise
SBP, mmHg
WBH	119 ± 8	112 ± 6*#	116 ± 6
NT	117 ± 7	119 ± 6	122 ± 11
LH	118 ± 10	117 ± 10	122 ± 11
DBP, mmHg
WBH	76 ± 7	71 ± 6*#	72 ± 8*#
NT	77 ± 5	79 ± 5	80 ± 8*
LH	76 ± 8	76 ± 8	78 ± 6†
MSNA BI, bursts/100 beats
WBH	56.5 ± 12.6	41.4 ± 13.5*#	51.7 ± 18.5‡
NT	54.3 ± 13.4	54.8 ± 14.5	55.7 ± 18.1
LH	53.1 ± 14.8	54.5 ± 14.0†	53.1 ± 15.6
SkBF, %
WBH	100	167 ± 71*#	235 ± 118*#‡
NT	100	107 ± 22	104 ± 42
LH	100	127 ± 60	139 ± 52†
CVC, %
WBH	100	180 ± 93*#	249 ± 138*#‡
NT	100	105 ± 23	102 ± 47
LH	100	128 ± 61	135 ± 53†
Tsk, °C
WBH	33.9 ± 0.6	36.7 ± 1.0*#	36.6 ± 0.7*#
NT	34.2 ± 0.5	34.2 ± 0.5	34.3 ± 0.5
LH	34.2 ± 0.5	34.3 ± 0.5†	34.3 ± 0.5†
Tcore, °C
WBH	36.8 ± 0.3	36.9 ± 0.3	37.1 ± 0.3*#‡
NT	36.8 ± 0.2	36.8 ± 0.2	36.8 ± 0.2
LH	36.9 ± 0.2	36.8 ± 0.2	36.8 ± 0.2†
Tbody, °C
WBH	36.5 ± 0.3	36.8 ± 0.3*#	37.1 ± 0.3*#‡
NT	36.5 ± 0.2	36.5 ± 0.2	36.5 ± 0.2
LH	36.6 ± 0.2	36.6 ± 0.2†	36.6 ± 0.2†
Tarm, °C
WBH	30.3 ± 1.1	31.5 ± 1.7*	31.5 ± 1.7*#
NT	30.9 ± 0.8	31.3 ± 1.5	30.4 ± 1.0
LH	31.4 ± 1.9	34.2 ± 1.8*#†	35.3 ± 1.3*#†
Tmuscle, °C
WBH	33.1 ± 1.4	33.1 ± 1.4	34.3 ± 1.6*‡
LH	33.5 ± 0.8	34.1 ± 0.8	34.8 ± 1.0

Two-way repeated-measures ANOVA was used for statistical analysis. *P < 0.05 vs. respective baseline; #P < 0.05 vs. normothermic control trial; †P < 0.05 LH vs. WBH; ‡P < 0.05 prior exercise vs. mild WBH. Subject number n = 14 for the WBH and normothermic control trials. n = 11 for the LH trial. CVC, cutaneous vascular conductance; DBP, diastolic pressure; HR, heart rate; LH, local forearm heating trial; MAP, mean arterial blood pressure; MSNA BI, muscle sympathetic nerve activity burst incidence; NT, normothermic time control trial; SBP, systolic pressure (SBP); SkBF, skin blood flow; T_arm, arm skin temperature; T_body, mean body temperature; T_core, internal temperature; T_muscle, muscle temperature; T_sk, mean skin temperature; WBH, whole body heating trial.

Figure 3.

Mean arterial pressure (MAP), heart rate (HR), and muscle sympathetic nerve activity (MSNA) during baseline, mild whole body heating or time control period, and prior exercise in the whole body heating (WBH), normothermic control, and local forearm heating (LH) trials. The small dots represent the individual data. P values for trial, time, and the interaction between trial and time with a two-way repeated-measures ANOVA: MAP: 0.001, 0.069, <0.001; HR: <0.001, <0.001, <0.001; MSNA burst rate: 0.637, 0.030, 0.007; MSNA total activity: 0.047, 0.053, 0.080. Post hoc: *P < 0.05 vs. respective baseline; #P < 0.05 vs. normothermic control trial; †P < 0.05 LH vs. WBH; ‡P < 0.05 prior exercise vs. mild WBH. subject number n = 14 for MAP and HR and n = 13 for MSNA in the WBH and normothermic control trials. n = 11 for MAP and HR and n = 9 for MSNA in the LH trial. NT, nonheating trial.

When the results of the present group of older subjects are compared with the data of the younger subjects in the prior study (8), the changes in MSNA burst rate (Δ−7.9 ± 6.0, Δ−10.1 ± 13.3, Δ2.7 ± 6.6 bursts/min, one-way ANOVA P = 0.002) evoked by mild WBH in the present study (Δ−7.9 ± 6.0 bursts/min) were significantly (post hoc P = 0.01) different from those in young healthy subjects (Δ2.7 ± 6.6 bursts/min). The changes in MSNA induced by mild WBH in the present study were not significantly different from the older subjects in the prior study (Δ−10.1 ± 13.3 bursts/min, P = 0.82). The changes in MAP (P = 0.90) and increase in HR (P = 0.20) were not different between the groups.

Effects of Moderate Whole Body Heating

In the WBH trial, with continuous heating, the T_core during the prior to exercise under moderate WBH condition increased by 0.39 ± 0.13°C from the normothermic baseline. During the prior to exercise under moderate WBH condition, DBP and MAP were lower and HR, SkBF, and CVC were higher than the normothermic baseline values in the WBH trial. Sweating was observed (0.103 ± 0.108 mg/cm²/min). MSNA and SBP were not different from the normothermic baseline values in the WBH trial. In the normothermic trial, the thermoregulatory variables did not change significantly with time (Table 1).

Stroke volume after WBH was not different from the normothermic baseline in the WBH trial (Table 2), whereas cardiac output was significantly greater than the normothermic baseline. TDI-E_m and E decreased after WBH, whereas TDI-S_m, TDI-A_m, and E/E_m were not different from normothermic baseline values (Table 2). These cardiac function indices did not change during the similar time control period in the normothermic trial.

Table 2.

Cardiac function during whole body heating and normothermic time control trials

	Baseline	Prior to Exercise
SV, mL
WBH	117 ± 14	123 ± 17#
NT	114 ± 16	116 ± 16
CO, L/min
WBH	7.3 ± 1.4	8.4 ± 1.9*#
NT	6.9 ± 1.3	7.0 ± 1.3
TPR, mmHg·min/L
WBH	12.9 ± 3.2	10.9 ± 3.0*#
NT	13.6 ± 2.8	14.0 ± 3.0
Sm, cm/s
WBH	9.54 ± 1.26	10.04 ± 1.79
NT	9.77 ± 1.51	9.59 ± 1.72
E, cm/s
WBH	74.0 ± 14	68.9 ± 13.8*
NT	71.8 ± 15.5	72.3 ± 12.5
Em, cm/s
WBH	10.35 ± 1.42	9.37 ± 1.52*#
NT	10.56 ± 1.81	10.53 ± 1.75
E/Em
WBH	7.20 ± 1.22	7.55 ± 2.08
NT	6.93 ± 1.62	6.96 ± 1.28
Am, cm/s
WBH	12.18 ± 2.29	12.59 ± 3.97
NT	12.11 ± 2.73	12.05 ± 2.96

Two-way repeated-measures ANOVA was used for statistical analysis. *P < 0.05 vs. respective baseline; #P < 0.05 vs. normothermic control trial. Subject number n = 14. A_m, second negative diastolic velocity when the mitral annulus moves toward the base away from the apex; CO, cardiac output; E, early diastolic mitral flow; E/E_m, ratio of early diastolic mitral flow over E_m; E_m, first rapid negative diastolic velocity when the mitral annulus moves toward the base away from the apex; NT, normothermic time control trial; S_m, positive systolic velocity when the mitral ring moved toward the cardiac apex; SV, stroke volume; TPR, total peripheral resistance; WBH, whole body heating trial.

MSNA and Hemodynamic Responses to the Fatiguing Handgrip

Recordings of handgrip force, HR, integrated MSNA, and BP during the PECO and stretch in the WBH and the normothermic trial in a representative subject are shown in Fig. 2. T_sk, T_core, and T_body did not change during the exercise trials, regardless of the thermal condition (Table 3). The measured handgrip force (7.12 ± 2.35, 7.13 ± 2.20, 7.14 ± 2.04 kg, P = 0.54) and the handgrip exercise duration (150 ± 65, 143 ± 56, 183 ± 77 s, P = 0.22) were not different in the WBH, normothermic control, and LH trials. There was no difference in the end-exercise Borg scale values during the three trials (19.5 ± 0.3, 19.4 ± 0.2, 19.4 ± 0.2, P = 0.51).

Table 3.

Temperatures and MSNA burst incidence during handgrip exercise in the three trials

	Prior Exercise	HG	PECO	Stretch
MSNA BI, bursts/100 beats
WBH	51.7 ± 18.5	57.2 ± 18.1	62.7 ± 24.8	63.5 ± 25.7*
NT	55.7 ± 18.1	67.2 ± 21.9*	67.4 ± 18.1*	74.8 ± 21.5*
LH	53.1 ± 15.6	64.3 ± 18.8	66.2 ± 14.6*	68.3 ± 16.3*
Tsk, °C
WBH	36.6 ± 0.7#	36.6 ± 0.7#	36.6 ± 0.7#	36.6 ± 0.7#
NT	34.3 ± 0.5	34.3 ± 0.5	34.3 ± 0.6	34.3 ± 0.6
LH	34.3 ± 0.5†	34.3 ± 0.5†	34.4 ± 0.5†	34.4 ± 0.5†
Tcore, °C
WBH	37.1 ± 0.3#	37.1 ± 0.3#	37.2 ± 0.3#	37.1 ± 0.3#
NT	36.8 ± 0.2	36.8 ± 0.2	36.8 ± 0.2	36.8 ± 0.2
LH	36.8 ± 0.2	36.8 ± 0.2	36.8 ± 0.2	36.8 ± 0.1
Tbody, °C
WBH	37.1 ± 0.3#	37.1 ± 0.3#	37.1 ± 0.3#	37.1 ± 0.3#
NT	36.5 ± 0.2	36.6 ± 0.2	36.55 ± 0.2	36.55 ± 0.2
LH	36.6 ± 0.2†	36.6 ± 0.2†	36.6 ± 0.2†	36.6 ± 0.1†
Tarm, °C
WBH	31.5 ± 1.7#	31.9 ± 1.7#	31.8 ± 1.6#	31.9 ± 1.5#
NT	30.4 ± 1.0	30.5 ± 0.8	30.6 ± 0.8	30.6 ± 0.8
LH	35.3 ± 1.3#†	35.7 ± 1.1*#†	35.9 ± 1.1*#†	36.1 ± 1.1*#†
Tmuscle, °C
WBH	34.3 ± 1.6	34.7 ± 1.2	34.8 ± 1.1*	34.6 ± 1.2
LH	34.8 ± 1.0	35.7 ± 0.6*†	35.9 ± 0.6*†	35.8 ± 0.6*†

Two-way repeated-measures ANOVA was used for statistical analysis. *P < 0.05 vs. prior exercise; #P < 0.05 vs. normothermic control trial; †P < 0.05 LH vs. WBH. Subject number n = 14 for the WBH and normothermic control trials. n = 11 for the LH trial. HG, last minute of the handgrip exercise; LH, local forearm heating trial; MSNA BI, muscle sympathetic nerve activity burst incidence; NT, normothermic time control trial; PECO, postexercise circulatory occlusion; stretch, passive stretch during the occlusion; T_arm, arm skin temperature; T_body, mean body temperature; T_core, internal temperature; T_muscle, muscle temperature; T_sk, mean skin temperature; WBH, whole body heating trial.

Isometric handgrip exercise, PECO, and the passive stretch during PECO evoked significant increases in MSNA, HR, and MAP (Fig. 4) from the respective before exercise baseline in both the WBH trial and the normothermic trial. The absolute HR during these stages of the WBH trial was greater than those seen in the normothermic trial (Fig. 4). The absolute MAP during these stages of the WBH trial was lower than those seen in the normothermic trial. During the last minute of fatiguing handgrip, the increases from the before exercise baseline for MAP and MSNA (i.e., changes) in the WBH trial were lower than those in the normothermic trial (Fig. 5).

Figure 4.

Absolute MAP, HR, and MSNA during prior to exercise baseline (prior exer), the last minute of the handgrip exercise (HG), PECO, and passive stretch during the occlusion (stretch) in the WBH, normothermic, and LH trials. MAP was calculated from the blood pressure waveform obtained from the Finometer. The small dots represent the individual data. P values for trial, stage of exercise paradigm, and the interaction between trial and stage with two-way repeated-measures ANOVA: MAP: <0.001, <0.001, 0.013; HR: <0.001, <0.001, 0.078; MSNA burst rate: 0.203, <0.001, <0.001; MSNA total activity: 0.101, <0.001, <0.001. Post hoc: *P < 0.05 vs. prior exer. #P < 0.05 vs. normothermic control trial; †P < 0.05 LH vs. WBH; ‡P < 0.05 stretch vs. PECO. Subject number n = 14 for MAP and HR and n = 13 for MSNA in the WBH and normothermic control trials. n = 11 for MAP and HR and n = 9 for MSNA in the LH trial. HR, heart rate; LH, limb heating trial; MAP, mean arterial pressure; MSNA, muscle sympathetic nerve activity; NT, nonheating trial; PECO, postexercise circulatory occlusion; WBH, whole body heating.

Figure 5.

Increases in MAP, HR, and MSNA relative to prior exercise baseline (i.e., change) during the last minute of the handgrip exercise (HG), PECO, and passive stretch during the occlusion (stretch) in the WBH, normothermic, and LH trials. P values for trial, stage of exercise paradigm, and the interaction between trial and stage with two-way repeated-measures ANOVA. MAP: 0.001, <0.001, 0.953; HR: 0.194, <0.001, 0.060; MSNA burst rate: 0.191, 0.044, 0.191; MSNA total activity: 0.073, 0.004, 0.013. Post hoc: #P < 0.05 vs. normothermic control trial; ‡P < 0.05 stretch vs. PECO. ^P < 0.05 vs. HG. Subject number n = 14 for MAP and HR and n = 13 for MSNA in the WBH and normothermic control trials. n = 11 for MAP and HR and n = 9 for MSNA in the LH trial. HR, heart rate; LH, limb heating trial; NT, nonheating trial; MAP, mean arterial pressure; MSNA, muscle sympathetic nerve activity; PECO, postexercise circulatory occlusion; WBH, whole body heating.

The changes in MSNA (two-way ANOVA, P < 0.001) and HR (P < 0.001) seen during the exercise paradigm in the WBH trial in the present study were significantly different from changes seen in young subjects from a prior study (22). The changes seen in the last-minute handgrip in MSNA burst rate (Δ8.4 ± 6.5 vs. Δ17.9 ± 11.4 bursts/min, post hoc P = 0.03), MAP (Δ16.3 ± 7.5 vs. Δ23.4 ± 5.6 mmHg, P = 0.007), and HR (Δ8.7 ± 5.7 vs. Δ22.7 ± 9.8 mmHg, P < 0.001) in the present study were significantly lower than those in young healthy subjects under WBH conditions in the prior study (22).

MSNA and Hemodynamic Responses to Metaboreceptor Stimulation and Effects of Passive Stretch

During PECO and PECO + stretch period, the absolute MAP was lower, and absolute HR was higher in the WBH trial than those in the normothermic trial (Fig. 4). Moreover, the MAP response to PECO and PECO + stretch (i.e., change from the prior exercise baseline) in the WBH trial was lower than that in the normothermic trial (Fig. 5). In both trials, the MAP and MSNA total activity responses during the PECO + stretch period were greater than those during the PECO alone period (Fig. 5).

The changes by PECO in MSNA burst rate (Δ8.7 ± 6.7 vs. Δ17.3 ± 13.0 bursts/min, P = 0.046) and HR (Δ1.6 ± 2.7 vs. Δ10.0 ± 7.8 mmHg, P < 0.001) in the present study were significantly lower than those in young healthy subjects under the WBH condition in the prior study (22). The changes by PECO + stretch in MSNA (P = 0.10) and MAP (P = 0.05) in the present study were not significantly different from that in young healthy subjects under the WBH condition in the prior study (22).

Muscle Temperature and Effects of Local Forearm Heating

In the WBH trial, T_muscle (subject number n = 12) slightly increased during the handgrip exercise paradigm (Table 3). The LH trial was performed on 11 of the 12 subjects who had T_muscle data in the WBH trial. The probe for T_muscle measurement did not induce pain sensation during the heating procedure and the exercise paradigm. During before exercise baseline in the LH trial, the increase in T_muscle by limb heating was not different from that seen in the WBH trial (Δ1.4 ± 0.5°C vs. Δ1.2 ± 0.8°C P = 0.67). LH did not alter the T_sk, T_core, and T_body, which were lower than those in the WBH trial (Table 1). No sweating was observed in the LH trial. LH did not alter resting MSNA, HR, and MAP (Fig. 3). There was no difference in BPs, HR, and MSNA during before exercise baselines between the LH trial and the normothermic control trial (Fig. 3 and Table 1).

In the LH trial, T_muscle significantly increased during the handgrip exercise paradigm (Table 3). Handgrip exercise, PECO, and the passive stretch during PECO evoked significant increases in MSNA, HR, and MAP from the respective before exercise baseline values in the LH trial (Fig. 4). There was no difference in the responses of MAP, HR, and MSNA during the fatiguing HG, PECO, and PECO + stretch periods between the LH trial and the normothermic trial (Fig. 5). The passive stretch during PECO also induced a significant MAP increase from the PECO alone condition (Fig. 5).

DISCUSSION

The main findings of this study are as follows: 1) mild WBH (ΔT_core of <0.3°C) decreases resting MSNA and BPs in older healthy individuals; 2) moderate WBH (ΔT_core of ∼0.4°C) increases HR, cardiac output, and SkBF and decreases BPs, and MSNA under these conditions is not different from baseline values; 3) the MAP and MSNA responses to handgrip exercise are attenuated under the moderate WBH condition; and 4) elevating the T_muscle of the exercising muscles to the level seen during the moderate WBH condition does not attenuate MAP and MSNA responses to handgrip exercise.

Resting MSNA and BP during Mild and Moderate WBH

It is known that adverse cardiac events occur with a higher frequency during extremely hot summers (e.g., heat wave) (47, 48). However, other epidemiological reports suggest that hemodynamic values such as BP tend to be lower in the summer than in the winter (49, 50). Moreover, cardiovascular diseases, including myocardial infarction (51, 52) and heart failure (53, 54), are more frequent and have a higher mortality in the winter than in the summer. Our prior report (55) has shown that resting MSNA in healthy individuals is lower in the summer than in the winter. We speculated that the warmer environment seen during the summer might contribute to the observation in that study. However, other factors (e.g., diet, daily physical activity) could also contribute to the seasonal differences in resting MSNA. The present study shows that MSNA and BP in older healthy individuals decrease under a mild heat exposure condition.

In one of our prior studies (8), MSNA decreased during mild WBH in patients with heart failure but not in young subjects. When compared with the data from that study (8), the changes (i.e., decreases) in MSNA evoked by mild WBH in the older subjects in the present study were significantly greater than those seen in young healthy subjects. It should be noted that the conditions for “mild WBH” were similar in the two studies. These findings suggest that a period of (e.g., tens of minutes) mild WBH may only induce MSNA decrease in older subjects.

The MSNA and BP decreases seen during the period of mild WBH were not an effect of the resting time period since decreases in MSNA and BP were not observed during the normothermic control trial. We can only speculate factors that may contribute to the decrease in MSNA and BP seen during the mild WBH condition. Skin temperature at 34°C was supposed to be a thermoneutral condition in the present and prior studies (9). However, in the present studies, the subjects reported that they were “comfortable” during the mild WBH period. Subject relaxion might contribute to the decrease in sympathetic activities.

With continuing heating, T_core rose by ∼Δ0.4°C from baseline. The moderate WBH raised SkBF and CVC and decreased total peripheral resistance. The HR and cardiac output increased to maintain enough BP when total peripheral vasodilation was induced, whereas stroke volume was not significantly altered. These changes in these variables are consistent with hemodynamic changes seen during “heat stress” studies (30, 56, 57). However, the magnitude of the changes seen with moderate WBH was much smaller than that seen with “heat stress.” For example, SkBF increased by ∼Δ330% by whole body heat stress in older healthy subjects when T_core was elevated by ∼Δ0.9°C in a prior study (58), whereas SkBF only increased by ∼Δ135% by the moderate WBH in the present study. We speculate that the direct heating seen with the heating suits evokes cutaneous vasodilation via local mechanisms (e.g., production of nitric oxide and axon reflex) under the moderate WBH condition, although a prior study suggested that the cutaneous vasodilation by local heating was attenuated in older individuals (59). On the other hand, the possible role of a withdrawal of cutaneous vasoconstrictor activity could not be excluded, since a prior study (60) showed that the skin sympathetic nerve activity (SSNA) decreases in the early stage of whole body heat exposure. Under the moderate WBH condition, the MSNA was greater than that seen during the mild WHB period; however, it was not different from the “normothermic” baseline. This was different from the MSNA activation seen under the “heat stress” condition in prior reports (4–9, 61). We postulate this was due to the smaller elevation in body temperature in the present report. It should be noted that DBP and MAP were lower than the baseline values without MSNA activation due to the moderate heating condition, since these were not observed during the normothermic control trial.

Prior studies suggested that cardiac systolic function “improved” in young and middle-aged subjects during heat stress (31). In the present study, moderate WBH raised cardiac output and did not alter stroke volume. Cardiac function indexes for systolic (S_m) and diastolic (E/E_m) did not change from baseline. After heating, E and E_m decreased, which were simply induced by the decrease in RR interval. Thus, the present data suggest that moderate WBH only raises cardiac output and may not significantly alter myocardium function in older individuals.

Responses to Fatiguing Handgrip Exercise

During the WBH exercise trial, the absolute MAP was lower and HR was higher than during the normothermic trial. Moreover, MSNA and MAP changes during the WBH exercise trial were lower than those during the normothermic trial. These observations are different from those seen in a prior study performed on young healthy subjects (22). When compared with the data in that study (8, 22), the changes in MSNA, HR, and MAP evoked by fatiguing handgrip in older subjects were less than the responses seen in the young subjects. Thus, we speculate that aging might predominantly contribute to the lower MSNA response to handgrip under heating conditions, although the role of the difference in the heating condition cannot be excluded thoroughly.

To examine the effects of T_muscle on muscle afferent activity in the exercising muscles during moderate WBH, the T_muscle was raised to a similar level during the WBH and LH trials. LH did not alter resting hemodynamics or MSNA. Hemodynamic and MSNA responses (both absolute values and the changes) during the LH exercise paradigm were not different from those seen with exercise under normothermic conditions. This suggests that the attenuated MSNA and BP seen with exercise during moderate WBH conditions were not due to temperature effects on the local exercising muscle.

The results from the LH trial in the present studies were different from those seen in prior studies (36, 62, 63), which demonstrated that local heating applied to an isometrically exercising forearm muscle group in young and older subjects augmented the increase in MSNA and BP observed during active fatiguing exercise. In these prior studies, the forearm muscle temperature was raised from ∼34°C to 39°C. In the present study, the T_muscle only rose by ∼Δ1.4°C. We speculate that the differences in results between the present report and the prior studies may have been due to the smaller increases in muscle temperature achieved in the present experiments.

Responses to Stimulation of Metaboreceptors and Mechanoreceptors

In the present study, MSNA responses during PECO under moderate WBH condition were not different from that in normothermic control trial. This is consistent with the findings of a prior study on young subjects (22). Moreover, MSNA responses during PECO in the LH trial were not different from that in the normothermic control trial, which was consistent with prior observations (36, 63). Compared with young subjects (8, 22), the MSNA responses to PECO under WBH condition in the older subjects were lower.

Moderate WBH led to a decrease in absolute MAP during PECO (Fig. 4) and a decrease in the MAP response to PECO (Fig. 5). MSNA and HR responses to PECO during the WBH trial (Fig. 5) were not different from those seen during the normothermic trial. These were similar to the observations in the prior study on young subjects (22). We speculate that the following factors may contribute to this observation. The moderate WBH raised by ∼Δ135% in SkBF and Δ149% in CVC. After a large increase in blood flow/volume in the cutaneous vascular beds, a similar increase in sympathetic vasoconstrictor activity directed toward the skeletal muscles under WBH conditions may not have been sufficient to increase systemic BP to the same degree as was seen under normothermic conditions.

The present results show that passive muscle stretch during PECO evoked significant MSNA and MAP increases as compared with MSNA and MAP seen during PECO alone during normothermia, a finding consistent with prior reports (21, 22). Under moderate WBH conditions, passive muscle stretch during PECO also evoked MSNA (total activity) and MAP increases from the PECO alone condition. The data suggest that stimulation of muscle mechanoreceptor contributes to the sympathetic and pressor response to exercise under the moderate WBH condition. In a prior study (22), “heat stress” abolished the MSNA and MAP increases by PECO + stretch from the PECO alone condition in young healthy subjects. The difference in the observations we found in this report compared with the prior studies may be due to the differences in subject populations studied and the heating conditions. The MAP response during stretch under the moderate WBH condition was lower than that in the normothermic trial. This could be due to the greater total peripheral vasodilation with WBH.

The results from whole exercise paradigm suggest that the MSNA and BP responses to exercise are attenuated under the moderate WBH condition, while the sensitivities of the metaboreceptor and mechanoreceptors in the forearm muscles may not be altered. We speculate that the following systemic factors may contribute to attenuated MSNA and MAP responses to handgrip exercise under moderate WBH conditions. First, the moderate WBH directly induced increases in whole body cutaneous vasodilation via local mechanisms, which led to increases in peripheral blood flow and HR and cardiac output. These changes were different with limb heating. The increased resting blood flow might alter/increase tissue perfusion, which might not be limited to the skin. Prior studies have shown that whole body “heat stress” increases SkBF, muscle blood flow (64), and MSNA (7). There could be a counterbalance between the vasodilatory effects of heating and the vasoconstricting effects of MSNA activation under the “heat stress” condition. In the present study, MSNA was not different from the normothermic baseline, whereas the T_muscle was elevated. Thus, we speculate that muscle blood flow might be increased by heating via local mechanisms. Under this prior exercise condition, a less BP increase might induce an elevation in the tissue perfusion that is enough for the exercise. Second, whole body temperature was elevated with WBH. Thus, the temperature in the entire neuropathway for exercise pressor reflex should be elevated, which was different from the LH trial. Although the sensitivities of the metaboreceptors and mechanoreceptors in the forearm muscles might not be altered, the receptors in the pathway (e.g., dorsal root ganglion neurons) could be altered by the heating, which might in turn contribute to the attenuated MSNA and MAP responses. This speculation can be examined with animal studies in the future.

Study Limitations

In this study, we tried to match the T_muscle elevation in the LH trial with those seen in the WBH trial. However, the body temperature in humans is controlled by multiple physiological mechanisms whose strength and influence vary from subject to subject. Thus, it is very difficult to make fine adjustments to control (internal and muscle) temperature while adjusting water temperature in the suit. Thus, the conditions in the two trials were only roughly “matched” (i.e., no significant difference in T_muscle during before exercise baseline under the two study conditions). The studies were performed in the supine position, which might not be an optimal body position for echocardiography (e.g., left lateral position). However, all echocardiography measurements were performed in the same position, and the influences of the body position on the data should be the same for all conditions. Thus, the body position for echocardiography measurements should not affect the conclusion on cardiac function in the study. The present study was not designed to investigate race-related differences, and hence, the results cannot be extrapolated to black individuals. The race-related differences should be examined in future studies.

Perspectives and Significance

The presented results suggest that in older individuals, mild WBH decreases resting MSNA and BP, whereas moderate WBH decreases BP without evoking a rise in MSNA. Moreover, as compared with exercise performed under normothermic conditions, the MSNA and BP responses to exercise are attenuated during moderate WBH. The mild and moderate heating exposures used in these studies are frequently encountered by normal older individuals under normal environmental and activity levels, and we speculate the findings may be important for the following reasons: First, we postulate that mild heat exposure can decrease resting MSNA. Higher resting MSNA levels are seen with aging (25) and with many cardiovascular conditions including hypertension and heart failure (65, 66). Indeed, heightened sympathetic activity is associated with worsening cardiac performance in heart failure and may predispose patients to the development of atrial and ventricular arrhythmias (67–70). Second, lowering BP responses to exercise may also be important since it is known that higher BP responses to exercise are associated with a higher incidence of cardiovascular disease (71). On the other hand, it should be considered that the absolute HR during exercise under the moderate WBH condition will be higher than under normothermic conditions. A higher (∼Δ11 beats/min) resting HR before exercise would decrease the buffer space for the HR increase during exercise. Prior epidemiological studies (72) suggested that a higher resting HR and a less increase in HR during exercise might be linked with sudden death. Although a period of higher HR induced by moderate WBH is different from the high resting HR in daily life, the risk associated with high HR in older individuals should be considered during intensive exercise under heating conditions.

In conclusion, the present data show that in older individuals, mild WBH decreases resting MSNA and BP and increases HR and cardiac output. Moreover, the results suggest that moderate WBH attenuates MSNA and BP responses to exercise in older healthy humans.

GRANTS

This work was supported by National Institutes of Health Grants MPI R01 HL141198, UL1 TR002014, and the American Heart Association Grant 0565399U.

DISCLOSURES

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

AUTHOR CONTRIBUTIONS

J.C. and L.I.S. conceived and designed research; J.C., Z.G., C.B., J.C.L., and K.B. performed experiments; J.C., Z.G., C.B., J.C.L., and K.B. analyzed data; J.C. and L.I.S. interpreted results of experiments; J.C. prepared figures; J.C. and L.I.S. drafted manuscript; J.C. and L.I.S. edited and revised manuscript; J.C., Z.G., C.B., J.C.L., K.B., and L.I.S. approved final version of manuscript.