Local arginase inhibition does not modulate cutaneous vasodilation or sweating in young and older men during exercise

Abstract

Age-related impairments in cutaneous vascular conductance (CVC) and sweat rate (SR) during exercise may result from increased arginase activity, which can attenuate endogenous nitric oxide (NO) production. We therefore evaluated whether arginase inhibition modulates these heat-loss responses in young (n = 9, 23 ± 3 yr) and older (n = 9, 66 ± 6 yr) men during two 30-min bouts of moderate-intensity cycling (Ex1 and Ex2) in the heat (35°C). CVC and SR were measured at forearm skin sites perfused with 1) lactated Ringer’s (control), 2) N^G-nitro-_L-arginine methyl ester (L-NAME; NO synthase-inhibited), or 3) N^ω-hydroxy-nor-arginine and S-(2-boronoethyl)-l-cysteine (Nor-NOHA + BEC; arginase-inhibited). In both groups, CVC was reduced at L-NAME relative to control and Nor-NOHA + BEC (both P < 0.01). Likewise, SR was attenuated with L-NAME compared with control and Nor-NOHA + BEC during each exercise bout in the young men (all P ≤ 0.05); however, no influence of treatment on SR in the older men was observed (P = 0.14). Based on these findings, we then evaluated responses in 7 older men (64 ± 7 yr) during passively induced elevations in esophageal temperature (∆T_es) equal to those in Ex1 (0.6°C) and Ex2 (0.8°C). L-NAME reduced CVC by 18 ± 20% CVC_max at a ∆T_es of 0.8°C (P = 0.03) compared with control, whereas Nor-NOHA + BEC augmented CVC by 20 ± 18% CVC_max, on average, throughout heating (both P ≤ 0.03). SR was not influenced by either treatment (P = 0.80) Thus, arginase inhibition does not modulate CVC or SR during exercise in the heat but, consistent with previous findings, does augment CVC in older men during passive heating.

NEW & NOTEWORTHY In the current study, we demonstrate that local arginase inhibition does not influence forearm cutaneous vasodilatory and sweating responses in young or older men during exercise-heat stress. Consistent with previous findings, however, we observed augmented cutaneous blood flow with arginase inhibition during whole-body passive heat stress. Thus, arginase differentially affects cutaneous vasodilation depending on the mode of heat stress but does not influence sweating during exercise or passive heating.

INTRODUCTION

Aging is associated with a decline in cutaneous vasodilatory (20–23, 26, 46, 47) and sweating (24, 25, 30–33, 46) responses to both exercise- and passively induced heat stress, with attenuated whole-body heat dissipation during exercise evident in adults as young as 40 yr of age (30). Contributing to these age-related impairments are alterations in factors affecting heat loss at the level of the end organ (i.e., cutaneous vasculature and eccrine sweat gland). For instance, nitric oxide (NO) is required for full expression of cutaneous vasodilation and sweating during heat stress (exercise or passive heating) in the young (9, 12, 20–23, 37–39, 47, 52). In older adults however, its contribution to these heat-loss responses is diminished (12, 13, 20–23, 31, 33, 47, 48).

Potentially underpinning the attenuated NO-dependent cutaneous vasodilation and sweating responses in older adults is elevated arginase activity in the skin. In comparison with NO synthase (NOS), arginase preferentially binds l-arginine, the common precursor for both enzymes. Thus, an age-related increase in arginase activity may reciprocally inhibit endogenous NO production because of substrate depletion (5, 53). During whole-body passive heating in older adults, local arginase inhibition has been shown to augment cutaneous vasodilation to a level comparable to their younger counterparts (22, 23, 47). Additionally, studies of eccrine sweat urea nitrogen (a primary end-product of arginase) suggest that arginase is also localized in the sweat gland (18), and increased urea in the sweat of older versus younger adults (2) may indicate upregulated sweat gland arginase activity with aging. That said, the role of arginase in mediating cutaneous vasodilation and sweating during exercise is currently unknown.

Restoring endogenous NO synthesis through arginase inhibition might alter thermoregulatory responses during exercise-heat stress in older adults. Therefore, the purpose of this study was to assess the influences of local arginase inhibition on cutaneous vasodilation and sweating in young and older men during exercise in the heat (35°C). It was hypothesized that arginase inhibition would augment cutaneous vasodilation and sweating in older but not younger men.

MATERIALS AND METHODS

Ethical Approval

This study was in accordance with the Declaration of Helsinki and was approved by the University of Ottawa Health Sciences and Science Research Ethics Board. Prior to their participation in the study, informed consent was obtained from all volunteers.

Participants

A total of 9 young and 16 older men participated in the study (see results for physical characteristics). Of these individuals, all of the young and nine of the older men completed an exercise-heat stress test, and the remaining seven older men participated in a follow-up whole-body passive heating secondary study. All participants were healthy (i.e., no history of cardiovascular, metabolic, or respiratory disease and not currently taking medication related to these conditions) and habitually active (i.e., performing ≥30 min of structured physical activity a minimum of 2 times per week) as determined by a standardized questionnaire (29). Men were chosen given the potential modulation in the mechanisms underpinning heat loss associated with fluctuations in female sex hormone levels (i.e., estrogen and progesterone) (6, 7). Furthermore, women exhibit an altered capacity to dissipate heat (for a given level of heat stress), independent of sex-related differences in body morphology and aerobic capacity, in comparison with men (14–16). Whereas in young adults sex-related differences in thermoregulatory function typically occur at higher heat loads, work from our laboratory suggests that aging differentially influences the heat-loss responses between sexes (32, 49). We are currently exploring the interaction between age and sex on thermoregulatory function. However, at the inception of the current study, we chose to test men to avoid the potentially confounding influence of sex on thermoregulatory responses.

Experimental Design

Each participant completed one preliminary and one experimental session (exercise-induced or passively induced heat stress). Prior to the start of each session, participants were instructed to refrain from alcohol, caffeine, and/or strenuous physical activity for a minimum of 24 h. Furthermore, participants were instructed to drink 500 ml of water the night before as well as ~2 h before each session to ensure adequate hydration. During the preliminary session, body height, mass, and density were measured. For those participating in the exercise-heat stress protocol, peak rate of oxygen uptake (V̇O_2peak) was also determined. Body height and mass were measured with a stadiometer (Detecto, model 2391, Webb City, MO) and a digital high-performance weighing terminal (model CBU150X, Mettler Toledo Inc., Mississauga, Canada), respectively. Body surface area was calculated from the measurements of body height and mass (8). Resting systolic and diastolic pressures were measured by manual auscultation using a validated mercury column sphygmomanometer (Baumanometer Standby Model, WA Baum Co, Copiague, NY). Body density was measured using the hydrostatic weighing technique and was used to calculate body fat percent (45). V̇o_2peak was assessed using an automated indirect calorimetry system (MCD Medgraphics Ultima Series; MGC Diagnostics, MN) during a progressive incremental cycling protocol on a semirecumbent cycle ergometer (Corival, Lode B.V., Groningen, the Netherlands). The protocol began with 1 min of cycling at a starting workload of 100 W for the younger adults and 60 W for the older adults. Following the first minute, the workload was increased by 20 W/min until volitional fatigue and/or the participant could no longer maintain a pedaling cadence >50 revolutions/min.

As mentioned above, the experimental session consisted of either an exercise-induced heat stress or whole-body heating at rest. In both protocols, participants provided a urine sample for the measurement of urine specific gravity (Reichert TS 400 total solids refractometer, Reichert Inc., Depew, NY) upon arrival to the laboratory, before a baseline measurement of body mass. Participants with a urine specific gravity of <1.020 were deemed adequately hydrated to begin the experimental session (3). Participants exceeding this threshold (n = 4) were given an additional 500 ml of tap water (~2 h before the start of exercise). Thereafter, 3 microdialysis fibers (Bioanalytical Systems Inc., West Lafayette, IN) were inserted in the dermal layer of the dorsal forearm skin while participants rested in an upright, seated position in a thermoneutral (~24°C) experimental room. Fibers were inserted under aseptic conditions by advancing a 25-gage needle 20–25 mm. The needle was withdrawn after the fiber was passed through the lumen, leaving a 10-mm dialysis membrane within the dermal layer. Each fiber was secured to the skin using surgical tape and separated by ~2–4 cm. All fibers were then perfused with lactated Ringer’s solution for ~30 min at a rate of 4 μl/min via a microperfusion pump (model 400, CMA, Microdialysis, Solna, Sweden). Each skin site was instrumented for the measurement of local cutaneous blood flow and sweat rate.

Exercise-heat stress.

In the exercise-heat stress protocol, participants entered a thermally controlled chamber regulated to 35°C and 20% relative humidity following microdialysis fiber insertion. Participants were seated on a semirecumbent cycle ergometer while the microdialysis fibers were perfused at a rate of 4 μl/min via a microinfusion pump (Model 400, CMA Microdialysis) with either 1) lactated Ringer’s (control), 2) 10 mM N^G-nitro-_L-arginine methyl ester (L-NAME; Sigma-Aldrich, St. Louis, MO) to inhibit NOS activity, or 3) 5 mM N^ω-hydroxy-nor-arginine (Nor-NOHA; Sigma-Aldrich) + 5 mM S-(2-boronoethyl)-l-cysteine (BEC; Sigma-Aldrich) arginase inhibitors to enhance endogenous NO synthesis (Nor-NOHA + BEC). Perfusion of each fiber was maintained for 60 min before baseline data collection, as well as throughout the subsequent exercise-heat stress protocol (described below).The concentration of each drug was based on previous work (9, 12, 13, 20–23, 38, 39, 48). Importantly, a period of ~90 min was allowed between the insertion of the microdialysis fibers and start of baseline data collection (30-min instrumentation + 60-min drug perfusion), which is sufficient for resolution of the local inflammatory response associated with fiber insertion (4).

Following the ~60-min pretrial perfusion period, 10 min of baseline data collection ensued, followed by 2 successive 30-min bouts of semirecumbent cycling at a metabolic heat production of 400 W. Exercise was performed at a fixed rate of heat production to ensure similar drives for whole-body heat loss between groups (27, 40), and this workload was chosen because age-related reductions in NO-dependent CVC and sweating have been observed at this heat load (9, 10). Each exercise bout was followed by 20 min of recovery. An intermittent exercise protocol was employed, given the potential influence of multiple exercise bouts on the mechanisms underpinning the heat-loss responses. Briefly, recovery from exercise is associated with an attenuation of cutaneous vasodilation and sweating to near-baseline levels, followed by a rapid increase in these responses upon initiation of a subsequent exercise bout (27). Although this response is primarily associated with nonthermal factors (i.e., baroreceptor unloading), it is unknown whether the mechanisms governing heat loss at the level of the end organ are affected by factors associated with the performance of multiple exercise bouts (i.e., graded hyperthermia). Following the second recovery, each microdialysis fiber was perfused with 50 mM sodium nitroprusside at a rate of 6 μl/min to evaluate maximum cutaneous blood flow. A measurement of blood pressure was taken once a stable plateau in cutaneous blood flow was achieved (~20–30 min) for the determination of maximal cutaneous vascular conductance (CVC_max). The remaining instrumentation was then removed, and a final nude body mass was obtained.

Passive heating secondary study.

As discussed in results (see below), we observed no major influence of arginase inhibition on the heat-loss responses in the older men during exercise, findings seemingly at odds with previous work that employed whole-body passive heating (20, 21). For this reason, a whole-body passive heating secondary study was performed following completion of the exercise-heat stress study for the purpose of comparing the effects of arginase inhibition on heat-loss responses between active (exercise) and passively induced heat stress. In that protocol, participants remained in the thermoneutral room (~24°C) in the semirecumbent position following insertion of the 3 microdialysis fibers while wearing a tube-lined water-perfused suit. The fibers were then perfused for ~60 min with the same agents as in the exercise-heat stress protocol (i.e., Ringer, L-NAME, and Nor-NOHA + BEC) while suit temperature was maintained at 34°C. Therefore, as with the exercise-heat stress protocol, ~90 min elapsed between insertion of the microdialysis fibers and the start of data collection. The temperature of the water perfusing the suit was then increased to 49.5°C to initiate whole-body heating. The perfusion of each microdialysis fiber was maintained throughout passive heating. Once esophageal temperature had increased by 0.8°C (equivalent to the peak increase in esophageal temperature during exercise-heat stress), the suit water temperature was reduced to 40°C, and CVC_max was determined as described in the preceding section.

Measurements

Esophageal temperature was measured in both protocols with a pediatric thermocouple probe ~2 mm in diameter (Mon-a-therm, Mallinckrodt Medical, St. Louis, MO) inserted through the nose and advanced 40 cm. In one older participant (exercise-heat stress), tympanic temperature was taken because of intolerance of the esophageal probe. Mean skin temperature was calculated using 4 skin sites weighted to the regional proportions: 30% chest, 30% bicep, 20% quad, and 20% calf (44), measured during the exercise protocol. Because of technical difficulties, mean skin temperature was not reported for one young participant. Temperature data were collected with a data acquisition module at a 15-s sampling rate and were displayed and recorded with LabVIEW software (version 7.0; National Instruments, TX).

During both exercise-heat stress and whole-body passive heating, systolic and diastolic pressures were measured by manual auscultation every 5 min using a validated mercury column sphygmomanometer (Baumanometer Standby Model, WA Baum Co). Mean arterial pressure was calculated as diastolic pressure plus one-third the difference between systolic and diastolic pressure (i.e., pulse pressure). Heart rate data were recorded and stored every 5 s during exercise using a Polar Coded WearLink and transmitter, Polar RS400 interface, and Polar Trainer 5 software (Polar Electro, Oy, Kempele, Finland).

Cutaneous red blood cell flux, an index of cutaneous blood flow expressed in perfusion units, was locally measured at a sampling rate of 32 Hz via laser Doppler flowmetry (PeriFlux System 5000, Perimed, Stockholm, Sweden) using integrated laser Doppler flowmetry probes with a 7-laser array (Model 413, Perimed). Each probe was placed over the center of each microdialysis fiber and housed within a specialized, ventilated sweat capsule (see below). CVC was calculated as cutaneous red blood cell flux divided by mean arterial pressure (perfusion units/mm Hg) and presented as a percentage of CVC_max.

Specialized ventilated capsules, designed to cover the entire area of skin perfused by the microdialysis fiber (1.1 cm²), were used for the measurement of local sweat rate (41). Each capsule was placed directly over the center of each fiber membrane and affixed to the skin using a double-sided adhesive and topical skin glue (Collodion HV, Mavidon Medical Products, Lake Worth, FL). Dry compressed air was passed through each capsule at a constant flow rate (0.4 l/min) while capacitance hygrometry (Model HMT333, Vaisala, Helsinki, Finland) was used to measure the water content of the effluent air. Connections between the gas tanks and sweat capsules, and between the sweat capsules and hygrometers, comprised long vinyl tubes to allow for gas temperature to equilibrate with ambient temperature. Local sweat rate was calculated every 5 s using the water content of the effluent air times the flow rate and normalized to capsule surface area (mg·min⁻¹·cm⁻²).

During exercise, endogenous heat production was estimated as metabolic rate minus external work (27). The oxygen and carbon dioxide concentrations of expired air were analyzed using electrochemical gas analyzers (AMETEK model S-3A/1 and CD3A, Applied Electrochemistry, Pittsburg, PA), which were calibrated using a gas mixture of known concentrations. A 3-liter syringe was used to calibrate the turbine ventilometer. Participants wore a full-face mask (Model 7600 V2, Hans-Rudolf, Kansas City, MO) attached to a 2-way T-shape nonrebreathing valve (Model 2700, Hans-Rudolf). Oxygen uptake and respiratory exchange ratio were obtained as 30-s averages and subsequently used to calculate metabolic rate (42).

Prior to each experimental protocol, urine specific gravity (an index of body fluid status) was evaluated using a handheld total solids refractometer (Model TS400, Reichter Inc., Depew, NY).

Data Analysis

For the exercise-heat stress protocol, local forearm CVC and sweat rate at each treatment site as well as esophageal and skin temperatures (both absolute values and elevations from baseline) and heart rate were presented as 5-min averages of data recorded at the end of each of the following time periods: baseline, exercise 1, and exercise 2. Blood pressure data were presented as an average of the 2 measurements taken over 10 min at the end of each time period. CVC_max was determined from averaged CVC data over a minimum of 2 min once a plateau was attained during the maximal cutaneous vasodilation procedure at the end of each protocol. In the passive heating protocol, average CVC and sweat rate data were determined over the time interval corresponding to increases in esophageal temperature of 0.6°C (0.55°C–0.64°C) and 0.8°C (0.75°C–0.84°C) (similar to the increase in esophageal temperature noted in the older adults during exercise 1 and 2 of the exercise-heat protocol, respectively).

Statistical Analysis

Local forearm CVC and sweat rate during the exercise-heat stress were analyzed using a three-way mixed model analysis of variance (ANOVA) with the factors of time period (repeated; three levels: baseline, exercise 1, and exercise 2), treatment site (repeated; three levels: control, L-NAME, and Nor-NOHA + BEC), and age group (nonrepeated; two levels: young and older). A two-way mixed model ANOVA was performed with the factors of time (three levels) and age (two levels) for body temperatures (i.e., esophageal and mean skin) and cardiovascular (i.e., mean arterial pressure and heart rate) variables between age groups. Similarly, a two-way mixed model ANOVA with the factors of treatment site (three levels) and age (two levels) was also used to evaluate local forearm CVC_max (expressed in perfusion units/mm Hg) obtained during the maximal cutaneous vasodilation protocol. For the whole-body passive heating protocol, CVC and sweat rate were analyzed with a two-way repeated measures ANOVA with the factors of stage (repeated; two levels: 0.6°C and 0.8°C increase in esophageal temperature) and treatment (three levels). CVC_max was evaluated with a one-way repeated measures ANOVA (factor of treatment site).

For the exercise-heat stress test, post hoc pair-wise comparisons were carried out using two-tailed paired (time, treatment) or unpaired (age group) t-tests. Based on the previous observations of attenuated and augmented CVC with L-NAME and Nor-NOHA + BEC, respectively (22, 23), responses during the passive heating protocol were compared using one-tailed t-tests. The Holm-Bonferroni correction was employed to account for multiple comparisons. Differences in participant physical characteristics, urine specific gravity and the percent change in body mass between age groups in the exercise-heat stress protocol were assessed with Student’s independent sample t-tests (two-tailed). P ≤ 0.05 was set as the level of statistical significance. All values were reported as mean (SD).

RESULTS

Exercise-Heat Stress

Participant physical characteristics.

Table 1 contains the physical characteristics of the young and older participants. By design, age was greater in the older men versus the young men (P < 0.01). No differences were noted for body height (P = 0.69), mass (P = 0.30), surface area (P = 0.40) or resting systolic (P = 0.56) or diastolic (P = 0.40) blood pressures. By contrast, V̇o_2peak was lower (P < 0.01) and body fat percent was greater (P < 0.01) in the older men compared with their younger counterparts.

Table 1.

Participant characteristics for the exercise-heat stress test

	Young	Older
Age, yr	23 (3)	66 (6)*
Body height, m	1.73 (0.06)	1.72 (0.09)
Body mass, kg	75.6 (5.8)	71.4 (10.3)
Body surface area, m2	1.90 (0.09)	1.84 (0.18)
Systolic BP, mm Hg	116 (9)	118 (11)
Diastolic BP, mm Hg	75 (4)	78 (7)
V̇o2peak, ml·min−1·kg−1	45.9 (7.1)	32.1 (2.7)*
Body fat, %	13.6 (4.2)	25.2 (4.5)*

n = no. of participants; young adults n = 9, and older adults n = 9. Presented values are mean (SD). Body fat % was assessed using the hydrostatic weighing technique. BP, blood pressure; V̇o_2peak, rate of peak oxygen uptake as determined during a maximal incremental cycling protocol.

^*Significant difference vs. young; P ≤ 0.05.

Body fluid status.

No differences in presession urine specific gravity were noted between the young (1.013 ± 0.008) and older (1.016 ± 0.006; P = 0.41) men. Furthermore, the change in body mass over the experimental protocol was similar between age groups [young: −1.61 (SD 0.39)% vs. older: −1.50 (SD 0.24)%; P = 0.49].

CVC.

During the exercise-heat stress, no time × treatment × age interaction was observed for CVC (P = 0.81; Fig. 1). There was, however, a time × age interaction (P < 0.01), such that CVC was elevated during both exercise bouts compared with baseline in each age group (all P < 0.01) but was not different between exercise bouts (both P ≥ 0.14). No differences were noted between age groups at any time point (all P ≥ 0.10). Additionally, there was an effect of treatment site (P < 0.01), whereby CVC was reduced at L-NAME compared with control (P < 0.01) and Nor-NOHA + BEC (P < 0.01) but was not different between the latter two sites (P = 0.55). Finally, CVC_max induced via 50 mM sodium nitroprusside administration following the intermittent exercise protocol was similar between and within the age groups at all sites (effects of treatment and age, both P ≥ 0.77; Table 2).

Fig. 1.

Forearm cutaneous vascular conductance (CVC; %CVC_max) combined for the young (n = 9) and older (n = 9) men during 2 successive exercise bouts performed at a heat production of 400 W. Three skin sites were continuously perfused via intradermal microdialysis with 1) lactated Ringer’s (control), or drugs to inhibit; 2) nitric oxide synthase (L-NAME); or 3) arginase (Nor-NOHA + BEC). Values are mean (SD) and represent an average of the final 5 min of each time period. *Significant difference vs. baseline in young and older men (collapsed across treatment sites), †L-NAME significant difference vs. control (collapsed across time-points), and ‡L-NAME significant difference vs. Nor-NOHA + BEC (collapsed across time-points). P ≤ 0.05. BEC, S-(2-boronoethyl)-l-cysteine; L-NAME, N^G-nitro-_L-arginine methyl ester; Nor-NOHA, N^ω-hydroxy-nor-arginine.

Table 2.

CVC_max during the exercise-heat stress

CVCmax, perfusion units/mm Hg	Young	Older
Control	1.85 (0.63)	1.86 (0.79)
L-NAME	1.89 (0.69)	1.79 (0.63)
Nor-NOHA + BEC	1.69 (0.56)	1.98 (0.65)

n = no. of participants. Presented values are mean (SD). CVC_max (perfusion units/mm Hg) measured in young (n = 9) and older (n = 9) men during administration of 50 mM sodium nitroprusside at 3 skin sites previously perfused with 1) lactated Ringer’s (control), 2) 10 mM NG-nitro-l-arginine methyl ester (L-NAME), or 3) 5 mM Nω-hydroxy-nor-Arginine + 5 mM S-(2-boronoethyl)-l-cysteine (Nor-NOHA + BEC) during the preceding intermittent exercise protocol. No statistical differences were observed. CVC_max, maximum cutaneous vascular conductance.

Sweat rate.

A time × treatment × age interaction was observed for the local sweating response (P = 0.05). In the young men (time × treatment interaction: P < 0.01; Fig. 2A), local sweat rate was elevated from baseline values at all treatment sites during both exercise 1 (all P < 0.01) and exercise 2 (all P < 0.01) but was not different between bouts (all P ≥ 0.07). Furthermore, whereas no between-treatment differences were observed at baseline (all P ≥ 0.89), local sweat rate was reduced at L-NAME compared with control (both P ≤ 0.05) and Nor-NOHA + BEC (both P ≤ 0.03) during both exercise bouts.

Fig. 2.

Forearm sweat rate (mg·min⁻¹·cm⁻²) in young (n = 9; A) and older (n = 9; B) men during two successive exercise bouts performed at a heat production of 400 W. Three skin sites were continuously perfused via intradermal microdialysis with 1) lactated Ringer’s (control), or drugs to inhibit; 2) nitric oxide synthase (L-NAME); or 3) arginase (Nor-NOHA + BEC). Values are mean (SD) and represent an average of the final 5 min of each time period. *Significant difference vs. baseline (comparisons are for individual treatment sites in the young but collapsed across sites in the older men), †significant difference vs. control, and ‡significant difference vs. Nor-NOHA + BEC. P ≤ 0.05. BEC, S-(2-boronoethyl)-l-cysteine; L-NAME, N^G-nitro-_L-arginine methyl ester; Nor-NOHA, N^ω-hydroxy-nor-arginine.

In the older men (effect of time: P < 0.01; Fig. 2B), local sweat rate (averaged across treatment sites) was elevated during each exercise compared with baseline (both P < 0.01) but was not different between bouts (P = 0.38). However, in contrast to the young, no influence of treatment was observed (P = 0.14). Finally, there were no differences in sweat rate between age groups at any time point during the exercise-heat stress (effects of age group: all P ≥ 0.28).

Body temperature and cardiovascular responses.

Esophageal temperature (time × age interaction: P < 0.01; Table 3) was elevated from baseline values in the young and older men during exercise (all P < 0.01) and increased from the first to second bout (both P ≤ 0.01). No differences between age groups were noted (all P ≥ 0.25). When expressed as a change from baseline, however, elevations in esophageal temperature (time × age interaction: P < 0.01; Table 3) were greater in the second compared with the first exercise in both age groups (both P = 0.01) and greater in the older versus young men during both exercise bouts (both P ≤ 0.01).

Table 3.

Esophageal responses during the exercise-heat stress test

	Baseline	Exercise 1	Exercise 2
Esophageal temperature, °C
Young	37.12 (0.22)	37.55 (0.22)*	37.67 (0.22)*†
Older	37.07 (0.11)	37.70 (0.21)*	37.86 (0.20)*†
ΔEsophageal temperature, °C
Young		0.43 (0.04)	0.54 (0.08)†
Older		0.63 (0.11)‡	0.78 (0.10)†‡

n = no. of participants. Presented values are mean (SD). Young (n = 9) and older (n = 9) men performed an intermittent exercise protocol consisting of 2 exercises at a fixed rate of heat production of 400 W (each followed by a 20-min recovery). Absolute esophageal change from baseline (Δ) represents an average of the final 5 min of the corresponding time period.

^*Significant difference vs. baseline.

^†Significant difference in exercise 1 vs. exercise 2.

^‡Significant difference vs. young; P ≤ 0.05.

An effect of time (P < 0.01) but not age group (P = 0.48) was noted for mean skin temperature, which, when compared with baseline [across-group average: 34.87 (SD 0.20)°C] was elevated during the first [35.51 (SD 0.15)°C; P < 0.01] and second [35.58 (SD 0.17)°C; P < 0.01] exercise bouts. Similarly, when expressed as a change from baseline (effect of time: P < 0.01), skin temperature was elevated during both exercise bouts [exercise 1: 0.64 (SD 0.14)°C; exercise 2: 0.70 (SD 0.16)°C; both P < 0.01], and not different between age groups (effect of age group: P = 0.98).

A time × age interaction was found for mean arterial pressure (P = 0.02). In the young men, mean arterial pressure was not different from baseline [89 (SD 5) mmHg] in either exercise bout [exercise 1: 92 (SD 5) mmHg; exercise 2: 91 (SD 5) mmHg; both P ≥ 0.23] and was also not different between bouts (P = 0.63). By contrast, in the older men, mean arterial pressure was elevated from baseline [92 (SD 7) mmHg] during both exercise bouts [exercise 1: 101 (SD 10) mmHg; exercise 2: 98 (SD 10) mmHg; both P < 0.01] and was also elevated in the first relative to the second bout (P = 0.03). That said, no differences between age groups were observed at any time point (all P ≥ 0.10). Finally, heart rate (effect of time, P = 0.01) increased from baseline [across condition average: 72 (SD 7) beats/min] during the first [113 (SD 9) beats/min; P < 0.01] and second [117 (SD 10) beats/min; P < 0.01] exercise bouts and was also elevated in the latter compared with the former (P < 0.01). No influence of age was noted (P = 0.97).

Whole-Body Passive Heating Secondary Study

The characteristics of the older men who participated in the whole-body passive heating protocol were: age, 64 (SD 7) yr; height, 1.75 (SD 0.09) m; body mass, 77.8 (SD 10.4) kg; surface area, 1.93 (SD 0.13) m²; systolic pressure, 127 (SD 13) mmHg; diastolic pressure, 73 (SD 6) mmHg; and body fat, 22.3 (SD 8.1)%. Participants were euhydrated [urine specific gravity <1.020 (3)] before the heating protocol [urine specific gravity: 1.012 (SD 0.009)].

At all treatment sites, CVC (temperature × treatment interaction, P = 0.01; Fig. 3A) was elevated when esophageal temperature was passively increased from 0.6°C to 0.8°C (all P ≤ 0.02). In comparison with control, CVC at the L-NAME skin site was similar when esophageal temperature was elevated to 0.6°C (P = 0.09), but it was reduced at an increase of 0.8°C (P = 0.04). Arginase inhibition via Nor-NOHA + BEC augmented CVC relative to control (both P ≤ 0.03) and L-NAME (both P ≤ 0.01) at both levels of hyperthermia. CVC_max was 2.01 (SD 0.67), 1.79 (SD 0.65), and 1.91 (SD 0.74) perfusion units/mmHg at control, L-NAME, and Nor-NOHA + BEC, respectively, and was similar between treatments (effect of treatment site: P = 0.80). Finally, whereas sweat rate at each skin site increased when the change in esophageal temperature from baseline was elevated from 0.6°C to 0.8°C (P < 0.01; effect of temperature: P < 0.01; Fig. 3B), responses were not affected by any treatment (effect of treatment site: P = 0.42).

Fig. 3.

Forearm cutaneous vascular conductance (CVC, %CVC_max; A) and sweat rate (mg·min⁻¹·cm⁻²; B) in a group of older men (n = 7) during passive elevations in esophageal temperature (ΔT_es) of 0.6°C and 0.8°C. Three skin sites were continuously perfused via intradermal microdialysis with 1) lactated Ringer’s (control), or drugs to inhibit, 2) nitric oxide synthase (L-NAME), or 3) arginase (Nor-NOHA + BEC). Values are mean (SD). *Significant difference versus ∆T_es of 0.6, †significant difference vs. control, and ‡significant difference vs. Nor-NOHA + BEC. P ≤ 0.05. Nor-NOHA + BEC. P ≤ 0.05. BEC, S-(2-boronoethyl)-l-cysteine; L-NAME, N^G-nitro-_L-arginine methyl ester; Nor-NOHA, N^ω-hydroxy-nor-arginine.

DISCUSSION

In the present study, the influence of age-related increases in arginase activity on cutaneous vasodilation and sweating during intermittent exercise in the heat (35°C) was evaluated. Contrary to our hypothesis, local arginase inhibition to increase synthesis of endogenous NO via NOS did not independently modulate cutaneous vasodilation or sweating in either young or older men performing moderate-intensity intermittent exercise in the heat (35°C). Based on these findings, we then performed a whole-body passive heating secondary study, in which we observed augmented cutaneous vasodilation (but not sweating) with arginase inhibition at increases in esophageal temperature similar to those achieved during the exercise-heat stress test. This study, therefore, provides novel information regarding the control of cutaneous perfusion and sweating during exercise while also highlighting the differential modulation of heat-loss responses between exercise and passive heat stress.

Cutaneous vascular response.

Our hypothesis that age-related changes in the activity of the arginase enzyme (5, 28, 34, 53) influence heat-loss responses in older men during exercise was not supported by the current findings. Specifically, increasing local synthesis of endogenous NO via arginase inhibition did not augment the cutaneous vasodilatory or sweating responses in either age group. This contrasts with the findings of Holowatz et al. (22, 23) during whole-body passive heating (verified by our secondary study; Fig. 3) in which augmented cutaneous vasodilation was observed in older adults when arginase was locally inhibited. Although we currently have no explanation for the dissimilar roles for arginase between exercise-induced and passively induced heat stress, it is known that control of the heat-loss responses is influenced by the mode of heating. For example, McNamara et al. (37) observed that NO-dependent cutaneous vasodilation was mediated primarily via endothelial NOS (eNOS) during exercise, whereas the neuronal isoform mediated this response during whole-body passive heating. Furthermore, Fujii et al. (9) demonstrated that cyclooxygenase contributes to the cutaneous vasodilatory response during exercise in the heat (in an interactive manner with NO), whereas McCord et al. (36) did not observe a role for this enzyme during whole-body passive heating.

A potential contributor to the discrepancies in the influence of arginase on cutaneous vasodilation may be differences in skin temperature between heating modalities. Exposure to a hot environment (35°C) in the current study resulted in unclamped mean skin temperatures of ~35.0°C–35.5°C, whereas the temperature of each treatment site in the study by Holowatz et al. (22, 23) was maintained at ~33°C. McNamara et al. (37) observed that NO produced via eNOS contributed to cutaneous vasodilation during exercise in a temperate environment (~24°C) once core (ΔT_es of 0.8°C) and presumably skin temperatures (not reported) were elevated. By contrast, Fujii et al. (12) recently demonstrated a similar magnitude of eNOS-mediated cutaneous vasodilation before and throughout an exercise-heat stress (~25% CVC_max) protocol identical to that of the current study. Given that esophageal temperature in the latter study was not appreciably elevated (~37.2°C) during pre-exercise rest, it would appear that skin temperature is the primary mediator of eNOS activity. In fact, the contribution of eNOS to cutaneous vasodilation was similar between end-exercise in the study by McNamara et al. (37) and throughout intermittent exercise in the report by Fujii et al. (12) (~20%–30% CVC_max), despite a 0.7°C increase in body core temperature from baseline to the end of the second exercise in the latter report.

Although the activation of eNOS secondary to the increase in skin temperature appears to be the primary mediator of NO-dependent cutaneous vasodilation associated with exercise and/or exposure to a hot environment (12), it is the neuronal isoform that mediates this response during whole-body passive heating via a water-perfused suit (37). Thus, a given change in core and/or skin temperatures may differentially influence NO-dependent cutaneous vasodilation between heating modes, especially given in vitro work suggesting that the direct influence of temperature on each isoform of the enzyme (i.e., endothelial, neuronal) is inconsistent (50). These factors combined with the differential localization of NOS isoforms in the skin (51) may also contribute to the greater influence of arginase inhibition on cutaneous vascular relaxation during passive heating compared with exercise, although experimental support for this notion is required. Finally, exercise increases cutaneous vessel shear stress, which can influence both NOS (35) and arginase (43) activation, further complicating comparisons between heating modes. Future research should be directed at evaluating the factors associated with exercise (i.e., skin temperature, shear stress) that not only influence NO-dependent heat loss but also its interaction with the age-related changes in arginase activity.

Sweating response.

The contribution of NO to the sweating response during exercise is blunted by aging (12, 13, 48). Based on previous suggestions of urea (a primary end-product of arginase) production in the eccrine sweat gland (18), along with elevated urea nitrogen in the sweat of older compared with younger adults (2), we surmised that age-related increases in arginase activity may reciprocally inhibit NO production via NOS. This was not the case, however, as we observed no differences in sweat rate between the control and arginase-inhibited skin sites in either age group during exercise (Fig. 2) or in the older adults who underwent passive heating (Fig. 3). A potential explanation for these findings may be that the elevated urea in sweat (18), especially in older adults (2), may simply represent differences in the relative size of the urea pool sequestered within the epidermis (17).

In addition to elevated arginase, age-related increases in oxidative stress have been shown to impair NO-dependent cutaneous vasodilation during passive heating (22). That said, we recently demonstrated that local administration of the antioxidant ascorbate did not influence the contribution of NO to the sweating response in older adults (11). Those findings, in combination with the current data, raise the possibility that age-related impairments in sweating stem from alterations in pathways downstream from NO. In support of this notion, we have demonstrated that cyclooxygenase contributes to the sweating response in young adults during moderate exercise in an NO-interactive manner (9). Similarly, recent work suggests that in younger adults, purinergic receptor activation is also an important mediator of sweating during exercise, again via NO-dependent mechanisms (1). However, in parallel to age-related reductions in NO-dependent sweating, neither cyclooxygenase (13) nor purinergic receptor (1) blockade modulate exercise-induced sweating in older adults. Clearly, more work is required to elucidate age-related factors affecting both sweating and cutaneous vasodilation at end-organ level.

Considerations.

Cutaneous vasodilation and sweating at the control skin site were similar between the young and older men throughout the exercise-heat stress protocol. Although age-related impairments in heat exchange are well established (27, 30, 31), they are not always reflected in local measurements (31, 49). Differences in heat-loss responses between young and older adults can be influenced by the site of measurement (46), and a more pronounced separation between groups may have been seen if measurements were performed on the chest or back, skin sites that typically exhibit greater sudomotor activity than the forearm (27). The forearm was chosen to reduce movement artifacts during cycling and to facilitate comparison between previous studies, given that the majority of work aimed at evaluating the mechanisms underpinning heat loss previously employed forearm skin sites (9, 12, 13, 19, 20, 22, 23, 38, 39, 41, 47, 48, 52). That said, we observed greater elevations in esophageal temperature during exercise in the older versus younger men (Table 3), despite having both groups exercise at the same absolute heat load (400 W) to elicit similar drives for whole-body heat loss between age groups (27, 40). Thus, whole-body heat loss was attenuated for a given body core temperature in the older men. Our findings are therefore still indicative of age-related modulation in the control of heat loss during exercise.

Conclusions.

Our findings suggest that age-related elevations in arginase activity do not influence cutaneous vasodilation or sweating during moderate-intensity intermittent exercise in the heat. By contrast, cutaneous vasodilation was augmented by local arginase inhibition in older men during passive heating. In all, these outcomes highlight the differential mechanisms underpinning the heat-loss responses to both exercise-induced and passively induced heat stress. Future work is required to delineate the putative modulators of cutaneous perfusion and sweating with both heating modalities along with their implications for body temperature regulation.

GRANTS

This project was supported by the Canadian Institutes of Health Research (Grant no. 286363; funds held by G. P. Kenny). G. P. Kenny was supported by a University of Ottawa Research Chair Award. R. D. Meade was supported by a Natural Sciences and Engineering Research Council of Canada Alexander Graham Bell graduate scholarships (CGS-D). G. W. McGarr was supported by the Human and Environmental Physiology Research Unit. R. J. Sigal was supported by a Health Senior Scholar award from Alberta Innovates-Health Solutions.

DISCLOSURES

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

AUTHOR CONTRIBUTIONS

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