Purinergic receptor antagonism: a viable strategy for the management of autonomic dysreflexia?

Abstract

The purinergic receptor ligand, ATP, may participate in reflex induced vasoconstriction through sympathetic efferent and sensory afferent mechanisms. However, the role of the purinergic system in contributing to autonomic dysreflexia following spinal cord injury is unclear. The present study investigates the involvement of P2X receptors in contributing to pressor responses during autonomic dysreflexia. Twenty rats were subjected to spinal cord injury and 24-hours later hemodynamic responses to colorectal distension were recorded. Animals were randomized to receive intravenous administration of the P2X receptor antagonist, NF023, or vehicle control. The data indicate that NF023 attenuates pressor responses to colorectal distension.

Introduction

Purinergic P2 receptors bind ATP and are now recognized to be involved in a wide range of physiological actions involving blood vessel tone and blood pressure regulation [1, 2]. ATP is co-released with norepinephrine from sympathetic terminals innervating blood vessels [2–4] and binds to P2X receptors located on vascular smooth muscle cells where it potentiates vasoconstriction [4, 5]. Distinctly, when ATP binds P2Y receptors located on vascular endothelial cells, it promotes vasodilation [2]. Release of ATP from sympathetic terminals produces vasoconstrictor effects which are abolished by inhibition of postsynaptic P2 receptors [6, 7]. Importantly, purinergic receptors have been implicated in the development of systemic hypertension in both the spontaneously hypertensive rat [8], and in Angiotensin II-dependent hypertension [9]. Taken together these studies support the important roles of ATP released from sympathetic terminals and P2 receptors in controlling systemic blood pressure in health and disease states.

Blood pressure control is particularly important for people living with spinal cord injury (SCI) [10, 11]. SCI occurring at or above the 6^th thoracic (T) segment interrupts descending control of sympathetic preganglionic neurons, which leaves spinal reflexes below the level of the injury unregulated. Consequently, these individuals experience transient and recurring episodes of increased blood pressure called autonomic dysreflexia (AD) [12, 13]. AD can range in severity with most severe cases leading to life threatening complications. The recurring nature of AD presents added burden to the cardiovascular system as it promotes maladaptive changes in the heart and blood vessels [14, 15].

AD increases sympathetic drive to blood vessels; however, the specific neuroeffector mechanisms are complex and still the subject of investigation. Studies have shown that selective blockade of α and β adrenoreceptors using pharmacological agents markedly attenuates AD [16, 17]; however, this therapeutic approach is not suitable for chronic AD prophylaxis, as individuals living with SCI generally suffer from hypotension which is exaggerated by anti-hypertensive therapy. Furthermore, two independent clinical studies have confirmed that α adrenergic blockade does not entirely abolish AD [18, 19].

Non-adrenergic vasoconstrictor mechanisms involved in AD are poorly understood. Al Dera and Brock observed that SCI increases the reactivity of visceral arterial vasculature to angiotension II [20], while the involvement of neuropeptide Y was deemed negligible in AD-associated increases in blood pressure in spinalized rats [16]. Perhaps the most convincing case for the involvement of non-adrenergic neurotransmission in AD was recently made by Sangsiri and colleagues who studied purinergic neurotransmission in perivascular sympathetic neurons of the mesenteric arteries following SCI [21]. They concluded that SCI leads to altered purinergic signaling, which results from an increased number of ATP-containing vesicles within sympathetic terminals, and/or increased ATP content within vesicles. This ultimately results in exaggerated constriction of vascular smooth muscle upon neurotransmitter release into postjunctional sites. Collectively, these studies support the possibility for non-adrenergic neurotransmitters, such as ATP, in contributing to AD. We hypothesized that ATP acting via P2X receptors plays a role in modulating increases in blood pressure in response to visceral stimulation following SCI. This initial study indicates that P2X receptor antagonism partially attenuates increases in blood pressure triggered by colorectal distension (CRD) in spinalized rats and suggests that therapeutic targeting of the purinergic system may constitute a novel therapeutic approach for the management of AD.

Materials and Methods

All animal protocols and surgical procedures employed in this study were reviewed and approved by the Wayne State University Institutional Animal Care and Use Committee and were performed in accordance with the Guide for the Care and Use of Laboratory Animals.

The effect of P2X receptor blockade on the generation of visceral-vasoconstrictor reflexes was studied in 20 male, spinalized, Wistar rats (Charles River, Wilmington, MA). On day zero animals were subjected to spinal cord transection of the 3trd thoracic segment. 24 hours later, animals were randomized to receive (i) 0.03 mg/kg NF023 (n=4), (ii) 0.3 mg/kg NF023 (n=4), (iii) 3.0 mg/kg NF023 (n=6), or (iv) vehicle control (saline; n=6). HR, and AP responses to CRD were quantified in unanesthetized decerebrate animals.

On day 0, the average animal weight was 369 ± 21 g. The SCI surgery was performed as described previously [22–26]. Briefly, animals were anesthetized with 2 % isoflurane. Rectal temperature was maintained at 37.0 °C using a homeothermic blanket (Physio Suite, Kent Scientific, Torrington, CT). Carprofen (5mg/kg) and lidocaine (0.5mg/kg) were administered subcutaneously and animals were prepared for aseptic survival surgery according to the Handbook for Laboratory Animal Care and Use. Following a skin incision a dorsal laminectomy between T2 and T3 vertebral segments was performed. The dura mater was excised, and the spinal cord was transected using microscissors. Transection was deemed complete following visual confirmation. Overlaying muscle and skin incisions were closed using 3-0 suture and wound clips (Roboz Surgical Instruments), respectively. Animals were allowed to recover in a temperature-controlled chamber for 24 hours (Techniplast, West Chester, PA) before undergoing the terminal experiment. Bladders were expressed twice the day of the surgery and one time before the terminal experiment next day. Food pellets and water were provided ad libitum.

Twenty-four hours following SCI animals underwent the terminal experiment. Anesthesia was induced with 3 % isoflurane and maintained with 1-2 % isoflurane during instrumentation. Rectal temperature was maintained at 37 °C via a water circulating blanket (Stryker, Kent Scientific). Animals were spontaneously ventilated. The left femoral artery and vein were cannulated using PE50 tubing for blood pressure monitoring and infusion of drugs, respectively. HR, SBP, diastolic blood pressure, and mean arterial pressure were derived from the arterial pressure tracings. The signal was digitized using Micro 1401-3 and Spike2 software (Cambridge Electronic Designs, Cambridge UK). Following instrumentation, animals underwent the decerebration procedure. Isoflurane was discontinued, and animals were allowed to recover for 1 hour prior to initiation of the experimental protocol. This general experimental design is similar to our recently published experimental model for investigating visceral-sympathetic reflex reactivity following SCI [27].

Decerebration was performed as described previously [27–29]. Carotid arteries were temporarily occluded using aneurysm clips which were removed following the procedure. With head secured in a stereotaxic frame (Kopf, Germany), craniotomy between the sagittal and lambdoid sutures was performed. The brain was transected pre-collicularly no more than 1mm rostral to superior colliculus and all the remaining CNS structures rostral and lateral to the transection were aspirated. The cavity was packed with cotton balls (Ethicon, Johnson & Johnson). Isoflurane was discontinued and a minimum recovery of 1h was employed [30].

CRD was induced by inflation of an 8 French latex, fluid-filled balloon catheter (Covidien., Ltd., Dublin, Ireland) which was inserted 2 cm past the anal verge, as described previously by us and others [27, 31, 32]. This is an effective and reproducible means of eliciting viscero-sympathetic reflexes [31]. The catheter was secured to the tail to prevent retraction. CRD was initiated over several seconds and maintained for 1 min.

Selective P2X receptor antagonist NF023, an analog of suramin, 8,8′-[carbonylbis(imino-3,1-phenylenecarbonylimino)] bis-1,3,5-naphthalene-trisulphonic acid, hexasodium salt was purchased from Sigma Aldrich (St Louis, MO) [33, 34]. NF023 was dissolved in sterile saline immediately before use and administered intravenously to animals at a dose of 0.03mg/kg, 0.3mg/kg, or 3.0mg/kg. The choice of the drug and the dose were determined based on literature search which showed that NF023 has analgesic properties in acute thermal nociception [35].

Average hemodynamic data were expressed as a maximum change during CRD from baseline levels (average of 5 min) immediately before CRD. Average data were analyzed using the Kruskal-Wallis test followed by Dunn’s multiple comparisons test when appropriate. Baseline hemodynamic data from vehicle and 3.0 mg/kg NF023 treated animals were analyzed using one-tailed Student’s t-test. An alpha level of <0.05 was considered significant.

Results

Representative examples of arterial pressure (AP) and heart rate (HR) responses to 2 ml CRD in a vehicle (saline)-treated animal are provided in Figure 1A. CRD evoked increases in AP and decreases in HR. Figure 1B, C & D show the averaged maximal responses in HR, systolic blood pressure (SBP), and diastolic blood presssure (DBP) to CRD among animals randomized to receive intravenous administration of the P2X receptor antagonist, NF023 (0.03, 0.3 or 3.0 mg/kg) or vehicle control. The mean CRD-evoked increase in SBP among the vehicle treated animals was 30.3 ± 7.1 mmHg, which was associated with a reflex decrease in HR of 45.7 ± 24.2 bpm. Administration of the highest dose of NF023, 3.0mg/kg, significantly attenuated CRD-evoked increases in SBP and decreases in HR (p<0.05). Administration of the lower doses of NF023 did not have significant effects on CRD responses.

Figure 1.

The effect of NF023 on blood pressure and heart rate (HR) during colorectal distension (CRD) in spinalized rats. (A) typical response in HR and arterial pressure (AP) during CRD in a vehicle-treated spinalized animal. (B - D) depict average maximal changes in HR, systolic blood pressure (SBP) and diastolic blood pressure (DBP); n=4-6 per group, p<0.05.

Baseline changes in hemodynamic parameters following administration of 3.0 mg/kg of NF023 or vehicle control are presented in Table 1. No significant change in diastolic blood pressure, SBP, HR or mean arterial pressure was observed following administration of 3.0mg/kg of NF023 when compared to vehicle control.

Table 1.

Effect of NF023 treatment on baseline hemodynamic parameters

	Diastolic pressure (mmHg)	Systolic pressure (mmHg)	Heart rate (bpm)	Mean arterial pressure (mmmHg)
Vehicle	57±11	85±19	326±79	68±11
NF023	62±9	85±16	293±44	71±9

Discussion

The principle finding of the present study is that blockade of P2X receptors using NF023, attenuates AD responses in the rat model of SCI. Intravenous administration of NF023, had negligible effects on baseline blood pressure, which is important as subjects living with SCI exhibit baseline hypotension which is exacerbated by most antihypertensive treatments. Given this severe side effect, antihypertensive agents are not generally administered prophylactically for the management of AD. Since blockade of P2X receptors did not dramatically effect baseline blood pressure, P2X receptor antagonism may represent a suitable prophylactic approach for the management of AD and is worthy of further investigation. While we observed moderate attenuation of blood pressure responses to visceral stimulation following P2X receptor antagonism, the potential role of P2Y receptors remains to be determined. P2Y receptors which are similarly activated by ATP are known to have vasodilatory effects which may be beneficial during acute AD episodes.

The overall contribution of P2 receptors in modulating AD is further complicated by actions of these receptors on signaling within sensory nerves. Notably, both P2X and P2Y receptors are involved in modulation of neurotransmission and may alter the spinal reflex neurocircuitry that drives AD responses. These two receptor subtypes modulate synaptic transmission quite differently. P2X receptors are non-selective cation channels and as such mediate transient and rapid effects which facilitate neurotransmission [2]. P2Y receptors are G-protein coupled receptors which are characterized by slower and more diverse neuromodulatory functions [2]. Ando et al. observed that blockade of P2Y receptors had a superior effect over blockade of P2X receptors for attenuating mechanical allodynia in rodent models of chronic neuropathic and inflammatory pain, while P2X receptor blockade was superior for attenuating acute thermal pain [35]. The beneficial effects of P2 receptor antagonism for increasing pain threshold likely involves modulation of the sensory C fibers and inhibition of Ca²⁺ currents within sensory neurons [36]. Given the similar pathophysiological mechanisms underlying AD and pain, it is tempting to speculate that P2 receptor antagonism may effectively target both vasoconstrictor and sensory afferent components of AD reflex responses. Future studies incorporating microneurography are warranted to evaluate the potential contribution of the receptors in the afferent and efferent arms of this pathophysiological reflex arc.

ATP and norepinephrine are stored in synaptic vesicles within sympathetic fibers. Even though there is evidence for existence of purinergic co-transmission, this does not necessarily indicate that this contribution is physiologically relevant. Indeed, the type and size of the blood vessels as well as experimental conditions are important to consider. The contribution of ATP as a contractile neurotransmitter is thought to increases as the diameter of the blood vessel decreases; such that, in the small branches of the mesenteric arteries, ATP neurotransmission significantly contributes to contractile responses [37]. Interestingly, low frequency firing of sympathetic nerves has been associated with the release of non-adrenergic neurotransmitters while high frequency firing is believed to consist of predominantly adrenergic components [38, 39]. Finally, baseline blood vessel tone is an important determinate of vascular reactivity to ATP [40]. Given that SCI is associated with baseline hypotension, it is possible that the contribution of ATP as a contractile co-neurotransmitter may be decreased. However, it is also possible that as blood vessel tone increases during AD episodes, so do the contractile effects of ATP co-neurotransmission leading to even higher blood pressure increases.

Limitations

The present study implicates the role of ATP, which is likely released from the sympathetic nerve terminals and acts on P2X receptors, in contributing to constriction of blood vessels and increasing blood pressure during AD. The observed reduction in magnitude of AD responses following P2X receptor blockade may, in part, occur via modulation of the sensory afferent end of this reflex arc. A convincing case for the involvement of ATP in mediating inflammatory, neuropathic and visceral pain has been made [41, 42] and in particular for the types of pain being caused by distension of the hollow organs which is a common natural stimulus for AD. In our study, P2X receptors antagonism was achieved using NF023, which blocks both P2X1 and P2X3 receptors. While the vasoconstrictor effects of ATP are mainly attributed to P2X1 receptors located on vascular smooth muscle cells, P2X3 receptors are likely involved in modulation of sensory afferent neurotransmission [43]. Thus, the precise the mechanism(s) of action of NF023 in this study remain unclear.