Mast cell degranulation and de novo histamine formation contribute to sustained postexercise vasodilation in humans

Blood flow to previously active skeletal muscle remains elevated following an acute bout of aerobic exercise and is dependent on activation of histamine H₁ and H₂ receptors. The intramuscular source of histamine that drives this response to exercise has not been identified. Using intramuscular microdialysis in exercising humans, we show both mast cell degranulation and formation of histamine by histidine decarboxylase contributes to the histamine-mediated vasodilation that occurs following a bout of aerobic exercise.

Abstract

In humans, acute aerobic exercise elicits a sustained postexercise vasodilation within previously active skeletal muscle. This response is dependent on activation of histamine H₁ and H₂ receptors, but the source of intramuscular histamine remains unclear. We tested the hypothesis that interstitial histamine in skeletal muscle would be increased with exercise and would be dependent on de novo formation via the inducible enzyme histidine decarboxylase and/or mast cell degranulation. Subjects performed 1 h of unilateral dynamic knee-extension exercise or sham (seated rest). We measured the interstitial histamine concentration and local blood flow (ethanol washout) via skeletal muscle microdialysis of the vastus lateralis. In some probes, we infused either α-fluoromethylhistidine hydrochloride (α-FMH), a potent inhibitor of histidine decarboxylase, or histamine H₁/H₂-receptor blockers. We also measured interstitial tryptase concentrations, a biomarker of mast cell degranulation. Compared with preexercise, histamine was increased after exercise by a change (Δ) of 4.2 ± 1.8 ng/ml (P < 0.05), but not when α-FMH was administered (Δ−0.3 ± 1.3 ng/ml, P = 0.9). Likewise, local blood flow after exercise was reduced to preexercise levels by both α-FMH and H₁/H₂ blockade. In addition, tryptase was elevated during exercise by Δ6.8 ± 1.1 ng/ml (P < 0.05). Taken together, these data suggest that interstitial histamine in skeletal muscle increases with exercise and results from both de novo formation and mast cell degranulation. This suggests that exercise produces an anaphylactoid signal, which affects recovery, and may influence skeletal muscle blood flow during exercise.

NEW & NOTEWORTHY Blood flow to previously active skeletal muscle remains elevated following an acute bout of aerobic exercise and is dependent on activation of histamine H₁ and H₂ receptors. The intramuscular source of histamine that drives this response to exercise has not been identified. Using intramuscular microdialysis in exercising humans, we show both mast cell degranulation and formation of histamine by histidine decarboxylase contributes to the histamine-mediated vasodilation that occurs following a bout of aerobic exercise.

in humans, acute aerobic exercise elicits a sustained vasodilation within previously active skeletal muscle (12, 13, 27). This sustained postexercise vasodilation occurs following both whole body exercise (14, 15, 38) and isolated small muscle-mass exercise (2). In humans, both central neural mechanisms and local vascular mechanisms contribute to sustained postexercise vasodilation following large muscle mass exercise, where combined histamine H₁- and H₂-receptor blockade reduces but does not eliminate sustained postexercise vasodilation (28, 31, 32). Conversely, following unilateral dynamic knee-extension exercise, neurovascular control of the skeletal muscle circulation is unchanged (3), and combined histamine H₁- and H₂-receptor blockade abolishes sustained postexercise vasodilation (2, 41). Thus the response is dependent on histaminergic signaling and occurs independent of central neural mechanisms following isolated small muscle-mass exercise.

The reduction in blood flow observed with histamine H₁/H₂-receptor blockade illustrates that activation of histamine receptors is critical to the genesis of the sustained postexercise vasodilation. Presumably, histamine is the ligand activating histamine receptors during the recovery from exercise. However, studies in our laboratory and others have failed to document any rise in circulating plasma or whole blood histamine concentrations during recovery from exercise, which elicited postexercise vasodilation (17, 28, 31, 32, 34), but elevations in circulating histamine concentrations have been found following exercise in a few studies (5, 7, 16). Conflicting reports may be due to differences in subject selection, exercise intensity, or choice of histamine assay (10). However, a likely explanation is that histamine is released and acting locally in response to exercise, and, owing to its short half-life in the circulation (23), circulating levels do not reflect local concentrations within skeletal muscle tissue. Along these lines, Barrett-O’Keefe et al. (2) found that, following unilateral dynamic knee-extension exercise, sustained postexercise vasodilation is only present in the previously active leg, and not in the contralateral unexercised leg. Furthermore, they showed that histamine blockade only affected blood flow in the exercised leg.

To date, histamine concentrations have not been measured within skeletal muscle tissue during or following acute aerobic exercise. Likewise, the source of histamine responsible for sustained postexercise vasodilation is unknown. There are two likely sources of histamine within skeletal muscle. First, there are histamine-containing mast cells located within skeletal muscle tissue, adjacent to blood vessels and nerves and also within connective tissue (33, 52). Degranulation of mast cells releases histamine (along with the enzyme tryptase, which is a biomarker of mast cell degranulation) (22). Second, there are a number of non-mast cells that can potentially produce histamine de novo without storing it, via the inducible enzyme histidine decarboxylase, which catalyzes the formation of histamine by decarboxylation of the amino acid l-histidine (35, 49). Histidine decarboxylase activity is transcriptionally regulated and could increase in response to acute aerobic exercise. Romero et al. (42) recently showed that 1-h aerobic exercise in humans is sufficient to upregulate mRNA for histidine decarboxylase, which has previously been observed in rodent hindlimbs following more prolonged (>3 h) exercise (1, 9, 37, 52).

Thus the aims of this study were threefold: 1) measure interstitial histamine concentrations in skeletal muscle during and after exercise; 2) assess the involvement of de novo histamine production by histidine decarboxylase through administration of the irreversible inhibitor α-fluoromethylhistidine; and 3) assess the involvement of mast cell degranulation by measuring the mast cell biomarker tryptase in skeletal muscle during and after exercise. We hypothesized that histamine and tryptase concentrations would increase in response to 1-h unilateral dynamic knee-extension exercise. We also hypothesized that inhibition of histidine decarboxylase would reduce histamine concentrations and would blunt local blood flow during recovery from exercise, perhaps to the same extent as local histamine H₁/H₂-receptor blockade. These hypotheses were tested using intramuscular microdialysis in humans.

METHODS

Subjects

This study was approved by the Institutional Review Board at the University of Oregon and was performed in accordance with the principles outlined by the Declaration of Helsinki. Written, informed consent was obtained from all subjects subsequent to a verbal and written briefing of all experimental procedures. Subjects were deemed healthy following a standard health history questionnaire. All subjects were required to abstain from caffeine, alcohol, and exercise for 24 h before studies. Additionally, subjects reported to the laboratory 2 h postprandial. No subjects were using over-the-counter or prescription medications at the time of study, with the exception of oral contraceptives. Female participants were studied during the early follicular phase of their menstrual cycle or during the placebo phase of their oral contraceptive.

Preliminary Study

We performed several trials in a small number of young healthy subjects that were intended to provide methodological “proof of principle” for the primary work described herein. The goal of this preliminary microdialysis work was twofold: 1) demonstrate that mast cells in skeletal muscle were capable of degranulation, and that this response would induce measurable changes of interstitial histamine and the mast cell biomarker tryptase (n = 3); and 2) demonstrate that interstitial histamine is not elevated in nonexercising skeletal muscle (n = 7).

In the first preliminary study, we infused (5 µl/min) 10 mM of the mast cell degranulator compound 48/80 (Sigma-Aldrich, St. Louis, MO) for 45 min via skeletal muscle microdialysis. Compound 48/80 induces mast cell degranulation in a noncytotoxic manner and independent of antigen-immunoglobulin E coupling (33). Microdialysis probes were inserted (custom probes, 3-kDa molecular mass cutoff with a 30-mm regenerated cellulose membrane) into the biceps brachii. Relative to baseline, the infusion of compound 48/80 increased interstitial histamine and tryptase concentrations (Fig. 1). We note that intramuscular degranulation of mast cells did not generate any symptoms or sensations, which is in sharp contrast to the typical intradermal response: wheal and flare and accompanying pruritus.

Fig. 1.

Effect of mast cell degranulation on dialysate histamine and tryptase concentrations ([Histamine]_Dialysate and [Tryptase]_Dialysate, respectively) collected via intramuscular microdialysis in the biceps brachii. Samples were obtained before degranulation (control; open bars) and after administration of compound 48/80, a mast cell degranulator (compound 48/80; solid bars). Samples represent absolute histamine and tryptase concentrations in the dialysate and were not corrected for probe recovery. Values are means ± SE.

In the second preliminary study, we measured interstitial histamine in the biceps brachii via skeletal muscle microdialysis in subjects before, during, and following 60-min leg cycle exercise. Microdialysis probes (custom probes, 3-kDa molecular mass cutoff with a 30-mm regenerated cellulose membrane) were inserted in the rested arm. Saline was infused (5 µl/min) through the microdialysis probes, and the dialysate was collected before, during, and following 60-min leg cycle exercise. Interstitial histamine concentrations within the biceps brachii remained unchanged throughout the protocol (Fig. 2).

Fig. 2.

Effect of exercise on dialysate histamine concentrations ([Histamine]_Dialysate) collected via intramuscular microdialysis in the biceps brachii. Samples were obtained before, during, and after moderate-intensity leg cycle exercise with the arm resting. Samples represent absolute histamine concentration in the dialysate and were not corrected for probe recovery. Values are means ± SE.

These data demonstrate that, by using skeletal muscle microdialysis, interstitial histamine concentrations within skeletal muscle are measurable. Likewise, if skeletal muscle mast cells degranulate, interstitial tryptase concentrations rise. Prior studies have established the utility of the microdialysis-based ethanol washout technique to measure vasodilatory responses to exogenous histamine, and the ability of locally administered histamine H₁/H₂-receptor blockade to blunt this response (39). Together, these data support the methodological approaches used in the present study to determine whether interstitial histamine is increased in active or previously active skeletal muscle, and to determine the source(s) of histamine in this response.

Primary Study

Screening visit.

A total of 29 subjects participated in these studies (16 men and 13 women). A screening visit was used to determine peak power output during a unilateral dynamic knee-extension exercise test performed to volitional fatigue. Dynamic knee-extension exercise during all visits was performed using a custom-built knee-extension ergometer based on a computer-controlled step-motor that provided resistance against the subject’s lower leg. Using real-time measures of angular velocity and torque, power was calculated, and a feedback loop maintained measured power at the assigned level. Subjects were seated with their back at 60° upright, and knee-extension exercise was performed with the right leg over a 45° range of motion, starting with the leg hanging at ~90° of flexion. Subjects were asked to maintain a cadence of 45 kicks/min, while being provided with visual feedback of both kicking cadence and range of motion. Workload was ramped incrementally at a rate of 3 W/min.

Experimental approach.

Following the screening visit, subjects were randomly assigned to sham (n = 13) or exercise (n = 16) experiments. Subjects in the sham experiments sat upright for 60 min. Subjects in the exercise experiments performed 60 min of unilateral dynamic knee-extension exercise at 60% of peak power and a cadence of 45 kicks/min. Power was ramped at the onset of exercise to 60% peak power over the first 15 min. Power output was recorded continuously throughout 60-min dynamic knee-extension exercise. For all subjects, skeletal muscle microdialysis probes were implanted in the right leg 2.5 h before upright rest or exercise and used to sample interstitial fluid from the vastus lateralis. Microdialysis effluent (dialysate) was collected every 30 min before, during, and following upright rest or exercise. The number of functional microdialysis probes varied across subjects. Thus, in some subjects, we measured interstitial tryptase concentrations (9 sham, 10 exercise). In some subjects, we measured interstitial histamine concentrations with probes that were 1) “control” (did not receive any drugs: 4 sham, 6 exercise); 2) histamine H₁/H₂-receptor blockade (4 sham, 6 exercise); and 3) histidine decarboxylase inhibition (4 sham, 6 exercise). As interstitial histamine concentrations from control probes and histamine H₁/H₂-receptor blockade probes did not differ, they were pooled and referred to collectively as control when comparing histamine concentrations against histidine decarboxylase inhibition. For all subjects, hemodynamic measurements were made before exercise and at 30, 60, and 90 min postexercise.

Skeletal muscle microdialysis.

All probes were inserted using sterile technique. The skin and underlying fascia were anesthetized using 1% lidocaine HCl (Hospira Worldwide, Lake Forest, IL) buffered with sodium bicarbonate (Hospira Worldwide). Care was taken to ensure that the lidocaine was not injected into the skeletal muscle. Probes were inserted in the vastus lateralis in a direction parallel with muscle fiber orientation (~19°, relative to long-axis of muscle) using a splitable introducer. After insertion, probes were held in place by covering the entry site with a sterile transparent medical dressing. Probes that were used to measure interstitial histamine and local blood flow had a 20-kDa molecular mass cutoff with a 30-mm polyarylethersulphone membrane (63 MD Catheter, MDialysis, Stockholm, Sweden). The molecular mass cutoff of the semipermeable membrane used in this probe allows histamine to freely diffuse from the interstitial space, but not the histamine metabolizing enzymes diamine oxidase (91 kDa) and histamine N-methyltransferase (30 kDa). Probes that were used to measure interstitial tryptase, which is a gold-standard biomarker of mast cell degranulation (22), had a 100-kDa molecular mass cutoff with a 30-mm polyarylethersulphone membrane (71 High Cut-off Brain MD Catheter, MDialysis, Stockholm, Sweden). Subsequent to insertion, microdialysis probes were perfused continuously at a rate of 5 µl/min (CMA 400 Microdialysis pump, CMA, North Chelmsford, MA). Probes used to measure histamine were perfused with a 0.9% saline solution that also contained 0.40 µCi/ml of tritiated [³H]histamine and 5 mM ethanol. Tritiated histamine was used for in vivo calibration of probe recovery (44), which averaged ~24.5 ± 0.8%, and ethanol was used to assess local blood flow via the ethanol washout technique. Probes used to measure tryptase were perfused with a 0.9% saline solution.

Several drugs were added to the perfusate of some probes. In combination in the same probe, pyrilamine maleate (1 mM) was used to block H₁ histamine receptors, and cimetidine (3 mM) was used to block H₂ histamine receptors (Sigma-Aldrich, St. Louis, MO). These doses have been used previously in our laboratory to locally block histamine H₁/H₂ receptors and reduce sustained postexercise vasodilation (39). In a separate probe, α-fluoromethylhistidine dihydrochloride (α-FMH, 200 µM) (Santa Cruz Biotechnology, Dallas, TX) was used to irreversibly inhibit histidine decarboxylase activity (50). This dose was extrapolated from studies using α-FMH to inhibit histidine decarboxylase in cell culture (45, 51).

Dialysate was collected every 30 min throughout the study. Microtubes were covered with nonporous tape during each sampling period to prevent dialysate and ethanol evaporation. Microtube weight was documented before and after dialysate collection to assess fluid loss and estimate perfusion rate of the probe. Dialysate was stored at −20°C for the duration of the study. Thereafter, the samples were stored at −80°C until analysis.

Interstitial histamine was measured using an enzyme-linked immunosorbent assay (Rocky Mountain Diagnostics, Colorado Springs, CO) and performed in accordance with the manufacturer’s instructions. Interstitial tryptase was measured using an enzyme-linked immunosorbent assay and performed in accordance with the manufacturer’s instructions adapted for microdialysis (Kamiya Biomedical, Seattle, WA). In addition, 10 µl of perfusate and dialysate were pipetted into 5-ml polypropylene scintillation vials containing 3 ml of scintillation cocktail (Ultima Gold, Shelton, CT) immediately after the sampling period. Radioactivity was then measured (counts/min) in duplicate using a liquid scintillation counter (Beckman LS 6000SC, Brea, CA) and were used to determine probe recovery.

Local blood flow responses were assessed by measuring ethanol washout (ethanol outflow-to-inflow ratio) from the interstitial space. Ethanol in the perfusate and dialysate was measured using a modified alcohol dehydrogenase enzymatic assay (19, 20, 40), as previously performed in our laboratory (39). The assay uses alcohol dehydrogenase to catalyze the oxidation of ethanol using nicotinamide adenine dinucleotide as the electron acceptor.

Hemodynamic measurements.

All resting measurements were made pre- and postexercise with the subjects in the supine position. Subjects were asked to remain quiet and relaxed during all hemodynamic measurements. Room temperature remained thermoneutral (~23°C) throughout the study. Arterial blood pressure was measured in the right arm using an automated sphygmomanometer (Tango+, SunTech Medical, Raleigh, NC). Heart rate was monitored using a three-lead electrocardiograph (Tango+, SunTech Medical). Heart rate and blood pressure were also measured during 60-min dynamic knee-extension exercise. Femoral artery blood flow velocity and diameter were measured via duplex ultrasonography. A linear-array vascular ultrasound probe (10 MHz, GE Vingmed System 5, Horton, Norway) and an insonation angle of 60° were used to measure blood flow in the common femoral artery, ~2–3 cm proximal to the bifurcation. Femoral artery diameter was measured in triplicate during end diastole using digital ultrasound calipers. The Doppler ultrasound was interfaced with a computer running custom audio recording software. Velocity measurements were determined using an intensity-weighted algorithm (custom software), subsequent to demodulation of forward and reverse Doppler frequencies. Velocity measurements were made at an average depth of 1.67 ± 0.03 cm for the sham condition and an average depth of 1.61 ± 0.03 cm for the exercise condition. Velocities were thin-beam corrected and based on a known beam width of 2.21 mm, which resulted in an average correction factor of 0.764 ± 0.003 for the sham condition and 0.767 ± 0.002 for the exercise condition, as outlined recently (4). Leg blood flow was calculated as cross-sectional area multiplied by femoral mean blood velocity and reported in milliliters per minute. Femoral vascular conductance was calculated by dividing femoral blood flow by mean arterial pressures and expressed as milliliters per minute per millimeters of mercury.

Statistical Analysis

Preliminary statistical analysis indicated that our primary outcome variables did not vary by sex. As such, all subsequent analyses were performed after grouping data for both men and women. Our primary outcome variables during the recovery from exercise were analyzed using a two-way mixed-model analysis of variance with repeated measures (JMP Pro 12; SAS Institute, Cary, NC). Planned comparisons were used to examine specific condition by time interactions. General interactions were examined using Tukey’s post hoc procedure. Significance was set at P < 0.05. Data are reported as means ± SE, unless stated otherwise (e.g., SD is used in Table 1 to document variability in the subject pool).

Table 1.

Subject characteristics

	Sham	Exercise
n	13	16
Age, yr	23.0 ± 4.4	20.9 ± 2.0
Height, cm	176 ± 10	174 ± 5
Weight, kg	73.6 ± 13.5	68.5 ± 6.5
Body mass index, kg/m2	23.5 ± 2.5	22.6 ± 1.9
Baecke sport index, arbitrary units	3.1 ± 0.7	3.2 ± 0.8
Physical activity index, MET·h−1 wk−1	31.7 ± 9.6	47.4 ± 25.4

Values are means ± SD. MET, metabolic equivalents.

RESULTS

Primary Study

Subject characteristics.

Subject physical characteristics and data obtained during the screening visit are shown in Table 1. Subject characteristics are similar to those obtained previously in our laboratory (8, 41) in young healthy subjects and are consistent with recreationally active individuals.

hemodynamics.

Systemic hemodynamics before and following sham or exercise are shown in Table 2. There were no differences between presham and preexercise for heart rate (P = 1.0), mean arterial pressure (P = 0.8), femoral blood flow (P = 1.0), or femoral vascular conductance (P = 1.0).

Table 2.

Systemic hemodynamics

	Time, min
	Baseline	Post
Heart rate, beats/min
Sham	56.9 ± 1.7	58.6 ± 1.7
Exercise	61.8 ± 1.5	65.5 ± 0.8*
Mean arterial pressure, mmHg
Sham	90.4 ± 1.4	87.5 ± 0.9*
Exercise	86.5 ± 1.0	85.9 ± 0.7
Femoral blood flow, ml/min
Sham	226 ± 10	227 ± 7
Exercise	224 ± 11	369 ± 11*
Femoral vascular conductance, ml min−1 mmHg
Sham	2.4 ± 0.1	2.5 ± 0.1
Exercise	2.5 ± 0.1	4.2 ± 0.1*

Values are means ± SE.

^*P < 0.05 vs. baseline.

Subjects maintained a power output of 17.0 ± 0.5 W over the final 45 min of unilateral knee-extension exercise. This power output matched the target 17.4 W, or 60% of peak power output. Compared with sham (68.1 ± 1.5 beats/min), heart rate was higher during exercise (104.1 ± 3.2 beats/min) (P < 0.05). Compared with sham (90.8 ± 2.8 mmHg), mean arterial pressure tended to be higher during exercise (96.9 ± 2.1 mmHg, P = 0.09).

The percent change (Δ) in heart rate from preexercise to 1 h postexercise did not differ between sham (Δ3 ± 1%) and exercise (Δ4 ± 2%, P = 0.7). Likewise, the percent change in mean arterial pressure from preexercise to 1 h postexercise did not differ between sham (Δ−2 ± 1%) and exercise (Δ0 ± 1%, P = 0.2). Relative to preexercise, blood flow at 1 h postexercise did not differ for sham (Δ−2 ± 4%), but was elevated following exercise (Δ36 ± 3%, P < 0.05). Likewise, the change in femoral vascular conductance from preexercise to 1 h postexercise did not differ for sham (Δ−2 ± 4%), but was elevated following exercise (Δ36 ± 3%, P < 0.05).

interstitial histamine concentration.

Muscle interstitial histamine concentrations are shown for sham and exercise in Fig. 3. Histamine concentrations for control (no drug) and histamine H₁/H₂-receptor blockade were grouped together and referred to collectively as “control” for comparison to histidine decarboxylase inhibition (α-FMH). Presham and preexercise histamine concentrations did not differ between control and α-FMH (P = 0.9). Likewise, histamine concentrations did not differ between control and α-FMH during (P = 1.0) or following (P = 0.2) sham. In contrast, histamine concentrations were elevated during exercise by Δ149 ± 27% for control, but only Δ87 ± 18% for α-FMH (both P < 0.05 vs. preexercise; P < 0.05 for α-FMH vs. control). During recovery from exercise, the histamine concentration remained elevated by Δ37 ± 8% for control (P < 0.05), but was not elevated for α-FMH (Δ−1 ± 7%; P = 0.8 vs. pre-exercise; P = 0.06 for α-FMH vs. control).

Fig. 3.

Effect of exercise and inhibition of histidine decarboxylase on interstitial histamine concentrations ([Histamine]_interstitial) collected via intramuscular microdialysis in the vastus lateralis. Samples were obtained before, during, and after sham (seated rest; top) and before, during, and after exercise (bottom). Open bars denote control (no drug or combined histamine H₁/H₂-receptor blockade, which does not affect histamine concentrations); solid bars denote inhibition of histidine decarboxylase with α-FMH. Samples represent interstitial histamine concentration after correcting for probe recovery. Values are means ± SE. *P < 0.05 vs. preexercise. †P < 0.05 vs. control during exercise and P = 0.06 vs. control during postexercise.

local blood flow responses.

Washout of ethanol from the interstitial space (ethanol outflow-to-inflow ratio) was used as a marker of local perfusion of the tissue. Data are presented as a change relative to presham or preexercise; a larger (more negative) reduction is reflective of a larger rise in local blood flow. Presham and preexercise outflow-to-inflow ratios did not differ between control, H₁/H₂ blockade, and α-FMH (P = 0.3). Changes in the ratio are shown in Fig. 4, expressed as the absolute change from presham or preexercise. Expressed in this way, a larger (more negative) reduction is reflective of a larger rise in local blood flow. For sham, there were no differences in local blood flow from presham to sham or postsham, and no differences between control, H₁/H₂ blockade, and α-FMH (all P > 0.3).

Fig. 4.

Effect of exercise, inhibition of histidine decarboxylase, and combined histamine H₁/H₂-receptor blockade on ethanol washout in the vastus lateralis. A larger (more negative) reduction is reflective of a larger rise in local blood flow. Top: sham. Bottom: exercise. Open bars denote control (no drug); shaded bars denote combined histamine H₁/H₂-receptor blockade; and solid bars denote inhibition of histidine decarboxylase with α-FMH. Values are means ± SE. *P < 0.05 vs. preexercise. †P < 0.05 vs. control.

In contrast, local blood flow was elevated during exercise for control, H₁/H₂ blockade, and α-FMH (i.e., ethanol outflow-to-inflow ratio was reduced; all P < 0.05 vs. preexercise), but the elevated blood flow was less for H₁/H₂ blockade compared with control (P < 0.05). Following exercise, local blood flow remained elevated for control (P < 0.05 vs. preexercise), but not for H₁/H₂ blockade or α-FMH (both P < 0.5 vs. preexercise). During recovery, compared with control, local blood flow was lower for both H₁/H₂ blockade and α-FMH (both P < 0.05 vs. control). Local blood flow during recovery did not differ between H₁/H₂ blockade and α-FMH (P = 0.6).

interstitial tryptase concentration.

Muscle interstitial tryptase concentrations for sham and exercise conditions are shown in Fig. 5. Presham and preexercise tryptase concentrations did not differ (P = 0.9). Tryptase was elevated during exercise (P < 0.05 vs. preexercise), but not sham (P = 0.1 vs. presham). Compared with sham, tryptase was greater during exercise (P < 0.05). Tryptase remained elevated during the recovery from exercise (P < 0.05 vs. preexercise), but not during sham recovery (P = 0.2). Compared with sham, tryptase was greater during the recovery from exercise (P < 0.05).

Fig. 5.

Effect of exercise on dialysate tryptase concentrations ([Tryptase]_Dialysate) collected via intramuscular microdialysis in the vastus lateralis. Samples were obtained before, during, and after sham (seated rest; top) and before, during, and after exercise (bottom). Values are means ± SE. *P < 0.05 vs. preexercise.

DISCUSSION

The purpose of this study was threefold. First, we wanted to determine whether interstitial histamine concentrations in skeletal muscle rise during or after exercise. Second, we wanted to delineate the contribution of de novo histamine formation within active skeletal muscle to any observed changes in histamine concentrations. Third, we wanted to determine whether or not mast cell degranulation occurs in skeletal muscle during or after exercise. In agreement with our hypothesis, we documented that interstitial histamine is elevated in response to 1 h of unilateral dynamic knee-extension exercise. Moreover, blockade of histidine decarboxylase, via the irreversible inhibitor α-fluoromethylhistidine, reduced interstitial histamine concentrations within skeletal muscle and reduced local blood flow. Importantly, this effect on blood flow was similar in magnitude to that observed with local histamine H₁/H₂-receptor blockade. However, it appears that mast cell degranulation is also triggered in response to 1 h of unilateral dynamic knee-extension exercise, as interstitial tryptase was elevated by exercise. Collectively, these observations show that both de novo formation of histamine and mast cell degranulation are involved in sustained postexercise vasodilation. The fact that histidine decarboxylase inhibition has a similar impact on local perfusion as histamine receptor blockade suggests that the primary histaminergic mechanism for vasodilation is due to augmented local histamine formation in skeletal muscle, along with a possible contribution from mast cell activation. In support of these current findings, Romero et al. (42) recently demonstrated that histidine decarboxylase mRNA expression is elevated in human skeletal muscle following exercise, an observation that is consistent with previous work in rodent models after prolonged exercise (1, 9, 52), and suggests that signaling mechanisms related to de novo formation are upregulated by acute exercise.

Taken together, our findings support the idea of parallel histaminergic pathways that cooperatively mediate sustained postexercise vasodilation, as first postulated by Endo and colleagues (1, 9, 37, 52). Their elegant work over the last decade has suggested that the induction of histidine decarboxylase replenishes the pool of mast cell histamine lost with degranulation occurring with exercise. Our study provides evidence that a similar mechanism may occur in humans, but we do not yet know how this synergism is established or maintained. Furthermore, it is possible that histamine formation by histidine decarboxylase is not limited to mast cells, but may take place in other cells within skeletal muscle tissue, such as vascular smooth muscle and endothelial cells (48).

Smoking Gun: Missing Trigger

As presented by the late John T. Shepherd (46) and revisited recently by Joyner and Casey (25), we suggest that a proposed mediator of sustained postexercise vasodilation must meet the following criteria: 1) the substance or its precursor should be present in skeletal muscle; 2) the substance should have access to the skeletal muscle resistance vessels; 3) the concentration in the interstitial fluid (or at the vascular endothelium) must be capable of causing and maintaining vasodilation; 4) the substance should be capable of producing dilatation of arterioles when administered; 5) antagonists should equally impact vasodilation caused by the substance and the physiological vasodilation; and 6) local administration of the substance should not produce any sensation or symptoms that do not normally accompany the sustained vasodilation. Along these lines, we believe we have a “smoking gun” for the cause of sustained postexercise vasodilation in humans. Our laboratory’s previous findings, coupled with those in this present investigation, demonstrate that histamine satisfies each of the above criteria. We have documented resident mast cells in skeletal muscle that corelease histamine and tryptase in response to a mast cell degranulator (compound 48/80). Furthermore, intramuscular histamine, whether endogenously released or exogenously administered via microdialysis, results in notable local vasodilation, which can be blocked by the combination of histamine H₁ and H₂-receptor blockers (39). Importantly, this vasodilation occurs without any subjective sensations of itchiness, consistent with normal exercise responses. Thus we can conclude that histamine is the vasodilator responsible for sustained postexercise vasodilation in humans. Moreover, induction of histidine decarboxylase is a major contributor to local histamine formation within the skeletal muscle. Mast cells also appear to be involved in the response to exercise. However, future work is necessary to determine how these mechanisms interact in active skeletal muscle and to determine in which cells these mechanisms occur. And while we have the smoking gun, we have yet to find a trigger.

What is the exercise-related factor which triggers this response? Several exercise-related factors, such as oxidative stress (47), cytokine release (33), and elevated temperatures (11), have been shown in other contexts to induce mast cell degranulation in an antigen-independent manner. Additionally, oxidative stress (21) and the transcription factor hypoxia-inducible factor-1α (24), both of which are induced by exercise, are associated with upregulated transcription of histidine decarboxylase. Histidine decarboxylase also appears to function optimally at elevated temperatures and reduced pH, two conditions that are well-associated with exercise (43). Lastly, a study by DeForrest et al. (6) suggests that shear stress may increase de novo histamine synthesis within blood vessels through a shear stress-dependent mechanism. To test one of these potential exercise-related mechanisms, Romero et al. (41) infused the potent antioxidant N-acetylcysteine before and during exercise. They determined that sustained postexercise vasodilation was unaffected, suggesting that exercise-induced oxidative stress is not a necessary signal for histidine decarboxylase and/or mast cells activation. To date, the trigger remains elusive.

Experimental Considerations

Skeletal muscle microdialysis is a powerful, yet complex technique that is used to continuously obtain biological samples from the extracellular fluid compartment in vivo. In general, fluid sampling and drug delivery will occur within 1 cm³ of the microdialysis probe (18, 29) and thus reflects the local expression of metabolic and regulatory events. The recovery of substances from the interstitial space is dependent on 1) perfusion rate, 2) probe length, 3) probe membrane characteristics, 4) the rate of diffusion in the interstitial space, and 5) the rate of clearance from the interstitial space by spillover into the circulation, breakdown (metabolism), or cellular uptake (26, 30, 36). The first three factors are usually fixed within an experiment and, therefore, do not affect recovery across a study, but can vary greatly between studies and need to be taken into account when estimating interstitial concentrations from recovered dialysate concentrations. The rate of diffusion is influenced by a number of factors and can vary greatly both between and within studies. The diffusion of substances is dependent on tissue tortuosity, the volume of interstitial space relative to the tissue (i.e., volume fraction), blood flow, and perfusate flow rate (18, 26, 30, 36). Experiments with dynamic conditions, such as exercise or pharmacological interventions, can greatly alter the factors that influence diffusion. Thus, ideally, an in vivo calibration should be performed to account for differences in probe recovery across probes. Because perfusion rates above 0.3 µl/min do not allow for complete equilibration with the interstitial fluid, the internal reference technique (44) was used for in vivo calibration of recovery by adding 0.40 µCi/ml tritiated [³H]histamine to the perfusate. The concentration of tritiated histamine is minimal and does not alter local blood flow or measurement of histamine by ELISA. For measurements where an internal reference was not available (i.e., our preliminary studies and measures of tryptase), we provide the uncorrected values as dialysate concentration, rather than interstitial concentrations, but presume there is a close correlation between dialysate and interstitial concentrations. Finally, evidence exists that suggests that there is an interaction between ethanol and histamine metabolites (53) that may influence our interpretation of the data, given that ethanol is included in our perfusate. Likewise, histamine H₂-receptor antagonists also appear to influence ethanol metabolism (53). However, the ethanol-histamine interaction has largely been described in gastrointestinal regions and cerebral tissue, but not, to our knowledge, in skeletal muscle. Additionally, it is unclear if the low concentrations of ethanol and/or histamine H₂-receptor antagonist in our perfusate can influence histamine formation/metabolism or ethanol metabolism in vivo. Nevertheless, the interstitial histamine concentration and ethanol washout data from the sham group suggest that these interactions are absent, as the perfusate was identical for both groups.

Perspectives

Our laboratory recently demonstrated that histamine H₁/H₂-receptor activation has a profound effect on the human exercise transcriptome (42), highlighting the importance of histamine as an exercise signaling molecule, or molecular transducer of the effects of physical activity. We now extended these observations by identifying key mechanisms by which histamine concentrations rise in skeletal muscle tissue with exercise. Understanding the complete histaminergic pathway in healthy humans, from the release/formation of histamine to the histamine-dependent modulation of gene expression and protein content, may provide insight into how the pathway might be affected in important patient groups, such as those within the cardiometabolic disease spectrum. Ultimately, this could lead to the development of therapies targeting this molecular transducer of physical activity responses.

Summary

The current study, using tryptase as a biomarker of mast cell release of histamine, shows that mast cell degranulation occurs within exercising skeletal muscle tissue in humans. Furthermore, by using a selective inhibitor of histidine decarboxylase, we have also shown an important contribution of de novo histamine formation to the exercise response. We conclude that mast cell degranulation and de novo histamine formation by histidine decarboxylase both contribute to vasodilation following exercise. To our surprise, many of these processes appear to occur during exercise, not just afterwards. This suggests that exercise produces an anaphylactoid signal that may influence skeletal muscle blood flow during exercise, an idea we intend to further pursue.

GRANTS

This research was funded by National Heart, Lung, and Blood Institute Grants HL-115027 and HL-107131, and American Heart Association Grant-in-Aid 555632Z.

DISCLOSURES

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

AUTHOR CONTRIBUTIONS

S.A.R., J.L.M., D.A.M., and J.R.H. conception and design of research; S.A.R., J.L.M., M.R.E., D.C.S., T.M.B., M.J.L., and J.R.H. performed experiments; S.A.R., J.L.M., and M.R.E. analyzed data; S.A.R., J.L.M., M.R.E., D.C.S., M.J.L., D.A.M., and J.R.H. interpreted results of experiments; S.A.R. prepared figures; S.A.R. drafted manuscript; S.A.R., M.R.E., D.C.S., T.M.B., M.J.L., D.A.M., and J.R.H. edited and revised manuscript; S.A.R., J.L.M., M.R.E., D.C.S., T.M.B., M.J.L., D.A.M., and J.R.H. approved final version of manuscript.