Sensory neuron inositol 1,4,5-trisphosphate receptors contribute to chronic mechanoreflex sensitization in rats with simulated peripheral artery disease

Abstract

The mechanoreflex is exaggerated in patients with peripheral artery disease (PAD) and in a rat model of simulated PAD in which a femoral artery is chronically (∼72 h) ligated. We found recently that, in rats with a ligated femoral artery, blockade of thromboxane A₂ (TxA₂) receptors on the sensory endings of thin fiber muscle afferents reduced the pressor response to 1 Hz repetitive/dynamic hindlimb skeletal muscle stretch (a model of mechanoreflex activation isolated from contraction-induced metabolite production). Conversely, we found no effect of TxA₂ receptor blockade in rats with freely perfused femoral arteries. Here, we extended the isolated mechanoreflex findings in “ligated” rats to experiments evoking dynamic hindlimb skeletal muscle contractions. We also investigated the role played by inositol 1,4,5-trisphosphate (IP₃) receptors, receptors associated with intracellular signaling linked to TxA₂ receptors, in the exaggerated response to dynamic mechanoreflex and exercise pressor reflex activation in ligated rats. Injection of the TxA₂ receptor antagonist daltroban into the arterial supply of the hindlimb reduced the pressor response to 1 Hz dynamic contraction in ligated but not “freely perfused” rats. Moreover, injection of the IP₃ receptor antagonist xestospongin C into the arterial supply of the hindlimb reduced the pressor response to 1 Hz dynamic stretch and contraction in ligated but not freely perfused rats. These findings demonstrate that, in rats with a ligated femoral artery, sensory neuron TxA₂ receptor and IP₃ receptor-mediated signaling contributes to a chronic sensitization of the mechanically activated channels associated with the mechanoreflex and the exercise pressor reflex.

INTRODUCTION

Skeletal muscle contraction during exercise stimulates mechanically activated (MA) channels located on the sensory endings of groups III and IV thin fiber skeletal muscle afferents (1–5). The stimulation of these channels activates the mechanoreflex which, along with the metaboreflex, is an important constituent of the exercise pressor reflex (6–8). The exercise pressor reflex plays an integral role in increased sympathetic nervous system activity, heart rate, and blood pressure during exercise (9, 10). The mechanoreflex is crucial for normal blood pressure regulation during exercise in health but becomes augmented during exercise in many forms of chronic disease which contributes to aberrant sympathoexcitation and augmented blood pressure reactivity (11–19). Of particular relevance for this investigation, the mechanoreflex was suggested to contribute to the exaggerated pressor response to rhythmic lower leg exercise in patients with peripheral artery disease (PAD) compared with that found in aged-matched healthy control counterparts (16, 17, 20). Moreover, the mechanoreflex was found to be exaggerated during 1-Hz rhythmic hindlimb muscle contractions in a rat model of simulated PAD in which a femoral artery was ligated ∼72 h before experimentation compared with the mechanoreflex found in rats in which the femoral artery was patent (21). Acute exaggerations in blood pressure reactivity during exercise are an independent risk factor for cardiovascular morbidity and mortality in both healthy and clinical populations (22, 23). Thus, identifying the mechanisms underlying the exaggerated mechanoreflex in PAD carries clinical importance.

Our laboratory has used a 1-Hz repetitive/dynamic rat hindlimb skeletal muscle stretch protocol that is based on original work by Stebbins et al. (24, 25) as an experimental model to investigate MA channel stimulation and mechanoreflex activation isolated from the influence of muscle contraction-induced metabolites (26–29). Using this model, we found that the pressor and renal sympathetic nerve activity (RSNA) response to 30 s of dynamic muscle stretch was larger when evoked from hindlimb muscles associated with a chronically ligated femoral artery compared with responses evoked from the contralateral hindlimb muscles which were freely perfused (27). Moreover, rat femoral artery ligation was found recently to have no effect on L₄ and L₅ dorsal root ganglia (DRG) protein expression of piezo channels (21), a novel class of MA channel that has been suggested to underlie mechanoreflex activation (11, 21, 29–31). Together, these findings suggest that the exaggerated mechanoreflex in rats with a ligated femoral artery is due, at least in part, to a chronic or persistent sensitization of MA channels located on the sensory endings of thin fiber muscle afferents. This chronic sensitization may couple with further sensitization that develops acutely during skeletal muscle contractions as metabolites are produced and accumulate. The concept of a chronic sensitization of MA channels is consistent with a large body of literature demonstrating that G protein-linked second messenger signaling cascades within sensory neurons may potentiate MA channel function, especially in chronic inflammatory conditions (32–35). In further investigation of the mechanisms of chronic mechanoreflex sensitization, we found recently that inhibition of the cyclooxygenase (COX) enzyme (26) and blockade of COX metabolite associated thromboxane A₂ (TxA₂) receptors (28) reduced the pressor response to dynamic hindlimb skeletal muscle stretch in rats with a ligated femoral artery. Conversely, we found no effect of either intervention on the pressor response to muscle stretch in rats with freely perfused hindlimbs (26, 28). These findings were supported by molecular evidence demonstrating that rat femoral artery ligation increased TxA₂ receptor protein expression in L₄ and L₅ DRG tissue (26). Thus, there is experimental support for the conclusion that, in rats with a ligated femoral artery, COX metabolite/TxA₂ receptor signaling increases the responsiveness of MA channels on thin fiber muscle afferents that contribute to mechanoreflex activation.

The specific intracellular signaling mechanisms within sensory neurons that mediate TxA₂ receptor-induced mechanoreflex sensitization have not been investigated. In sensory neurons, TxA₂ receptors are coupled to G_q proteins that, when stimulated, activate phospholipase C (PLC) to induce increases in diacylglycerol and inositol 1,4,5-trisphosphate (IP₃) formation (36–38). Although both of these ubiquitous intracellular signaling molecules have been linked to sensory neuron sensitization to various stimuli (39), a possible role for IP₃ in mechanoreflex sensitization holds indirect experimental support. IP₃ binds to its intracellular receptor on the endoplasmic reticulum which results in the release of stored calcium into the cytosol. Elevated cytosolic calcium concentration has been shown to sensitize MA piezo channel-mediated currents in HEK293 cells (33). Moreover, sensory neuron IP₃ receptors have been found to play an important role in the development of mechanical allodynia in mice (40). Together, these findings raise the possibility that amplified IP₃ receptor signaling may contribute to a chronic sensitization of MA channels and the mechanoreflex in rats with simulated PAD induced by chronic femoral artery ligation.

Based on the information above, the present investigation was undertaken with two primary goals. First, we sought to confirm that the role played by TxA₂ receptors in the pressor response to dynamic hindlimb muscle stretch in rats with a ligated femoral artery (28) extended to a role for TxA₂ receptors in evoking the pressor response to dynamic muscle contraction which produces physiological mechanoreflex activation consequent to muscle shortening and concurrent increases in intramuscular pressure (41). Second, we investigated the role played by sensory neuron IP₃ receptors in the chronic sensitization of MA channels and the exaggerated mechanoreflex and exercise pressor reflex in rats with a ligated femoral artery. We tested the hypotheses that, in decerebrate, unanesthetized rats, 1) injection of the TxA₂ receptor antagonist daltroban (80 µg) into the arterial supply of the hindlimb reduces the pressor response to 1 Hz dynamic hindlimb skeletal muscle contraction to a greater extent in rats with a ligated femoral artery than in sham-operated rats with freely perfused hindlimb muscles, and 2) injection of the IP₃ receptor antagonist xestospongin C (XeC, 5 µg) into the arterial supply of the hindlimb reduces the pressor response to 1 Hz dynamic hindlimb skeletal muscle stretch and contraction in rats with a ligated femoral artery but not in sham-operated rats with freely perfused hindlimb muscles.

METHODS AND MATERIALS

All experimental procedures were approved by the Institutional Animal Care and Use Committee of Kansas State University and were conducted in accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals. Experiments were performed on young adult (∼12–15 wk old) male Sprague-Dawley rats (n = 64, average body weight: 412 ± 5 g; Charles River Laboratories). The rats were housed two per cage in temperature (maintained at 22°C) and light (12-12-h light-dark cycle)-controlled accredited facilities with standard rat chow and water provided ad libitum. At the end of each experiment, the decerebrated rats (see Surgical Procedures for Experimental Protocols) were killed by an intravenous injection of saturated (>3 mg/kg) potassium chloride.

Femoral Artery Ligation/Sham Procedure

Of the 64 rats in this investigation, 49 had their left femoral artery ligated ∼72 h before the terminal experimental protocol was performed. Briefly, rats were anesthetized with 2% isoflurane anesthesia (balance O₂) and their left femoral artery was surgically exposed and ligated tightly with 5-0 silk suture ∼3–5 mm distal to the inguinal ligament. In 15 rats, a sham ligation procedure was performed in which the left femoral artery was surgically exposed, and 5-0 suture was passed under the femoral artery but not tied. Thus, the hindlimb remained freely perfused following the sham procedure. In both “ligated” and “freely perfused” rats (referred to as such from here forward for simplicity), the incisions were closed, meloxicam was administered (1–2 mg/kg sc) as an analgesic, and the rats were monitored daily until the final experiment.

Surgical Procedures for Experimental Protocols

On the day of the experiment, rats were anesthetized with ∼2% isoflurane (balance O₂). Adequate depth of anesthesia was confirmed by the absence of toe-pinch and blink reflexes. The trachea was cannulated, and the lungs were mechanically ventilated (Harvard Apparatus model 683) with the gaseous anesthetic until the decerebration was completed (see below). In all rats, the right jugular vein and both common carotid arteries were cannulated with PE-50 catheters for the administration of fluids/drugs, measurement of arterial blood pressure (Physiological Pressure Transducer, AD Instruments), and sampling of arterial blood gases (ABL80 Flex, Radiometer). Heart rate (HR) was measured by electrocardiogram. The left calcaneus bone was severed and the triceps surae (gastrocnemius, soleus, and plantaris) muscles were exposed by reflecting the overlying skin and skeletal muscles. A string was then tied to the distal Achilles tendon and severed calcaneus that linked the triceps surae muscles to a force transducer (Grass FT03) and rack and pinion that could be turned manually. In the 15 sham rats and 44 of the ligated rats, the left superficial epigastric artery was cannulated with a PE-8 catheter whose tip was located near the junction of the superficial epigastric artery and the femoral artery. In those rats in which a catheter was placed in the left superficial epigastric artery, a reversible snare was placed around the left iliac artery and vein (i.e., proximal to the location of the catheter placed in the superficial epigastric artery). For rats in which dynamic hindlimb skeletal muscle contraction was performed (n = 30), the left sciatic nerve was exposed.

After the initial surgical procedures, all rats were placed in a Kopf stereotaxic frame with clamps placed around the pelvis. Dexamethasone (0.2 mg iv) was injected to minimize brainstem edema. A precollicular decerebration was performed and all neural tissue rostral to the superior colliculus was aspirated. After the decerebration was completed, anesthesia was terminated and the lungs were mechanically ventilated with room air. The decerebration procedure was performed because anesthesia has been shown to depress the exercise pressor reflex in the rat (42). Arterial blood gases and pH were measured periodically with a blood gas analyzer and maintained within normal limits ($\text{P}{\text{a}}_{\text{CO}}{}_{{}_2}$: 35–45 mmHg, $\text{P}{\text{a}}_{\text{O}}{}_{{}_2}$: ∼100 mmHg, pH: 7.35–7.45) by adjusting ventilation and/or administering intravenous sodium bicarbonate (8.5%). Core temperature was measured by a rectal probe and maintained at ∼37°C–38°C by an automated heating system (Harvard Apparatus) and heat lamp. For all rats in which dynamic hindlimb muscle stretch was performed (34 of the 64 rats), the paralytic pancuronium bromide (1 mg/kg iv) was injected before the initiation of any stretch maneuver to prevent any spontaneous or reflex muscle contraction which would produce a metabolic stimulus.

Exercise Pressor Reflex Activation Protocol

The control dynamic hindlimb muscle contraction maneuver was performed at least 60 min following termination of isoflurane anesthesia. To begin, baseline muscle tension was set to ∼100 g and baseline blood pressure and HR were measured for ∼30 s. The sciatic nerve was then electrically stimulated using stainless steel electrodes for 30 s at a voltage of ∼1.5× motor threshold (0.01 ms pulse duration, 500 ms train duration, 40 Hz frequency), which produced 1-Hz repetitive/dynamic contractions of the triceps surae muscles. Approximately 10–15 min following the control contraction maneuver, the snare on the left iliac artery and vein was tightened and the appropriate solution (see protocol details below) was injected into the arterial supply of the hindlimb via the left superficial epigastric artery catheter. After the appropriate time had passed for a given injectate (see below), the left iliac artery and vein snare were released and the hindlimb was allowed to reperfuse before the muscle contraction protocol was repeated as aforementioned. At the end of the experiment, to ensure that the increase in blood pressure and HR during contraction was not due to electrical activation of the axons of the thin fiber afferents in the sciatic nerve, we administered the paralytic pancuronium bromide (1 mg/kg iv) and the sciatic nerve was stimulated for 30 s with the same parameters as those used to elicit contraction. No increase in blood pressure or HR was observed during the stimulation period after the administration of pancuronium bromide in any of the experiments for which data are reported which indicates the increases in blood pressure during contractions were reflex in nature. In addition, at the end of each experiment, Evans blue dye was injected in the same manner as the experimental solution to confirm that the injectate had access to the triceps surae muscle circulation.

The following muscle contraction protocols were performed. The injectate, the time the iliac artery and vein snare was pulled tight, the time the hindlimb was allowed to reperfuse, and the sample sizes are indicated. Protocol 1: Daltroban (TxA₂ receptor antagonist, 80 µg dissolved in 0.4 mL of 1% DMSO), snare tight for 5 min, hindlimb reperfused for 10 min, n = 6 freely perfused rats and n = 10 ligated rats. Protocol 2: 0.4 mL of 1% DMSO (vehicle for daltroban), snare tight for 5 min, hindlimb reperfused for 10 min, n = 5 ligated rats. Protocol 3: XeC (IP₃ receptor antagonist, 5 µg dissolved in 0.2 mL of 0.02% ethanol), snare tight for 10 min, hindlimb reperfused for 30 min, n = 4 freely perfused and n = 5 ligated rats.

Isolated Mechanoreflex Activation Protocol

The control dynamic hindlimb muscle stretch maneuver was performed at least 60 min following termination of isoflurane anesthesia. To begin, baseline muscle tension was set at ∼100 g and baseline MAP and HR were collected for 30 s. An experienced investigator then elicited repetitive/dynamic triceps suare muscle stretch for 30 s by manually turning the rack and pinion at a 1-Hz frequency with the aid of a metronome. The investigator aimed to develop ∼0.6 to 0.8 kg of tension during each dynamic stretch maneuver because that is the tension typically developed during hindlimb muscle contractions in decerebrate rat preparations (21, 27, 30). Moreover, the investigator aimed for consistent levels of tension development for each individual dynamic stretch although slight variability in tension development was often present. The dynamic stretch protocol was adapted from that described by Daniels et al. (24). Approximately 5 min after the control stretch maneuver, the snare on the left iliac artery and vein was tightened and the appropriate solution (see below) was injected into the arterial supply of the hindlimb via the left superficial epigastric artery catheter. After the appropriate time had passed, the left iliac artery and vein snare were released and the hindlimb was allowed to reperfuse before the dynamic stretch protocol was repeated as aforementioned. At the end of each experiment, Evans blue dye was injected in the same manner as the experimental solution to confirm that the injectate had access to the triceps surae muscle circulation.

The following isolated mechanoreflex protocols were performed. The injectate, the time the iliac artery and vein snare was pulled tight, the time the hindlimb was allowed to reperfuse, and the sample sizes are indicated. Protocol 4: XeC (IP₃ receptor antagonist, 5 µg dissolved in 0.2 mL of 0.02% ethanol), snare tight for 10 min, hindlimb reperfused for 30 min, n = 9 freely perfused rats and n = 14 ligated rats. Protocol 5: 0.2 mL of 0.02% ethanol (vehicle for XeC), snare tight for 10 min, hindlimb reperfused for 30 min, n = 6 ligated rats.

In an additional group of five ligated rats, dynamic hindlimb skeletal muscle stretch maneuvers were performed before and after XeC (5 µg) was injected into the right jugular vein and therefore allowed to circulate systemically. Forty minutes elapsed between the intravenous injection of XeC and the subsequent stretch maneuver exactly as aforementioned in protocol 1 when XeC was injected into the arterial supply of the hindlimb.

Lactic Acid Injection Protocol (Protocol 6)

In another group of ligated rats (n = 5), we performed a control experiment to determine if xestospongin had an “off-target” effect that resulted in a generalized reduction in sensory neuron responsiveness. About 0.2 mL of 24 mM lactic acid was injected into the arterial supply of the hindlimb before and after XeC (5 µg) was injected into the arterial supply of the hindlimb exactly as aforementioned in protocol 4. Forty minutes elapsed between the injection of the XeC and the subsequent lactic acid injection maneuver also as aforementioned in protocol 4.

Drugs

Daltroban was dissolved in 100% DMSO and diluted to a final concentration of 80 µg in 0.4 mL of 1% DMSO. Dose and timing of the daltroban protocol were based on previous use of the drug by our laboratory (28, 43) and Leal et al. (44) demonstrated the efficacy of TxA₂ receptor inhibition. The dose of XeC was calculated from in vitro molarity concentrations assuming a hindlimb blood volume of ∼15 mL (45). XeC was dissolved initially in 10% ethanol and diluted to a final concentration of 5 µg in 0.25 mL of 0.02% ethanol. This dose of XeC resulted in an estimated concentration of 746 nM that is sufficient to inhibit IP₃ receptors (46, 47).

Data Analysis

Data were collected with a PowerLab and LabChart data acquisition system (AD Instruments). Arterial blood pressure, electrocardiogram, and muscle tension were measured, mean arterial pressure (MAP) and HR were calculated, and all data were displayed in real time and recorded for offline analysis. Baseline MAP and HR were determined from the 30-s baseline periods that preceded each maneuver. The peak pressor (peak ΔMAP) and cardioaccelerator (peak Δ HR) responses were calculated as the difference between the peak values wherever they occurred during the 30-s maneuvers and their corresponding baseline value. The tension-time indexes (TTIs) and blood pressure indexes (BPIs) were calculated by integration of the area under signal during the stretch or contraction maneuver and subtracting the integrated area under the signal during the corresponding baseline period. The time course of the increase in MAP and HR was plotted as ΔMAP and ΔHR from baseline during the 30-s contraction or stretch maneuvers. Data are expressed as means ± SE. Across group comparisons were made with unpaired Student’s t tests. Effects of daltroban or XeC on peak ΔMAP, BPI, peak ΔHR, and ΔTTI were analyzed within freely perfused and ligated groups with Sidak multiple comparisons tests. The use of Sidak multiple comparisons tests for those analyses was determined a priori. ANOVA analyses were not performed because they would not have enhanced interpretation of those experiments that were focused only on the effect of the intervention within each group. The effects of the pharmacological interventions on the time courses of the MAP and HR responses to muscle contraction and stretch within freely perfused and ligated groups were analyzed with two-way repeated measures (time and condition) ANOVAs and Sidak multiple comparisons tests. Statistical significance was accepted at P ≤ 0.05.

RESULTS

Effect of Femoral Artery Ligation on the Exercise Pressor Reflex and Mechanoreflex

Data from the control condition of the main experimental groups in which either daltroban or XeC was injected into the arterial supply of the hindlimb indicate that the pressor (peak ΔMAP freely perfused: 24 ± 3, ligated: 35 ± 4 mmHg, P = 0.03) but not the cardioaccelerator (peak ΔHR freely perfused: 41 ± 6, ligated: 46 ± 4 beats/min, P = 0.23) response to muscle contraction was larger in ligated rats (n = 15) than in freely perfused (n = 10) rats. Moreover, the pressor (peak ΔMAP freely perfused: 24 ± 3, ligated: 37 ± 4 mmHg P < 0.03) and cardioaccelerator (peak ΔHR freely perfused: 9 ± 1, ligated: 25 ± 4 beats/min, P < 0.01) response to muscle stretch was larger in ligated rats (n = 14) than in freely perfused rats (n = 9).

Effect of the TxA₂ Receptor Blockade with Daltroban on the Exercise Pressor Reflex

In freely perfused rats (n = 6), injection of the TxA₂ receptor antagonist daltroban into the arterial supply of the hindlimb had no effect on the pressor (peak ΔMAP and BPI) or cardioaccelerator (peak ΔHR) response to dynamic hindlimb skeletal muscle contraction (Fig. 1). In ligated rats (n = 10), daltroban significantly reduced the pressor and cardioaccelerator response to contraction. The ΔTTI of the contraction maneuver was not different between control and daltroban conditions in either group (Fig. 1D). The time course of the pressor and cardioaccelerator response to contraction before and after daltroban is shown in Fig. 2. Baseline MAP and HR were not different between conditions in either group (Table 1).

Figure 1.

Effect of thromboxane A₂ (TxA₂) receptor blockade with daltroban (Dal.) on the exercise pressor reflex. The peak Δmean arterial pressure (peak ΔMAP, A), blood pressure index (BPI, B), and peak heart rate (peak ΔHR, C) response to 30 s of 1 Hz dynamic hindlimb skeletal muscle contraction before (Control) and after injection of the TxA₂ receptor antagonist daltroban (80 µg) into the arterial supply of the hindlimb of freely perfused (n = 6) and ligated (n = 10) rats. The tension-time index (ΔTTI, D) was not different between conditions for either group. Data were analyzed with Sidak multiple comparisons tests. Bars represent group mean. *Statistically significant (P ≤ 0.05) differences between conditions.

Figure 2.

Effect of thromboxane A₂ (TxA₂) receptor blockade with daltroban on the time course of the exercise pressor reflex. The Δmean arterial pressure (ΔMAP, A and C) and Δheart rate (ΔHR, B and D) response to 30 s of 1 Hz dynamic hindlimb skeletal muscle contraction before and after injection of TxA₂ receptor antagonist daltroban (80 µg) into the arterial supply of the hindlimb of freely perfused (top, n = 6) and ligated (bottom, n = 10) rats. Data were analyzed with two-way repeated-measures ANOVA and Sidak multiple comparisons tests. *Time points where statistically significant (P ≤ 0.05) differences exist between conditions. Cond., condition effect; Int., interaction.

Table 1.

Baseline mean arterial pressure (MAP) and heart rate (HR)

Experimental Group	Control	Postinjection	P Value
Baseline MAP, mmHg
Freely perfused contraction (TxA2-R antagonist, n = 6)	107 ± 8	108 ± 10	0.68
Ligated contraction (TxA2-R antagonist, n = 10)	98 ± 6	108 ± 6	0.11
Ligated contraction (1% DMSO, n = 5)	99 ± 12	101 ± 10	0.76
Freely perfused stretch (IP3-R antagonist XeC, n = 9)	128 ± 5	121 ± 4	0.26
Ligated stretch (IP3-R antagonist XeC, n = 14)	113 ± 5	119 ± 5	0.21
Ligated stretch (IP3-R antagonist XeC iv, n = 5)	133 ± 7	132 ± 5	0.81
Ligated stretch (0.02% ethanol, n = 6)	109 ± 12	115 ± 13	0.53
Ligated lactic acid inj. (IP3-R antagonist XeC, n = 5)	125 ± 15	118 ± 4	0.58
Freely perfused contraction (IP3-R antagonist XeC, n = 4)	98 ± 13	100 ± 7	0.82
Ligated contraction (IP3-R antagonist XeC, n = 5)	100 ± 9	101 ± 9	0.91
Baseline HR, beats/min
Freely perfused contraction (TxA2-R antagonist, n = 6)	391 ± 19	388 ± 22	0.63
Ligated contraction (TxA2-R antagonist, n = 10)	382 ± 13	412 ± 20	0.10
Ligated contraction (1% DMSO, n = 5)	430 ± 16	426 ± 21	0.68
Freely perfused stretch (IP3-R antagonist XeC, n = 9)	509 ± 13	517 ± 14	0.13
Ligated stretch (IP3-R antagonist XeC, n = 14)	494 ± 13	507 ± 11	0.11
Ligated stretch (IP3-R antagonist XeC iv, n = 5)	523 ± 4	516 ± 5	0.32
Ligated stretch (0.02% ethanol, n = 6)	517 ± 13	520 ± 13	0.85
Ligated lactic acid inj. (IP3-R antagonist XeC, n = 5)	544 ± 8	550 ± 8	0.26
Freely perfused contraction (IP3-R antagonist XeC, n = 4)	344 ± 7	357 ± 13	0.47
Ligated contraction (IP3-R antagonist XeC, n = 5)	398 ± 19	389 ± 16	0.31

Values are means ± SE. The experimental group is identified by the presence of a ligated or patent (freely perfused) femoral artery, the experimental maneuver, and the injectate. Data were analyzed with paired Student’s t tests. IP₃-R, inositol 1,4,5-trisphosphate receptor; TxA₂-R, thromboxane A₂ receptor.

In control experiments in ligated rats (n = 5), 0.4 mL of 1% DMSO in saline (the vehicle for daltroban) had no effect on the peak ΔMAP (control: 22 ± 4, 1% DMSO: 20 ± 2 mmHg, P = 0.51), BPI (control: 410 ± 95, 1% DMSO: 386 ± 57 mmHg·s, P = 0.80), or peak ΔHR (control: 27 ± 5, 1% DMSO: 33 ± 6 beats/min, P = 0.07) response to dynamic contraction. The ΔTTI of the dynamic contraction maneuver was not different between control (9 ± 1 kg·s) and 1% DMSO (8 ± 1 kg·s, P = 0.24) conditions. Baseline MAP and HR were not different between conditions (Table 1).

Effect of IP₃ Receptor Blockade with XeC on Isolated Mechanoreflex and Exercise Pressor Reflex Activation

In sham rats (n = 9), injection of the IP₃ receptor antagonist XeC into the arterial supply of the hindlimb had no effect on the peak pressor response, BPI, or peak cardioaccelerator response to dynamic stretch (Fig. 3). In ligated rats (n = 14), XeC significantly reduced the pressor and cardioaccelerator response to stretch. The ΔTTI of the stretch maneuver was not different between control and XeC conditions in either group (Fig. 3D). The time course of the pressor and cardioaccelerator response to stretch before and after XeC is shown in Fig. 4. Baseline MAP and HR were not different between conditions for either group (Table 1). An example of original recordings showing the pressor and cardioaccelerator response to stretch before and after XeC in a freely perfused and ligated rat is shown in Fig. 5.

Figure 3.

Effect of inositol 1,4,5-trisphosphate (IP₃) receptor blockade with xestospongin C (XeC) on isolated mechanoreflex activation. The peak Δmean arterial pressure (peak ΔMAP, A), blood pressure index (BPI, B), and peak Δheart rate (peak ΔHR, C) response to 30 s of 1 Hz dynamic hindlimb skeletal muscle stretch before (control) and after injection of the IP₃ receptor antagonist XeC (5 µg) into the arterial supply of the hindlimb of freely perfused (n = 9) and ligated (n = 14) rats. The tension-time index (ΔTTI, D) was not different between conditions for either group. Data were analyzed with Sidak multiple comparisons tests. Bars represent group mean. *Statistically significant (P ≤ 0.05) differences between conditions.

Figure 4.

Effect of inositol 1,4,5-trisphosphate (IP₃) receptor blockade with xestospongin C (XeC) on the time course of isolated mechanoreflex activation. The Δmean arterial pressure (ΔMAP, A and C) and Δheart rate (ΔHR, B and D) response to 30 s of 1 Hz dynamic hindlimb skeletal muscle stretch before and after injection of the IP₃ receptor antagonist XeC (5 µg) into the arterial supply of the hindlimb of freely perfused (top, n = 9) and ligated (bottom, n = 14) rats. Data were analyzed with two-way repeated-measures ANOVA and Sidak multiple comparisons tests. *Time points where statistically significant (P ≤ 0.05) differences exist between conditions. Cond., condition effect; Int., interaction.

Figure 5.

Original tracings in a freely perfused rat (A) and a ligated rat (B) of the blood pressure and heart rate response to 30 s of 1 Hz dynamic hindlimb skeletal muscle stretch before (left) and after (right) injection of inositol 1,4,5-trisphosphate (IP₃) receptor antagonist xestospongin C (5 µg) into the arterial supply of the hindlimb.

In vehicle control experiments in ligated rats (n = 6), 0.02% ethanol (the vehicle for XeC) had no effect on the peak ΔMAP (control: 35 ± 7, 0.02% ethanol: 32 ± 5 mmHg, P = 0.49), BPI (control: 568 ± 147, 0.02% ethanol: 534 ± 110 mmHg·s, P = 0.77), or peak ΔHR (control: 15 ± 4, 0.02% ethanol: 12 ± 3 beats/min, P = 0.37) response to stretch. The ΔTTI of the stretch maneuver was not different between control (13 ± 1 kg·s) and 0.02% ethanol (13 ± 1 kg·s, P = 0.26) conditions. Baseline MAP and HR were not different between conditions (Table 1).

In systemic control experiments in ligated rats (n = 5), injection of XeC into the jugular vein to allow it to circulate systemically had no effect on the peak ΔMAP (control: 47 ± 7, iv XeC: 43 ± 8 mmHg, P = 0.49), BPI (control: 585 ± 86, iv XeC: 549 ± 70 mmHg·s, P = 0.70), or peak ΔHR (control: 8 ± 1, iv XeC: 9 ± 1 beats/min, P = 0.52) response to stretch. The ΔTTI of the dynamic stretch maneuver was not different between control (12 ± 1 kg·s) and intravenous XeC (12 ± 1 kg·s, P = 0.37) conditions. The ΔTTI of the stretch maneuver was not different between control and intravenous XeC conditions. Baseline MAP and HR were not different between conditions (Table 1).

In “off target” control experiments in ligated rats (n = 5), injection of XeC into the arterial supply of the hindlimb had no effect on the peak ΔMAP (control: 35 ± 5, XeC: 41 ± 9 mmHg, P = 0.46) or peak ΔHR (control: 4 ± 2, XeC: 7 ± 2 beats/min, P = 0.07) produced in response to the injection of lactic acid into the arterial supply of the hindlimb. Baseline MAP and HR were not different between conditions (Table 1).

In experiments extending our isolated mechanoreflex findings to experiments evoking the exercise pressor reflex, injection of XeC into the arterial supply of the hindlimb of freely perfused rats (n = 4) did not reduce the pressor or cardioaccelerator response to dynamic contraction (Fig. 6). In contrast, in ligated rats (n = 5), XeC significantly reduced the pressor and cardioaccelerator response to contraction. The ΔTTI of the contraction maneuver was not different between control and XeC conditions in either group (Fig. 6D). The time course of the pressor and cardioaccelerator response to contraction before and after XeC is shown in Fig. 7. Baseline MAP and HR were not different between conditions in either group (Table 1).

Figure 6.

Effect of inositol 1,4,5-trisphosphate (IP₃) receptor blockade with xestospongin C on the exercise pressor reflex. The peak Δmean arterial pressure (peak ΔMAP, A), blood pressure index (BPI, B), and peak Δheart rate (peak ΔHR, C) response to 30 s of 1 Hz dynamic hindlimb skeletal muscle contraction before (control) and after injection of the IP₃ receptor antagonist XeC (5 µg) into the arterial supply of the hindlimb of freely perfused (n = 4) and ligated rats (n = 5). The tension-time index (ΔTTI, D) was not different between conditions. Data were analyzed with Sidak multiple comparisons tests. Bars represent group mean. *Statistically significant (P ≤ 0.05) differences between conditions.

Figure 7.

Effect of inositol 1,4,5-trisphosphate (IP₃) receptor blockade with xestospongin C (XeC) on the time course of the exercise pressor reflex. The Δmean arterial pressure (ΔMAP, A and C) and Δheart rate (ΔHR, B and D) response to 30 s of 1 Hz dynamic hindlimb skeletal muscle contraction before and after injection of the IP₃ receptor antagonist XeC (5 µg) into the arterial supply of the hindlimb of freely perfused (top, n = 4) and ligated (bottom, n = 5) rats. Data were analyzed with two-way repeated measures ANOVA and Sidak multiple comparisons tests. *Time points where statistically significant (P ≤ 0.05) differences exist between conditions. Cond., condition effect; Int., interaction.

DISCUSSION

We investigated the role of sensory neuron TxA₂ receptors and IP₃ receptors in the exaggerated mechanoreflex and exercise pressor reflex in a rat model of simulated PAD in which a femoral artery is ligated for 72 h. We first extended our recent finding that TxA₂ receptors contribute to chronic mechanoreflex sensitization in ligated rats (28) by confirming a role for TxA₂ receptors in the exaggerated exercise pressor reflex in ligated rats. The present investigation also provides the first evidence that IP₃ receptor signaling, a component of second messenger signaling associated with G_q protein coupled receptors such as TxA₂ receptors, contributes to the chronic sensitization of the MA channels that underlie dynamic mechanoreflex and exercise pressor reflex activation in rats with a ligated femoral artery.

The 1-Hz hindlimb skeletal muscle contraction maneuver used herein replicates the rhythmic nature of muscle contractions present during locomotor movement. The contraction protocol elicits a particularly robust mechanical stimulus as evidenced by its production of reflex RSNA bursts in sync with skeletal muscle tension development (27, 30) and the similar magnitude of the pressor and RSNA response to 1 Hz muscle stretch compared with 1 Hz muscle contraction in both freely perfused rats and ligated rats (27). The extension of our recent finding that, in ligated rats, TxA₂ receptor blockade with daltroban reduced to pressor response to dynamic stretch (28) to dynamic contraction experiments was important. Passive hindlimb muscle stretch is a useful but nonphysiological model of mechanoreflex activation isolated from contraction-induced metabolite production. The stretch model of mechanoreflex activation may be used to uncover mechanisms of chronic MA channel sensitization, but any findings produced using the model should be translated to muscle contraction experiments where the mechanoreflex is activated in a physiological manner. The present finding that TxA₂ receptor blockade with daltroban reduced the cardiovascular responses to dynamic muscle contraction in ligated rats likely reflects, at least in part, a TxA₂ receptor-mediated chronic sensitization of MA channels that exists even when contraction-induced metabolite production is absent (28). Acute MA channel sensitization mediated by contraction-induced elevations in COX metabolites (48, 49) may have occurred additively with the chronic sensitization to produce the overall TxA₂ receptor-mediated mechanoreflex contribution to the exercise pressor reflex in our experiments. A role for TxA₂ receptors in the metaboreflex component of the exercise pressor reflex in ligated rats is also possible (44). A limitation of the present investigation is that we did not measure muscle metabolite production during contraction and are unable to make confident conclusions regarding the possible roles of acute mechanoreflex sensitization or metaboreflex activation.

We did not find a role for TxA₂ receptors in the exercise pressor reflex evoked during dynamic contraction in freely perfused rats in this investigation, or in the mechanoreflex evoked in response to dynamic muscle stretch in our recent investigation (28). Conversely, Leal et al. (44) found that TxA₂ receptor blockade with daltroban reduced the pressor response to static muscle contraction and stretch in freely perfused rats. The reason(s) for the discrepancy between the present investigation and that of Leal et al. (44) is unknown but may involve the different modalities (static vs. dynamic) of reflex activation. Specifically, the different modalities are likely to result in very different levels of metabolite production and accumulation and may stimulate somewhat different populations of muscle afferents and/or classes MA channels (29, 50). The possibility of redundancy among receptors masking a TxA₂ receptor-mediated contribution to the exercise pressor reflex in the freely perfused rats in the present investigation must also be considered (51).

XeC is a cell membrane permeable molecule isolated from a marine sponge species (xestospongia) (47) that has been shown to inhibit IP₃ receptors and prevent the release of internal calcium stores into the cytosol in multiple cell types (47, 52–55). Its effect may be mediated by a noncompetitive mechanism in which the calcium channel pore is blocked and/or an allosteric mechanism that uncouples IP₃ binding from calcium release (47). We found that XeC did not reduce the pressor or cardioaccelerator response to dynamic hindlimb muscle stretch or contraction in freely perfused rats. In fact, the pressor response to stretch was higher following XeC compared with control at several time points early in the maneuver (Fig. 4A). Although speculative, this effect may be due to a XeC-induced inhibition of SERCA pumps (46, 56) and a transient elevation in cytosolic calcium levels within sensory neuron endings that briefly sensitized MA channels. Such an effect, however, did not occur during muscle contraction and therefore does not appear important for physiological mechanoreflex activation. Overall, the lack of marked effect of XeC on the mechanoreflex and exercise pressor reflex in freely perfused rats is consistent with our findings that TxA₂ receptor blockade had no effect on the pressor response to hindlimb muscle stretch (28) or contraction (present data) in freely perfused rats. Moreover, we reported recently that COX inhibition (26, 57) and blockade of the COX metabolite-associated endoperoxide 4 receptors (28, 43) had no effect on the pressor response to dynamic hindlimb muscle stretch in freely perfused rats. Collectively, the evidence suggests that in health, the function of the MA channels on thin fiber muscle afferents that underlie dynamic mechanoreflex activation is not modulated by COX metabolite receptor or IP₃ receptor-mediated signaling.

Our finding in ligated rats that IP₃ receptor blockade with XeC reduced the pressor and cardioaccelerator response to stretch identifies a second messenger-signaling component through which TxA₂ receptor-mediated chronic sensitization of MA channels and the mechanoreflex likely occurred in this simulated PAD model (28). In addition, in ligated rats, we found that the effect of XeC on the pressor response evoked during isolated mechanoreflex activation extended to dynamic muscle contraction. That finding suggests that IP₃ receptors contribute to the chronic sensitization of the MA channels that underlie mechanoreflex activation during rhythmic contraction. Although as mentioned above for the TxA₂ receptor blockade experiments, a role for IP₃ receptors in acute mechanoreflex sensitization and/or metaboreflex activation is also possible. The fact that injection of XeC into the jugular vein had no effect on the pressor response to stretch suggests that the effect of XeC when injected into the arterial supply of the hindlimb on the pressor response to stretch in ligated rats is attributable to effects on the sensory endings of thin fiber muscle afferents and not effects elsewhere in the mechanoreflex arc such as the brainstem and/or the spinal cord. The fact that XeC had no effect on the pressor response to lactic acid injection suggests that the effect of XeC on the pressor response to stretch in ligated rats is most likely attributable to an interruption of the cellular signaling between IP₃ receptors and MA channels and is not attributable to an “off-target” effect that produced a generalized reduction in sensory neuron responsiveness. Thus, XeC likely permeated sensory neuron cell membranes, blocked/inhibited IP₃ receptors, and reduced cytosolic calcium concentration within the endings of the sensory neurons that mediate dynamic mechanoreflex activation in ligated rats. The mechanism by which TxA₂ receptor and IP₃ receptor signaling is amplified by chronic femoral artery ligation in the rat likely includes the functional elevation in TxA₂ receptor protein expression in sensory neurons (26, 28). An increase in sensory neuron IP₃ receptor expression or phosphorylation may also contribute and such a determination is an important future direction. It is also important to note that other G_q protein coupled receptors in addition to TxA₂ receptors, such as bradykinin 2 receptors (58), are also likely to influence IP₃ receptor signaling in ligated rats.

Several experimental considerations warrant discussion. First, we used only male rats in this study, which eliminated the possibility of limit sex-dependent variability. Extension of the present findings to female rats and the investigation of possible sex differences is an important future direction. Second, as indicated above, XeC has been shown to inhibit SERCA pumps in addition to IP₃ receptors (46, 56). It seems unlikely, however, that such an effect could have contributed to a reduction in pressor response to stretch and contraction in ligated rats without also reducing those responses in freely perfused rats. Third, atherosclerosis develops slowly with a gradual narrowing of the arteries in patients with PAD, whereas the rat model of simulated PAD relies on instantaneous and complete femoral artery ligation. Nevertheless, femoral artery ligation followed by 72 h of recovery replicates the limb blood flow patterns during exercise and exaggerated exercise pressor reflex found in patients with PAD (59, 60). Fourth, we did not perform experiments in which daltroban was injected systemically. However, we found recently that the systemic (intravenous) injection of daltroban had no effect on the pressor response to mechanoreflex activation in ligated rats (28) or rats with heart failure (43). Based on those findings, we believe daltroban was unlikely to have systemic effects during contraction in the present investigation. Finally, investigating the specific downstream mechanism(s) by which IP₃ receptor signaling modulates MA channel function was beyond the scope of this investigation but constitutes an important future direction. For example, activation of protein kinases (61, 62) and/or modification of the cytoskeleton (63) resulting in altered gating properties of MA channels may be involved.

Perspectives and Significance

We investigated important extensions of our previous findings that COX metabolite (26) and TxA₂ receptor (28) signaling contributes to a chronic sensitization of MA channels and the mechanoreflex in rats with a ligated femoral artery (27). We first confirmed that the role for TxA₂ receptors in chronic mechanoreflex sensitization in ligated rats recently reported by our laboratory is reflected when the exercise pressor reflex is generated during skeletal muscle contraction. We also found that IP₃ receptor signaling, a component of the second messenger signaling linked to G_q protein-coupled receptors such as TxA₂ receptors, contributed importantly to the chronic mechanoreflex sensitization in rats with a ligated femoral artery. PAD patients experience exaggerated increases in blood pressure during exercise and reduced exercise tolerance compared with healthy counterparts (16, 17, 64–68). This investigation reveals important mechanisms within thin fiber sensory neurons that may contribute to reflex-mediated sympathoexcitation and exaggerated blood pressure increases during exercise in this patient population.

GRANTS

This work was supported by National Heart Lung and Blood Institute awards F31HL154779 (to K. S. Rollins) and R01HL142877 (to S. W. Copp).

DISCLOSURES

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

AUTHOR CONTRIBUTIONS

K.S.R., A.L.E.B., A.C.W., and S.W.C. conceived and designed research; K.S.R., A.L.E.B., A.C.W., and S.W.C. performed experiments; K.S.R., A.L.E.B., A.C.W., and S.W.C. analyzed data; K.S.R., A.L.E.B., A.C.W., and S.W.C. interpreted results of experiments; K.S.R., A.L.E.B., A.C.W., and S.W.C. prepared figures; K.S.R., A.L.E.B., A.C.W., and S.W.C. drafted manuscript; K.S.R., A.L.E.B., A.C.W., and S.W.C. edited and revised manuscript; K.S.R., A.L.E.B., A.C.W., and S.W.C. approved final version of manuscript.