Fluid replacement modulates oxidative stress- but not nitric oxide-mediated cutaneous vasodilation and sweating during prolonged exercise in the heat

Abstract

The roles of nitric oxide synthase (NOS), reactive oxygen species (ROS), and angiotensin II type 1 receptor (AT₁R) activation in regulating cutaneous vasodilation and sweating during prolonged (≥60 min) exercise are currently unclear. Moreover, it remains to be determined whether fluid replacement (FR) modulates the above thermoeffector responses. To investigate, 11 young men completed 90 min of continuous moderate intensity (46% V̇o_2peak) cycling performed at a fixed rate of metabolic heat production of 600 W (No FR condition). On a separate day, participants completed a second session of the same protocol while receiving FR to offset sweat losses (FR condition). Cutaneous vascular conductance (CVC) and local sweat rate (LSR) were measured at four intradermal microdialysis forearm sites perfused with: 1) lactated Ringer (Control); 2) 10 mM N^G-nitro-l-arginine methyl ester (l-NAME, NOS inhibition); 3) 10 mM ascorbate (nonselective antioxidant); or 4) 4.34 nM losartan (AT₁R inhibition). Relative to Control (71% CVC_max at both time points), CVC with ascorbate (80% and 83% CVC_max) was elevated at 60 and 90 min of exercise during FR (both P < 0.02) but not at any time during No FR (all P > 0.31). In both conditions, CVC was reduced at end exercise with l-NAME (60% CVC_max; both P < 0.02) but was not different relative to Control at the losartan site (76% CVC_max; both P > 0.19). LSR did not differ between sites in either condition (all P > 0.10). We conclude that NOS regulates cutaneous vasodilation, but not sweating, irrespective of FR, and that ROS influence cutaneous vasodilation during prolonged exercise with FR.

despite the importance of cutaneous vasodilation and sweating to the maintenance of a stable body core temperature during exercise and/or exposure to a hot environment, the physiological mechanisms governing these heat loss responses at the level of the end organ (i.e., the cutaneous vasculature and eccrine sweat gland) have yet to be fully elucidated. Numerous studies have shown nitric oxide (NO) to be a key modulator of cutaneous vasodilation and sweating during dynamic exercise lasting 30–45 min (10, 32, 35–38, 63) or passive heat stress (22, 24, 26, 60). However, exercise of longer duration (i.e., ≥60 min) may reduce NO bioavailability. Specifically, prolonged exercise is associated with the increased production of reactive oxygen species (ROS), such as superoxide, which may gradually overwhelm the scavenging capacity of endogenous antioxidants (19, 33, 46). The accumulation of superoxide can subsequently reduce NO bioavailability by directly interacting with NO to form peroxynitrate (43, 47, 50, 53), a potent ROS itself, and/or by uncoupling NO synthase (NOS). Indeed, a role for exercise-induced ROS in limiting NO-dependent cutaneous vasodilation was confirmed during 30-min bouts of high intensity (70% V̇o_2peak) exercise in the heat (35°C) (38). However, it remains to be determined if oxidative stress produced during prolonged exercise limits cutaneous vasodilation and sweating.

During prolonged exercise (3, 54), passive heat stress (7), and exercise in the heat (2), there can be increases in the plasma aldosterone and plasma renin activity (downstream indicators of angiotensin II concentration). Intradermal administration of angiotensin II (ANG II) has been shown to reduce cutaneous vasodilation and sweating during passive exposure to a hot environment (35°C) both before and after moderate-intensity exercise (12). ANG II-mediated reductions in cutaneous blood flow are thought to be due to ROS-dependent mechanisms (57–59). Specifically, the activation of ANG II Type 1 receptors (AT₁R) elevates ROS production through NADPH oxidase (15, 16, 18). Thus, the AT₁R-mediated reductions in cutaneous vasodilation and sweat rate could be a result of increased oxidative stress.

Prolonged exercise in the heat may cause substantial disruptions to the regulation of cutaneous vasodilation and sweating due to excess fluid losses. Hypohydration is well known to directly impair thermoregulatory function (48). Notably, passive heat stress-induced hypohydration has also been shown to increase markers of oxidative stress (45). Thus, prolonged exercise without fluid replacement could further exacerbate oxidative stress inherent to exercise of long duration. Furthermore, while some have suggested that levels of ANG II do not change with dehydration (34), others have shown that hypohydration resulting from prolonged exercise without fluid replacement increases circulating ANG II (42), ultimately potentiating further ROS production and the consequent reductions in cutaneous vasodilation and sweating as mentioned above.

Prolonged exercise without fluid replacement can cause significant disruptions to the physiological mechanisms of heat loss; therefore, gaining a better understanding of the ROS- and AT₁R-mediated regulations of cutaneous blood flow and sweating could have profound implications for body core temperature regulation. Thus, the present study was aimed to evaluate the influence of NO, ROS, and AT₁R on cutaneous vasodilation and sweating during prolonged exercise (90 min) in the heat (40°C) with and without fluid replacement. Given that previous work showed ROS-mediated reductions in cutaneous vasodilation through NO- (38) and ANG II-dependent mechanisms (58, 59), we hypothesized that cutaneous vasodilation and sweating would be augmented by local ascorbate (a nonselective antioxidant) administration and AT₁R blockade during prolonged moderate-intensity exercise in the heat. Second, we hypothesized that fluid replacement would reverse the ROS- and ANG II-mediated reductions, if any, in the heat loss responses.

METHODS

Ethical Approval

The current study was approved by the University of Ottawa Health Sciences and Science Research Ethics Board and is in accordance with the guidelines set forth by the Declaration of Helsinki. Verbal and written informed consent was obtained from all volunteers before their participation in the study.

Participant Information

Eleven young males participated in this study. All participants were healthy (i.e., no known history of cardiovascular, respiratory, and/or metabolic diseases), highly active, nonsmoking, and not taking prescription medications at the time of the study. The physical characteristics of the participants (means ± SD) were the following: age, 26 ± 5 yr; height, 1.81 ± 0.04 m; body mass, 78.4 ± 8.1 kg; body surface area, 2.0 ± 0.1 m²; body fat percentage, 11 ± 4%; and peak rate of oxygen consumption (V̇o_2peak), 60.0 ± 8.2 ml O₂·kg⁻¹·min⁻¹.

Experimental Procedures

Each participant completed one screening and two experimental sessions. All participants were instructed to refrain from over-the-counter medications (including nonsteroidal anti-inflammatory drugs and supplements) for a minimum of 48 h, as well as alcohol, caffeine, and heavy exercise at least 24 h before each session. Furthermore, on the day of each session participants were instructed not to consume food for 2 h before arriving to the laboratory. During the screening session, body height, mass, surface area, density, fat percentage, and V̇o_2peak were determined. Body height was measured using an eye-level physician stadiometer (model 2391; Detecto, Webb City, MO) while body mass was measured using a digital weight scale platform (model CBU150X, Mettler Toledo) with a weighing terminal (model IND560, Mettler Toledo). Subsequently, body surface area was calculated from the measurements of body height and mass (6). Body density was evaluated by means of hydrostatic weighing and used to calculate body fat percentage (52). V̇o_2peak was assessed using an automated indirect calorimetry system (MCD Medgraphics Ultima Series, Sun Tech Medical, Morrisville, NC) during a progressive incremental cycling protocol on a semirecumbent cycle ergometer (Corival; Lode BV, Groningen, The Netherlands). During this protocol, participants were instructed to maintain a pedaling cadence between 60 and 90 revolutions/min at a starting workload of 120 W for 1 min. The resistance was increased incrementally by 20 W/min, thereafter, until volitional failure and/or until a pedaling cadence above 50 revolutions/min could no longer be maintained. V̇o_2peak was then evaluated as the highest average rate of oxygen uptake measured over 30 s.

Participants performed two experimental trials on separate days, with a minimum of 48 h between trials. Upon arrival to the laboratory, participants provided a urine sample for the determination of urine specific gravity (USG) and voided their bladders before a nude body mass measurement was taken. Participants who arrived with USG measurements above 1.020 were deemed dehydrated (1) and therefore were ineligible to complete the experimental protocol. If adequately hydrated, participants rested passively in an upright seated position for a 30-min instrumentation period at ambient room temperature (~23°C). During this time, four microdialysis membranes fibers (30 kDa cutoff; MD 2000, Bioanalytical Systems, West Lafayette, IN) were placed in the dermal layer of the left dorsal forearm of the participants. All microdialysis fibers were separated by a minimum of 4 cm and inserted under aseptic conditions by using a 25-gauge needle. The needle entry and exit sites were 20–25 mm apart thereby ensuring that the full membrane rested below the surface of the skin. The microdialysis fiber was then threaded through the lumen of the needle, which when subsequently withdrawn, left the 10-mm dialysis membrane of the fiber beneath the skin. All fibers were secured in place for the remainder of the trial with surgical tape.

After fiber insertion, participants were moved to a thermally controlled chamber (Can-Trol Environmental Systems Limited, Markham, ON, Canada) regulated to 40°C and 20% relative humidity. In the chamber, participants rested quietly while seated on a semirecumbent cycle ergometer (Corival; Lode BV) while the microdialysis fibers were perfused in a counter-balanced manner with either: 1) lactated Ringer solution (Baxter, Deerfield, IL) (Control); 2) 10 mM N^G-nitro-l-arginine methyl ester (l-NAME, Sigma-Aldrich, St. Louis, MO) to nonselectively inhibit NOS; 3) 10 mM ascorbate (Sigma-Aldrich) to nonselectively reduce ROS; or 4) 4.34 nM losartan (Sigma-Aldrich) to inhibit the action of AT₁R at a rate of 4 μl/min via a perfusion pump (model 400; CMA Microdialysis, Solna, Sweden). Each site was instrumented for the measurement of local sweat rate and cutaneous blood flow (see Measurements). The concentration of l-NAME was determined based on previous literature of microdialysis on human skin (10, 11, 22, 24, 38, 55) as were the concentrations for ascorbate (23, 24, 38, 59) and losartan (31, 58, 59).

All fibers were perfused with their respective physiological agents throughout a habituation period to ensure complete pharmacological blockade and for needle trauma to subside (~90 min) (21). Moreover, 90 min of resting in the heat were employed to ensure that heat balance had been achieved in all participants before baseline measurements (29, 56). Next, a 10-min baseline resting data collection period ensued. After baseline, participants performed 90 min of semirecumbent cycling exercise at 600 W of metabolic heat production followed by a 40-min recovery period. A fixed rate of heat production of 600 W (calculated as metabolic rate minus external workload) was determined from pilot work that evaluated the metabolic heat load that was achieved during an established prolonged exercise protocol where participants exercised against a constant load of 120 W (13, 14, 39). This workload was equivalent to an exercise intensity of 46 ± 3% (means ± SD) V̇o_2peak for the study participants. In controlling for the rate of the metabolic heat production between trials rather than a given percentage of maximum oxygen consumption, we aimed to reduce potential differences in fluid loss between experimental sessions and between participants (14). At the end of the recovery period, each microdialysis fiber was perfused with 50 mM of sodium nitroprusside (SNP, Sigma-Aldrich) to determine the maximum cutaneous blood flow at each skin site. SNP administration continued at a rate of 6 μl/min until a stable plateau cutaneous blood flow was achieved for a minimum of 2 min. Thereafter, blood pressure was measured to evaluate a maximum cutaneous vascular conductance (CVC_max). Finally, all instrumentation was removed from the participants’ forearms and a posttrial nude body mass was measured. The difference between pretrial nude body mass and posttrial nude body mass was used to evaluate fluid loss via sweating. Participants also provided a posttrial urine sample to evaluate any changes in hydration status.

In the first experimental session, participants performed the above-mentioned experimental protocol with no fluid replacement (No FR). The change in body weight exhibited in the No FR trial was measured and used to determine the amount of water required to maintain the individual’s body weight during the second or fluid replacement trial (FR). Because of this design, the No FR condition was performed first in all participants. To minimize any carryover effect from the first trial, each participant’s trials were separated by a minimum of 72 h between FR and No FR trials. Fluid replacement consisted of tap water, a container of which was filled with temperate water before the start of the trial and kept within the environmental chamber to stabilize its temperature near to that of the room (40°C). The water was administered in boluses of 500–700 ml 5 min before the Baseline resting measurement and at the 30th, 60th, and 90th min of exercise. This range of fluid replacement is in line with a previous study using a similar experimental protocol (14). The No FR condition allowed for the observation of the influence of progressive exercise-induced fluid loss via sweating, while the FR condition allowed for the observation of responses due to prolonged exercise without progressive dehydration. Aside from the fluid replacement, the experimental protocol for the two sessions was identical.

Measurements

Cutaneous red blood cell flux, an index of cutaneous blood flow, was measured at a sampling rate of 32 Hz with laser Doppler flowmetry (PeriFlux System 5000, Perimed, Stockholm, Sweden). Four integrated 7-laser array Doppler flowmetry probes were each housed in a specialized sweat capsule positioned directly over the center of each of the four microdialysis membrane to allow for the simultaneous measurement of cutaneous red blood cell flux and local sweat rate (40). CVC was calculated as the red blood cell flux divided by mean arterial pressure and presented as a percentage of CVC_max obtained during the SNP-induced maximum vasodilation protocol. Mean arterial pressure was calculated as diastolic pressure plus one-third of the difference between systolic and diastolic pressures (i.e., pulse pressure), which were measured using manual auscultation with a validated mercury column sphygmomanometer (Baumanometer Standby model, WA Baum, Copiague, NY) at 5-min intervals through the experimental protocol.

Local forearm sweat rate was simultaneously evaluated at each skin site by a ventilated capsule (1.1 cm²) specifically designed to encompass the area of skin perfused by the intradermal microdialysis fiber (40). The sweat capsules were secured directly over the center of the microdialysis membrane using adhesive rings and topical skin glue (Collodion HV, Mavidon Medical Products, Lake Worth, FL). Anhydrous air was passed through each sweat capsule at a rate of 0.4 l/min, while the water content of the effluent air was measured using capacitance hygrometers (model HMT333, Vaisala, Helsinki, Finland). To ensure that the internal gas tanks were equilibrated to near room temperature, the gas tanks were located in the thermal chamber and connected to the sweat capsules and hygrometers via vinyl tubing. Local sweat rate was calculated every 5 s using the water content of the effluent air multiplied by flow rate and normalized for skin surface area beneath the capsule (expressed in mg·min⁻¹·cm⁻²).

Esophageal temperature was measured using a pediatric thermocouple probe of ~2 mm diameter (Mon-a-therm; Mallinckrodt Medical, St. Louis, MO) inserted ~40 cm past the entrance of the nostril and confirmed every 5 min with aural canal temperature measurement (Welch Allyn Braun ThermoScan Pro 6000, Braun, Kronberg, Germany). Mean skin temperature was calculated from the temperature measured at four skin sites using T-type copper thermocouples (Concept Engineering, Old Saybrook, CT) weighted to the following regional proportions: chest, 30%; biceps, 30%; quadriceps, 20%; and calf, 20% (17). All temperature data were collected using a data acquisition module (model 34970A; Agilent Technologies Canada, Mississauga, ON, Canada) at a sampling rate of 15 s and simultaneously displayed and recorded using LabVIEW software, version 7.0 (National Instruments, Austin, TX). Mean body temperature was calculated as (0.9 × esophageal temperature) + (0.1 × mean skin temperature). Heart rate was measured using a Polar coded WearLink and transmitter, Polar RS400 interface, and Polar Trainer 5 software (Polar Electro, Kempele, Finland).

A total solids refractometer (model TS400, Reichter, Depew, NY) was used to evaluate USG from the urine samples obtained before and after the experimental protocol.

Metabolic energy expenditure was measured using indirect calorimetry. Electrochemical gas analyzers (AMETEK model S-3A/1 and CD3A, Applied Electrochemistry, Pittsburgh, PA) measured expired air for concentrations of O₂ and CO₂. The gas analyzers were calibrated ~20 min before baseline using a gas mixture of known concentration. The turbine ventilometer was calibrated using a 3-liter syringe. Participants wore a partial face mask (model 7600 V2, Hans-Rudolph, Kansas City, MO) attached to a two-way T-shape nonrebreathing valve (model 2700, Hans-Rudolph). Metabolic rate was calculated using oxygen uptake and respiratory exchange ratio values averaged over 30s. The rate of metabolic heat production was taken as the difference between metabolic rate and external work.

Data Analysis

Baseline resting values were determined by averaging the data collected over the 5 min before the start of the 90-min exercise bout. CVC and local sweat rate at each skin treatment site, all temperature measurements, and heart rate were evaluated by averaging data collected during the final 10 min of each 30-min interval during exercise and the final 10 min of each 20-min interval during recovery. Furthermore, blood pressure data were presented as the average of the two measurements taken during the final 10 min of each interval mentioned above. CVC_max was obtained from data averaged over a 2-min period once a stabilized plateau occurred during the maximal vasodilation protocol.

Statistical Analysis

For the purpose of statistical comparison, the exercise and recovery periods of the experimental protocol were defined based on the following time periods: Baseline, −5 to 0 min; Exercise, 0 to 90 min; and Recovery, 90 to 130 min. To assess the influence of each treatment and condition on the heat loss responses throughout the experimental protocol, a three-way repeated measures ANOVA was performed with the factors of treatment site (4 levels: Control, l-NAME, ascorbate, and losartan), time (6 levels: baseline, 30, 60, 90, 110, and 130 min), and condition (2 levels: No FR and FR) for both CVC and local sweat rate. Furthermore, to assess the influence of fluid replacement on the body temperature (i.e., esophageal temperature, mean skin and mean body temperature) and cardiovascular (i.e., heart rate and blood pressure) responses, an additional two-way mixed model ANOVA was performed with the factors of time (6 levels: baseline, 30, 60, 90, 110, and 130 min) and condition (2 levels: No FR and FR) was used.

To compare CVC_max (expressed as perfusion units/mmHg) between each treatment site in both trials obtained during SNP administration, we performed a two-way repeated measures ANOVA with the factors of treatment site (4 levels: Control, l-NAME, ascorbate, and losartan) and condition (2 levels: No FR and FR). A Student’s paired samples t-test was used to compare pretrial body mass and the percent change in body mass during the experimental sessions between No FR and FR trials. Post hoc comparisons were conducted using Student’s paired two-tailed t-tests, in which P values were adjusted using a modified Bonferroni correction (Holm-Bonferroni’s method) when a significant interaction or main effect was detected. For all analyses, P ≤ 0.05 was considered statistically significant. All values were presented as means ± 95% confidence intervals (calculated as 1.96 × SEM) unless otherwise indicated. Statistical analyses were performed using software package SPSS 24.0 for Windows (IBM, Armonk, NY).

RESULTS

Hydration Satus and Fluid Balance

No differences in pretrial USG (No FR, 1.008 ± 0.002; FR, 1.009 ± 0.003; P = 0.78) were observed, whereas posttrial USG was greater following No FR (1.019 ± 0.004) trial in comparison to FR (1.007 ± 0.001; P < 0.01). All participants arrived with USG measurements below 1.020 and as such were eligible to complete the experimental protocol. Pretrial body weight (No FR, 79 ± 5 kg; FR, 79 ± 9 kg; P = 0.08) was not different between conditions; however, the percent change in body weight was greater during No FR (Table 1; −3.4 ± 0.5%) compared with FR (−0.1 ± 0.0%; P < 0.01). In FR, participants received an average of 2.6 ± 0.5 liters of water, accounting for 89% of total fluid loss in No FR.

Table 1.

Body weight changes before and after prolonged exercise in the heat

	Preweight, kg	Postweight, kg	% Change in Body Weight	Volume of Water Consumed, liters
Trial condition
No FR	78.8 ± 4.7	76.2 ± 4.5	−3.57 ± 0.44	0.4 ± 0.1
FR	79.4 ± 5.0	78.8 ± 4.8†	−0.01 ± 0.00†	2.6 ± 0.3†

Values are means ± 95% confidence interval. Eleven young adults performed a continuous 90-min exercise protocol followed by a 40-min recovery period. Exercise was performed at a fixed rate of heat production of 600 W (46% V̇o_2peak). No FR, no fluid replacement; FR, fluid replacment.

^†Significant difference vs. No FR; P ≤ 0.05.

Body Temperature and Cardiovascular Responses

Esophageal, mean skin, and mean body temperatures (Table 2; main effect of time, all P < 0.01) were elevated from baseline values throughout exercise and recovery (all P ≤ 0.01). Moreover, esophageal and mean body temperatures (main effect of condition, all P ≤ 0.01) were reduced during the FR condition relative to No FR from the 60-min time point of exercise until the end of recovery (all P ≤ 0.02). Mean skin temperature was reduced in FR relative to No FR at 20 min of recovery only (Table 2; P = 0.05).

Table 2.

Body temperatures and cardiovascular responses at rest, during, and after prolonged exercise in the heat

		Exercise			Recovery
	Resting Baseline	30 min	60 min	90 min	20 min	40 min
Esophageal temperature, °C
No FR	37.1 ± 0.1	37.9 ± 0.2*	38.5 ± 0.2*	39.0 ± 0.3*	38.3 ± 0.3*	37.9 ± 0.2*
FR	37.0 ± 0.1	37.7 ± 0.2*	38.3 ± 0.3*†	38.7 ± 0.3*†	37.9 ± 0.3*†	37.6 ± 0.3*†
Mean skin temperature, °C
No FR	35.6 ± 0.2	36.6 ± 0.2*	37.1 ± 0.3*	37.5 ± 0.3*	37.1 ± 0.5*	36.5 ± 0.5*
FR	35.7 ± 0.2	36.6 ± 0.2*	37.0 ± 0.2*	37.4 ± 0.3*	36.8 ± 0.3*†	36.2 ± 0.3*
Mean body temperature, °C
No FR	37.0 ± 0.1	37.7 ± 0.2*	38.3 ± 0.2*	38.9 ± 0.3*	38.2 ± 0.3*	37.7 ± 0.3*
FR	36.9 ± 0.1	37.6 ± 0.2*	38.1 ± 0.3*†	38.6 ± 0.3*†	37.8 ± 0.3*†	37.4 ± 0.3*†
Mean arterial pressure, mmHg
No FR	93 ± 3	99 ± 4*	97 ± 4	97 ± 4	85 ± 4*	86 ± 3*
FR	91 ± 3	95 ± 3*†	94 ± 4	94 ± 4	86 ± 5	89 ± 4
Heart rate, beats/min
No FR	70 ± 3	131 ± 6*	146 ± 8*	157 ± 8*	114 ± 8*	103 ± 7*
FR	69 ± 5	124 ± 8*†	132 ± 8*†	141 ± 9*†	97 ± 7*†	86 ± 7*†

Presented values are means ± 95% confidence interval. Eleven young adults performed a continuous 90-min exercise protocol followed by a 40-min recovery period. Exercise was performed at a fixed rate of heat production of 600 W (46% V̇o_2peak). Esophageal and mean skin temperatures as well as heart rate responses represent an average of the final 10 min of the corresponding time period. Mean arterial pressure values represent an average of two measurements taken over the final 10 min of the corresponding time period. No FR, no fluid replacement; FR, fluid replacment.

^*Significant difference vs. baseline;

^†significant difference vs. No FR; P ≤ 0.05.

In both conditions, mean arterial pressure (Table 2; main effect of time, P = 0.03) was increased relative to baseline during the first 30 min of exercise but similar thereafter until the end of exercise (all P ≥ 0.15). Mean arterial pressure (main effect of condition, P = 0.04) was elevated during the first 30 min of exercise in the No FR condition relative to the FR condition (P = 0.01) but not different at any other time point (all P > 0.07). Mean arterial pressure was reduced relative to baseline throughout the recovery period in the No FR (>7 mmHg decrease, both P < 0.01) but not the FR condition (both P ≥ 0.07). Heart rate (Table 2; main effect of time, P < 0.01) was increased relative to baseline throughout exercise and recovery in both conditions (all P < 0.01) and was increased in the No FR condition relative to the FR condition from 30 min of exercise until the end of the protocol (main effect of condition, all P ≤ 0.02).

Local Forearm Cutaneous Vascular Conductance Response

No fluid replacement condition.

During No FR, CVC (Fig. 1A; main effects of time and treatment site, both P ≤ 0.01) was elevated throughout exercise and recovery at all sites relative to their respective baseline values (all P ≤ 0.01). Throughout baseline, exercise and recovery, CVC (Fig. 1A; main effect of condition P ≤ 0.01) at the l-NAME-treated site was reduced relative to Control (all P ≤ 0.04); whereas, ascorbate or losartan sites were similar to Control (all P ≥ 0.14).

Fig. 1.

Time-course changes in cutaneous vascular conductance (CVC) for participants (n = 11) in both the no-fluid replacement (No FR) condition (A) and fluid replacement (FR) condition (B) at baseline resting (BL) during a 90-min prolonged exercise bout performed at a fixed rate of metabolic heat production (600 W) and during the postexercise recovery period. Four skin sites were continuously administered with 1) lactated Ringer solution (CON); 2) 10 mM N^G-nitro-l-arginine methyl ester (l-NAME); 3) 10 mM ascorbate (ASC); or 4) 4.34 nM losartan (LOS). *Control significantly different from l-NAME(P < 0.05); †control significantly different from ASC (P < 0.05).

Fluid replacement condition.

In parallel with the No FR condition, CVC (Fig. 1B; Fig. 1A; main effects of time and treatment site, both P ≤ 0.01) in the FR condition was elevated during exercise and recovery at all sites relative to their respective baseline values (all P ≤ 0.01). CVC was reduced at the l-NAME-treated site compared with Control during baseline resting (P < 0.01) and throughout exercise and recovery (all P < 0.01). In contrast, CVC was similar at the ascorbate- or losartan-treated sites during baseline resting (both P > 0.29). Whereas no differences were observed between the losartan-treated site and Control throughout the protocol (all P > 0.19), CVC at the ascorbate-treated site was increased relative to Control for the final 30 min of exercise (both P < 0.03). CVC at the ascorbate-treated site returned to values similar to the Control site throughout the 40-min recovery period (both P > 0.38).

Between fluid replacement conditions.

No effect of fluid replacement condition was observed on CVC (P = 0.25) condition.

Maximal CVC Response.

Maximum CVC during administration of 50 mM of SNP did not differ between treatment sites within each condition (Table 3; main effect of treatment site, P = 0.35). Furthermore, there was a main effect of condition on maximum CVC (Table 3; P = 0.04); however, no statistical differences were observed after correcting for multiple comparisons (all P ≥ 0.09).

Table 3.

Absolute maximal cutaneous vascular conductance at four skin sites

	Control	l-NAME	Ascorbate	Losartan
CVCmax, perfusion units/mmHg
No FR	2.3 ± 0.3	2.3 ± 0.5	2.1 ± 0.3	2.4 ± 0.6
FR	2.6 ± 0.3	2.7 ± 0.4	2.5 ± 0.5	3.0 ± 0.9

Values are means ± 95% confidence interval. Eleven young adults performed a continuous 90-min exercise protocol followed by a 40-min recovery period. Exercise was performed at a fixed rate of heat production of 600 W (46% V̇o_2peak). CVC_max measured at the four skin sites previously perfused with either 1) lactated Ringer solution (Control); 2) 10 mM l-NAME to nonselectively inhibit NOS; 3) 10 mM ascorbate, an antioxidant; or 4) 4.34 nM losartan to inhibit the action of AT₁R. No statistically significant differences were detected. No FR, no fluid replacement; FR, fluid replacment.

Local Forearm Sweating Response

Local sweat rate (Fig. 2; main effect of time P < 0.01) was elevated relative to baseline values throughout exercise and recovery at all treatment sites during both No FR and FR (all P ≤ 0.01) but did not differ between treatment sites (main effect of treatment site, P = 0.24) or fluid replacement condition (main effect of treatment site, P = 0.78).

Fig. 2.

Time-course changes in local sweat rate (LSR) for participants (n = 11) in both the no-fluid replacement (No FR) condition (A) and fluid replacement (FR) condition (B) at baseline resting (BL) during a 90-min prolonged exercise bout performed at a fixed rate of metabolic heat production (600 W), and during the postexercise recovery period. Four skin sites were continuously administered with 1) lactated Ringer solution (CON); 2) 10 mM N^G-nitro-l-arginine methyl ester (l-NAME); 3) 10 mM ascorbate (ASC); or 4) 4.34 nM losartan (LOS). No statistical differences between sites were detected.

DISCUSSION

The primary finding of the present study is that intradermal administration of ascorbate augmented cutaneous vasodilation but not sweating during moderate intensity (46% V̇o_2peak) prolonged exercise in the heat after 60 min of exercise when fluid was provided to offset losses via sweat. This observation indicates that long-duration exercise even at moderate intensity can cause a substantial physiological burden limiting heat loss, likely through increased ROS accumulation. Additionally, NOS was shown to be a key modulator of cutaneous vasodilation but not sweating in response to prolonged exercise in the heat, irrespective of fluid replacement. In contrast, we did not see an effect of local AT₁R inhibition on cutaneous vasodilation or sweating during or after exercise in the heat with or without fluid replacement. Altogether, these data demonstrate that during prolonged exercise in the heat, the role of NO on cutaneous vasodilation is not dependent on fluid replacement, albeit we showed that local ascorbate treatment augments cutaneous vasodilation with fluid replacement.

Cutaneous Vascular Response

Consistent with previous reports we show NOS inhibition (10, 11, 22, 24, 26, 38, 63), but not ascorbate (38) nor losartan (31) administration, to be an essential modulator of cutaneous vascular tone during passive exposure to a hot environment (40°C). At the initiation of exercise in both conditions, CVC at the NOS-inhibited site remained reduced relative to Control throughout the experimental protocol, thereby demonstrating the importance of NO to cutaneous vasodilation during and after prolonged exercise-induced heat stress. Importantly, this role for NOS inhibition was conserved irrespective of condition, suggesting NOS-mediated cutaneous vasodilation is not dependent on fluid replacement.

In the current study, we show that intradermal administration of ascorbate increased cutaneous vasodilation during prolonged exercise in the heat when fluid was provided to offset fluid losses via sweat. Recently, local ascorbate administration has been shown to augment cutaneous vasodilation in young adults performing 30 min of high intensity (70% V̇o_2peak) exercise in the heat in a NO-dependent manner (38). It was suggested that ascorbate administration served to reduce an increase in ROS resulting from high-intensity exercise and thereby promoted NO bioavailability and cutaneous vasodilation (38). However, while there is also a progressive accumulation of ROS during prolonged exercise (5, 19, 33, 46), it remained to be determined if it was sufficient to modulate the body’s heat loss responses. Importantly, no effect of ascorbate supplementation was observed until 60 min of exercise. While speculative, it may be that the later stages (≥60 min) of prolonged exercise in the heat resulted in the progressive accumulation of oxidative stress such that the accumulation of ROS limited the ability to increase cutaneous vasodilation.

Contrary to our hypothesis, the augmented cutaneous vasodilation at the ascorbate-treated site was only present during the FR but not the No FR condition (Fig. 1). It is possible that the progressive dehydration during the No FR condition could have caused a further increase in oxidative stress (i.e., increased superoxide levels) relative to the FR condition (20, 30). In the presence of superoxide dismutase, highly elevated levels of superoxide facilitate the conversion of superoxide to H₂O₂, which can act as a vasodilator and thereby increase cutaneous blood flow (41), counteracting the superoxide-mediated reduction in cutaneous vasodilation. Therefore, under these conditions ascorbate would have a minimal effect on CVC as it nonselectively reduces levels of both superoxide and H₂O₂, each pathway yielding the opposite action on cutaneous vascular regulation.

During recovery from exercise in the FR condition, CVC at the ascorbate-treated site returned to levels similar to Control (Fig. 1). This observation may indicate that once exercise was halted, ROS production declined and the body’s endogenous antioxidants were able to scavenge the ROS, restoring NO bioavailability and thereby NO-dependent cutaneous vasodilation. In fact, it has been shown that endogenous serum antioxidant capacity is raised following prolonged (~87 min) running (4). Moreover, it is believed that postexercise suppression of cutaneous vasodilation is largely mediated centrally through nonthermal factors (e.g., baroreceptors, etc.) (28) and is in part due to cutaneous vasoconstriction associated with adenosine and α-adrenergic mechanisms (35, 36). Therefore, perhaps vasoconstriction induced by such mechanisms may have an overriding influence on the ascorbate-mediated vasodilation during recovery from exercise in the heat.

Losartan did not influence cutaneous vasodilation during either the No FR or FR conditions (Fig. 1). This observation parallels the findings by Fujii et al. (12) wherein no role for ANG II administration was found during exercise in the heat. The authors postulate exercise-induced increases in NO bioavailability may be an overriding influence of ANG II-mediated vasoconstriction. Our observations contrasted with findings from Stewart and colleagues where local AT₁R inhibition diminished an exogenous ANG II-induced reduction in CVC during a local heating protocol (57–59). Importantly, AT₁R inhibition was evaluated in response to the administration of exogenous ANG II, which does not necessarily mimic physiological concentrations of circulating ANG II. Moreover, whole body heat stress, such as exercise in the heat, has been shown to affect NO-mediated cutaneous vascular regulation differently from local heating (27, 64). Finally, while FR was shown to influence cutaneous vasodilation with ascorbate supplementation, it is possible that endogenous levels of ANG II were below those previous used for ANG II administration. In addition, during prolonged exercise oxidative stress may arise from sources not directly linked to ANG II such as exercise-induced increases in oxidative metabolism in the mitochondria (62). Therefore, in our study ANG II alone may not have produced a dramatic increase in oxidative stress such that a further decrease in oxidative stress during FR relative to the No FR condition would not have played a role in ANG II-mediated regulation of thermoeffector function.

After prolonged exercise in the heat, we show that AT₁R inhibition did not influence CVC (Fig. 1). Previously, ANG II administration was shown to cause a reduction in cutaneous vasodilation after exercise in the heat (12); however, the administration of exogenous ANG II may not be an exact representation of true physiological concentrations as mentioned above. In addition, body core temperatures achieved in our study surpassed (~1.5°C greater) those in the study by Fujii et al. (9) and, as such, the greater requirement to dissipate heat (and therefore muscarinic receptor activation) in our study may have overridden ANG II-mediated vasoconstriction through increased NO bioavailability as mentioned above. Future research is warranted to determine at what elevation in body core and skin temperatures ANG II-mediated pathways regulate cutaneous vasodilation during postexercise recovery periods.

Sweating

In contrast to the comparatively well-established modulators of NO-dependent cutaneous vasodilation during whole body heat stress, the mechanisms underpinning the involvement of NO in the sweating response to exercise are less understood. Contrary to our hypothesis, we did not observe a role for l-NAME in the regulation of local forearm sweating during or after prolonged exercise in the heat (Fig. 2). Although increased NO bioavailability has been shown to be required for the full expression of local forearm sweating, its involvement has also been shown to be exercise intensity dependent (10). Specifically, this NOS inhibition has been shown to diminish local sweating during low exercise intensity (400 W metabolic heat load) but not high exercise intensity (700 W metabolic heat load) (10). Therefore, given that our participants exercised at a fixed rate of metabolic heat load of 600 W (calculated as metabolic rate minus external workload) and in conjunction with an increased ambient temperature relative to the study conducted by Fujii et al. (10) (i.e., room temperature in their study was 35°C relative to 40°C in the present study), the cumulative (metabolic + environmental) heat stress experienced in the current study may have caused sufficient thermal drive such that the skin treatment-specific changes in sweating could not be observed. Similarly, Welch et al. (63) demonstrated a role for NOS inhibition in the regulation of sweating with participants performing moderate intensity exercise at a fixed percentage of their predetermined V̇o_2peak and therefore metabolic heat production similar to that of our participants. However, the additional environmental thermal strain experienced by our participants and corresponding greater skin temperature (~5.5°C increase) may suggest that the total heat load (environmental + metabolic), rather than simply the metabolic heat load, influences the role of NO on local sweat rate.

Sweat production is induced via muscarinic stimulation of the sweat gland by acetylcholine released from cholinergic nerves (51). While NO alone can induce cutaneous vasodilation (10, 22, 24–26, 35–37), the current view is that NO plays a synergistic role to muscarinic sweating such that it acts to augment cholinergic sweating but cannot activate sweat production directly. In support of this conclusion, we have previously observed no contribution of NO to intradermal administration of relatively high doses of methacholine, an acetylcholine mimetic (1–2,000 mM) (9). In the context of whole body heat stress, these findings suggest that NO may not contribute to the sweating response at high levels of thermal drive and therefore cholinergic stimulation. Similarly, we did not observe a role for ascorbate administration during or after prolonged exercise in the heat. Given that ROS are produced during situations of high thermal drive, this would support the conclusion that ROS do not mediate NO-dependent sweating during prolonged exercise in the heat.

During rest in the heat (40°C) as well as during and after prolonged exercise in the heat we found no role for AT₁R inhibition irrespective of fluid replacement (Fig. 2). While ANG II administration reduced sweating at rest in a hot environment (35°C), both before and after moderate intensity exercise (12), exogenous administration of an agent does not necessarily reflect true physiological conditions as previously mentioned. Moreover, levels of hyperthermia at the end of exercise were substantially greater in our study relative to Fujii et al. (39.0°C vs. 37.5°C body core temperature, respectively). Therefore, the increased thermal drive (and therefore muscarinic receptor activation), to dissipate heat via sweating may have overridden ANG II-mediated attenuations in sweating.

Perspectives and Significance

Much of the previous research investigating the mechanisms underlying the regulation of heat loss responses did not incorporate compensation for the natural fluid loss due to sweating during exercise in the heat. In fact, previous work has demonstrated that vascular regulation is in part dependent on hydration status (8, 44, 49). As the role of ascorbate differed between trial conditions, our results could further indicate that fluid replacement may be necessary to delineate the mechanisms regulating cutaneous blood flow and sweating during prolonged exercise in the heat. Importantly, participants in our study lost 3.6% body weight during the No FR condition thereby indicating hypohydration levels sufficient to impair heat loss responses (48). While there were no differences seen in cutaneous vascular conductance or sweating at the Control site between FR and No FR conditions despite differences in esophageal temperature, this suggests that the thermoeffector response invoked by a specific increase in esophageal temperature changed according to the experimental condition. Moreover, changes at the forearm level do not necessarily reflect potential differences over the whole body.

In addition, posttrial USG was significantly greater in the No FR trials (1.019) relative to the FR trials (1.006; P < 0.01). Recent work has shown that even moderate hypohydration (~2% body mass loss) requires greater postsynaptic (cholinergic) output acting through the endothelium to elicit the same degree of cutaneous vasodilation as the euhydrated condition (61). Altogether, our results suggest that both hydration status and fluid replacement are important factors to consider when interpreting and comparing the results between different protocols.

The participants of the current study were highly active and aerobically fit individuals. As such, they may not be truly representative of the general population. It is important to note that the current participants exercised at a fixed rate of metabolic heat load of 600 W; therefore, individuals with lower aerobic fitness levels may have exhibited exaggerated oxidative stress responses to exercise at this fixed rate of metabolic heat load since it would correspond to an increased relative exercise intensity. Moreover, if participants of lower fitness were in fact able to complete the exercise protocol, they would have likely had reductions in overall fluid loss due to less efficient sweating and therefore have increased body core temperatures compared with the current group of participants. Therefore, fluid replacement may not have had a dramatic an effect on the regulation of cutaneous vasodilation in individuals of lower fitness.

The results of the current study have important physiological and clinical implications as augmented levels of oxidative stress occur during prolonged or high-intensity exercise but are also inherent to advanced aging. Given that older adults have been shown to have reduced heat loss responses relative to their young counterparts (28), hypohydration may further exacerbate these deficits, ultimately putting this population at a greater risk of heat-related injury during thermal challenges. As such future research is warranted to determine how exercise-induced fluid loss via sweating and subsequent fluid replacement may modulate heat loss responses in older adults.

To accurately determine total fluid loss during the experimental session, all participants first completed the No FR session. Then, the fluid loss measured during this trial (via change in weight) was used to determine the amount of water to be provided during the FR trial. Although performing the trials in this order enabled us to directly assess the amount of water required to maintain hydration and therefore ensure participants were not exercising under a dehydrated state, this procedure required the trials to be performed in a nonrandomized order. Given that there was a minimum of 72 h between the two trials we minimized any carryover effect from the first No FR session. In line with this, resting body core temperature, which is typically elevated in dehydrated states, was found to be not different between the two trials.

We showed that NOS modulates cutaneous vasodilation during prolonged moderate intensity exercise in the heat irrespective of progressive fluid replacement while ascorbate-sensitive ROS impair cutaneous vasodilation (exercise duration of >60 min) only with fluid replacement. In contrast, no role for AT₁R inhibition was found for the regulation of heat loss responses with or without fluid replacement.

GRANTS

This study was supported by the Natural Sciences and Engineering Research Council of Canada (Discover Grant, RGPIN-06313–2014 and Discovery Grants Program-Accelerator Supplement, RGPAS-462252–2014; funds held by G. P. Kenny). B. D. McNeely was supported by the Queen Elizabeth II Graduate Scholarship in Science and Technology. G. P. Kenny is supported by a University of Ottawa Research Chair Award. R. D. Meade is supported by a Natural Sciences and Engineering Research Council of Canada Alexander Graham Bell graduate scholarships (CGS-D). N. Fujii was supported by the Human and Environmental Physiology Research Unit.

DISCLOSURES

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

AUTHOR CONTRIBUTIONS

B.D.M., R.D.M., N.F., A.J.S., R.J.S., and G.P.K. conceived and designed research; B.D.M. and R.D.M. performed experiments; B.D.M. analyzed data; B.D.M., R.D.M., N.F., A.J.S., R.J.S., and G.P.K. interpreted results of experiments; B.D.M. prepared figures; B.D.M. drafted manuscript; B.D.M., R.D.M., N.F., A.J.S., R.J.S., and G.P.K. edited and revised manuscript; B.D.M., R.D.M., N.F., A.J.S., R.J.S., and G.P.K. approved final version of manuscript.