Myogenic tone in mouse mesenteric arteries: evidence for P2Y receptor-mediated, Na⁺, K⁺, 2Cl⁻ cotransport-dependent signaling

Abstract

This study examines the action of agonists and antagonists of P2 receptors on mouse mesenteric artery contractions and the possible involvement of these signaling pathways in myogenic tone (MT) evoked by elevated intraluminal pressure. Both ATP and its non-hydrolyzed analog α,β-ATP triggered transient contractions that were sharply decreased in the presence of NF023, a potent antagonist of P2X₁ receptors. In contrast, UTP and UDP elicited sustained contractions which were suppressed by MRS2567, a selective antagonist of P2Y₆ receptors. Inhibition of Na⁺, K⁺, 2Cl⁻ cotransport (NKCC) with bumetanide led to attenuation of contractions in UTP- but not ATP-treated arteries. Both UTP-induced contractions and MT were suppressed by MRS2567 and bumetanide but were insensitive to NF023. These data implicate a P2Y₆-mediated, NKCC-dependent mechanism in MT of mesenteric arteries. The action of heightened intraluminal pressure on UTP release from mesenteric arteries and its role in the triggering of P2Y₆-mediated signaling should be examined further.

Discovered more than 100 years ago, myogenic tone (MT) is the intrinsic ability of small arteries to constrict in response to increases in intraluminal pressure in the absence of blood flow [1]. Later on, MT was documented in nearly all arterial vessels with diameters below 150 μm [2]. Because of its key role in peripheral resistance as well as in the maintenance of constant capillary blood flow and hydrostatic pressure [3], the involvement of abnormal MT responses in the pathogenesis of cardiovascular diseases and protection against hypertension-induced organ damage is widely disputed [4–7].

Despite the obvious physiological and pathophysiological significance of MT, upstream mechanisms that translate pressure changes to vascular smooth muscle cell (VSMC) contractions are poorly understood [3, 8–10]. There are gaps in both our knowledge of the molecular make-up of sensory elements and the sensed signal(s) triggered by elevated intraluminal pressure. In accordance with the most widely discussed hypothesis, pressure-sensing is provided by conformational transitions of ion channels and other potential stretch-activated plasma-membrane-bound proteins and/or the cytoskeleton network [3, 11]. However, the molecular origin of pressure-sensitive protein transducers remains unknown. Our recent estimation of plasma membrane reserves indicates that in most physiological situations, tension of the plasma-membrane-unconstrained lipid bilayer is unlikely to rise to the level required for the activation of mechanosensitive proteins [12]. The disruption of actin-microfilaments by cytochalasin D blocked pressure-induced contractions but did not prevent VSMC depolarization, an obligatory step in MT development. It was also shown that mice lacking key elements of the VSMC cytoskeleton network, such as vimetin and desmin, did not exhibit altered MT (for more details, see [11]).

Searches for potential mechanosensitive systems revealed that very modest shear stress, extracellular fluid movement, and cell volume modulation led to the release of ATP, UTP, and other nucleotides from almost all cell types studied so far, including VSMC and endothelial cells [13–15]. It was also documented that activation of ligand-gated P2X receptor cation channels and G protein-coupled P2Y receptors with released nucleotides contributed to the autocrine regulation of blood vessel tone and remodeling [16, 17]. To the best of our knowledge, the role of purinoceptors in blood pressure-sensing and MT development has not yet been explored.

Ubiquitously expressed Na⁺, K⁺, 2Cl⁻ cotransporter NKCC1 belongs to the superfamily of Cl⁻-coupled carriers. Under baseline conditions ([Na⁺]_o>>[Na⁺]_i and [Cl^-]_o²>>[Cl⁻]_i²), it provides inwardly directed net electroneutral symport of Na⁺, K⁺, and Cl^- and maintenance of [Cl⁻]_i above values predicted by the Nernst equilibrium potential [18]. Importantly, unlike the dominant contribution of K⁺ permeability (P_K) to resting membrane potential (E_m) in skeletal and cardiac muscles, the P_K/P_Cl ratio in VSMC varies from 0.8 to 0.5 [19]. These data suggest that high-ceiling diuretics suppress MT of the renal afferent arterioles via NKCC inhibition, attenuation of [Cl⁻]_i and VSMC hyperpolarization. Indeed, both furosemide and bumetanide decrease [Cl⁻]_i [20–22], hyperpolarize VSMC [21], suppress the activation of L-type Ca²⁺ channels [22] and reduce the baseline tone of vascular and non-vascular smooth muscles [23–25] as well as contractions triggered by diverse stimuli, including [K⁺]_o-induced depolarization [22], phenylephrine [22, 26–28], histamine [29], angiotensin II [30], thromboxane A₂ [31, 32], and oxytocin [33, 34]. Recently, it was demonstrated that high-ceiling diuretics, such as bumetanide and furosemide that are known to be potent NKCC inhibitors, almost completely abolished the MT of renal afferent arterioles in perfused hydronephrotic rat kidneys in vitro [32]. We designed this study to examine the role of P2 receptors and NKCC in MT regulation in mouse mesenteric arteries.

Methods

Mice Experiments were performed on 2- to 4-month-old male C57B1/6 mice obtained from Charles River Canada (Montreal, PQ) and the International Biotechnological Research Centre (Moscow, Russia). Before experimentation, the mice were subjected to a lethal dose of CO₂ in accordance with the protocol approved by the respective institutional animal care and use committees.

Mesenteric artery contractions The animal intestines were removed surgically with their mesentery beds attached and placed in ice-cold, physiologically balanced salt solution (PSS): 135 mM NaCl, 5 mM KCl, 1.5 mM CaCl₂, 1.2 mM MgCl₂, 5 mM glucose, 20 mM HEPES; pH 7.4 was adjusted at 37°C with Tris. Dissected segments of the second- and third-order arteries, 1–2 mm in length, were isolated from the intestinal tract and fat tissues by dissection microscopy (Motic, Richmond, BC). They were transferred to a 15-ml arteriographic chamber (Living System Instrumentation, Inc., Burlington, VT), cannulated at both ends with micropipettes and mounted in a Motic microscope supplied with a video-monitored perfusion system. The signal derived from this system was processed by video dimension analyzer, for continuous measurement of intraluminal diameter. MT was triggered by elevation of ultraluminal pressure from 10 to 80 mmHg. We noted that it was sharply decreased in mesenteric arteries with intraluminar diameter below 120 μm and increased in the presence of bicarbonate. Considering this, we used vessels with intraluminar diameters in the range from 70 to 110 μm and PSS containing 25 mM NaHCO₃ and 120 mM NaCl; pH was adjusted to 7.4 with HEPES. Passive diameter at 80 mmHg (PD₈₀) was measured after 10-min incubation of arteries in Ca²⁺-free PSS containing 2 mM EGTA, and MT was quantified as , where D₈₀ was the baseline intraluminar diameter at the same pressure (Fig. 1). Note that MT was absent at 10 mmHg (PD₁₀∼D₁₀), showing that it was triggered by heightened intraluminal pressure. [K⁺]_o-induced contractions were evoked by equimolar substitution of NaCl with 80 mM KCl.

Fig. 1.

Typical recording showing the effect of intraluminal pressure and Ca²⁺-free medium on the internal diameter of mouse resistance mesenteric arteries in the absence (a) and presence (b) of 100 μM bumetanide

Chemicals Bumetanide, ATP, UTP, UDP, α,β-methylene-ATP (α,β-ATP), pyridoxal-5′-phosphate-6-azo(benzene-2′,4′-disulfonic acid) tetrasodium salt (PPADS), suramin, nicardipine, apyrase—Sigma (St. Louis, MO); 2′-deoxy-N⁶-methyl adenosine-3′,5′-biphosphate (MRS2567}—Tocris Chemicals, Ellisville, MO; 8,8′-carbonylbis(imino-3,1-phenylene)-bis-1,3,5-naphthalenetrisulfonic acid sodium salt (NF023)—Calbiochem, La Jolla, CA; salts, d-glucose and buffers—Sigma and Anachemia (Montreal, PQ).

Results

Mesenteric artery contractions evoked by P2 agonists

Figure 2A demonstrates that the addition of 100 μM ATP led to transient mesenteric artery contractions that were completely abolished in 3–5 min. We also noted that repeated stimulation with ATP in 10 to 30 min of washing in PSS had a negligible effect compared to the first exposure. Unlike ATP, 100 μM UTP elicited sustained contractions (Fig. 2B) which did not significantly change during 40-min treatment (data not shown). Significantly, both sustained contractions evoked by UTP and transient contractions induced by ATP were preserved in UTP-pretreated vessels (Fig. 2).

Fig. 2.

Typical recording charting the effects of 80 mM KCl, 100 μM ATP and 100 μM UTP on the internal diameter of mouse resistance mesenteric arteries

Addition of the non-hydrolyzed ATP analog α,β-ATP and UDP caused transient and sustained contractions, respectively (Fig. 3A). ATP- and α,β-ATP triggered half-maximal mesenteric artery contractions were somewhat similar (∼30 μM), whereas UDP was more effective than UTP (EC₅₀ ∼5 and 10 μM, respectively; Fig. 3B).

Fig. 3.

a Typical recording illustrating the effects of 80 mM KCl, 100 μM α,β-ATP, and 100 μM UDP on the internal diameter of mouse resistance mesenteric arteries. b Dose-dependent actions of ATP, UTP, UDP, and α,β-ATP on mesenteric artery contractions. Maximal values of contractions evoked by 80 mM KCl were taken as 100%. Means ± S.E. of three independent experiments are shown

Effect of P2 antagonists

It is well-documented that suramin modestly suppresses several subtypes of P2X and P2Y receptors [14]. We observed that this compound decreased UTP-induced contractions by 22% (p < 0.05), but did not significantly change contractions evoked by ATP (Fig. 4). Contractions triggered by UTP were also suppressed by PPADS, a modest antagonist of P2X₂, P2X₃, P2X₅, P2Y₂, and P2Y₆ receptors [14, 35], and by MRS2567, a selective antagonist of P2Y₆ receptors [35]. ATP-induced contractions were not affected by the above-listed compounds, but were sharply reduced in the presence of NF023, a potent antagonist of cloned P2X₁ and P2X₃ receptors [14].

Fig. 4.

Effect of purinoceptor antagonists on mesenteric artery contractions evoked by ATP (a) and UTP (b); 100 μM suramin, 100 μM PPADS, 10 μM MRS2567 and 10 μΜ NF023 were added 10 min before the treatment of mesenteric arteries with 100 μM ATP or UTP. Maximal values of contractions measured in the absence of purinergic antagonists were taken as 100%. Means ± S.E. of three independent experiments are reported. *p < 0.02

Effect of bumetanide

Figure 5 shows that NKCC inhibition with bumetanide led to twofold attenuation of mesenteric artery contractions caused by modest depolarization ([K⁺]_o = 30 mM), but did not affect contractions in vessels subjected to full-scale depolarization ([K⁺]_o = 80 mM). These results are consistent with data obtained in a study on contractions of depolarized endothelium-denuded strips from rat aortae [22]. We did not observe any inhibitory action of bumetanide on contractions triggered by ATP, whereas contractions of UTP-treated arteries were decreased by 60% (Fig. 5).

Fig. 5.

Effect of bumetanide on mesenteric artery contractions evoked by KCl, ATP, and UTP. Bumetanide (100 μM) was added 30 min before the treatment of mesenteric arteries with KCl (80 or 30 mM), ATP (100 μM) or UTP (100 μM). Maximal values of contractions triggered by 80 mM KCl and measured in the absence of bumetanide were taken as 100%. Means ± S.E. of three independent experiments are presented

Myogenic tone

MT was diminished by ∼3.5-fold in mesenteric arteries pretreated for 40 min with MRS2567 (p < 0.02; Fig. 6). It was also suppressed by suramin and PPADS, but these differences were not statistically significant. Bumetanide time-dependently increased mesenteric artery intraluminal diameter pressurized at 80 mmHg (Fig. 1B) which resulted in a twofold attenuation of MT (Fig. 6).

Fig. 6.

Effects of purinoceptor antagonists and bumetanide on MT in mesenteric arteries treated with 100 μM suramin, 100 μM PPADS, 10 μM MRS2567, 10 μΜ NF023 or 100 μM bumetanide 40 min before their transfer in Ca²⁺-free PSS, as illustrated in Fig. 1. In the absence of the tested compounds, MT varied in the range from 29 to 41% of passive diameter, and these values were taken as 100%. Means ± S.E. of four independent experiments are shown

Discussion

Seven P2X receptor subunits (P2X_1–7) and seven P2Y receptors (P2Y₁, P2Y₂, P2Y₄, P2Y₆, P2Y₁₁, P2Y₁₂, and P2Y₁₃) have been cloned from mammalian cDNA libraries [14]. Among them, P2X₁, P2X₂, P2X₄, P2X₅, P2Y₂, P2Y₄, and P2Y₆ mRNA species and proteins interacting with anti-P2X₁, anti-P2Y₂, and anti-P2Y₆ antibodies have been detected in rodent small arteries [36–40]. Our results show that transient and sustained constrictions of mouse mesenteric arteries evoked by ATP and UTP, respectively, are mainly mediated by P2X₁ and P2Y₆ receptors. Indeed, the transient kinetics of ATP-induced contractions followed by prolonged inactivation of this signaling pathway (Fig. 2) are consistent with rapid decay of ATP-induced cation current detected in cells transfected with P2X₁ and P2X₃ but not other P2X isoforms [41, 42]. Significantly, ATP-induced contractions were sharply suppressed by NF023, a compound that inhibited ion currents mediated by cloned P2X₁ and P2X₃ with IC₅₀ of 0.2 and 8 μM, respectively, but did not affect the activity of the other P2 receptors [14]. The major role of P2X₁ receptors in ATP-induced contractions is consistent with the absence of an inhibitory action of PPADS, a potent antagonist of P2X₂, P2X₃, and P2X₅ receptors [14, 35], and with the lack of contractions in ATP-treated mesenteric arteries from P2X1^−/− mice [39].

A key function of P2Y₆ receptors in sustained constrictions triggered by UTP was supported by several observations. First, sustained constrictions were absent in arteries treated with ATP and α,β-ATP, potent agonists of all cloned P2X and P2Y receptors with the exception of the P2Y₆ isoform [14, 40]. Second, the apparent affinity of sustained contractile responses was higher for UDP compared to UTP (Fig. 3B), which is consistent with the rank-order potency for P2Y₆ rather than for other UTP- and UDP-sensitive receptors, such as P2Y₂ and P2Y₄ [40, 43]. Third, UTP-induced constrictions were sharply suppressed in the presence of 10 μM MRS2567 (Fig. 4). At this concentration, the compound almost completely depressed the signaling evoked by P2Y₆ without any effect on the activity of cloned P2Y₂ and P2Y₄ receptors [35].

Our results indicate that sustained constrictions caused by P2Y₆ G protein-coupled receptors rather than transient activation of P2X₁ receptor-mediated, inwardly directed Ca²⁺ currents are involved in the long-term maintenance of slowly-developing MT in mesenteric arteries. Indeed, both UTP-induced contractions and MT were depressed by the NKCC inhibitor bumetanide and the P2Y₆ antagonist MRS2567 and were insensitive to the P2X₁ antagonist NF023 (Figs. 4B, 5, and 6). In contrast, constrictions of ATP-treated vessels were sharply suppressed by NF023 but were resistant to bumetanide (Fig. 5). The distinct action of bumetanide on ATP- and UTP-induced contractions is consistent with different mechanisms of action of these nucleotides on excitation-constriction coupling. Indeed, in rat mesenteric arteries, UTP triggered modest and sustained depolarization from −60 to −50 mV [44] that was probably caused by the activation of Ca²⁺-sensitive Cl⁻ channels and could be abolished by the addition of bumetanide that decreases [Cl⁻]_i and approaches Nernst Cl⁻ equilibrium potential to resting E_m [20–22]. Unlike UTP, ATP affected VSMC contractions via transient activation of Ca²⁺ influx through P2X ion channels and independently of E_m modulation [44]. Finally, the slow kinetic of MT development in mesenteric arteries is consistent with the presence of P2Y₆ receptors in endothelial cells triggering NO-mediated vasorelaxation [45] and suppressing the rapid phase of vasoconstriction evoked by interaction of UTP/UDP with VSMC P2Y₆ receptors.

The data considered above are consistent with the signaling mechanism presented schematically in Figure 7. In accordance with the proposed model, intraluminal pressure acting through unidentified sensors leads to the release of intracellular nucleotides, such as ATP and UTP. Extracellular ATP triggers inwardly directed Ca²⁺ currents via activation of non-selective cation channels assembled from P2X₁ subunits. These channels are, however, rapidly inactivated, which results in transient artery contractions rather than long-lasting MT. Unlike ATP, UTP activates P2Y₆ Gq/G₁₁-coupled receptors that, in turn, evoke Ca²⁺ release from the endoplasmic reticulum and activation of protein kinase C (PKC). Ca²⁺ release from the endoplasmic reticulum elicits activation of Cl⁻ channels, plasma membrane depolarization and activation of long-lasting L-type Ca²⁺ channels, whose involvement in MT is well-documented in an overwhelming number of vessels studied so far [3]. Significantly, NKCC inhibition partially blocked the depolarizing action of Cl⁻ channel activation via attenuation of [Cl^-]_i, as demonstrated in bumetanide- and furosemide-treated smooth muscle cells of different origins [20–22], and resulted in the MT suppression documented here (Fig. 6). Side-by-side with the [Ca²⁺]_i signaling precipitated by P2Y₆ receptors, MT may involve PKC activation that, in turn, leads to heightened [Ca²⁺]_i sensitivity of the contractile machinery via myosin light chain kinase phosphorylation [3].

Fig. 7.

Possible mechanisms involving P2Y₆ receptors and Na⁺, K⁺, 2Cl⁻ cotransporter (NKCC) in the maintenance of MT. IP₃ inositol 1,4,5-triphosphate, PIP₂ triphosphoinositide, PKC protein kinase C, PLC phospholipase C; 1 Ca²⁺-activated Cl⁻ channels, 2 voltage-gated L-type Ca²⁺ channels, 3 endoplasmic reticulum, ? unknown pressure-sensing mechanism. Compounds used in the study appear in italics

The model considered above assumes that MT is suppressed by apyrase (ATP diphosphohydralase), providing ATP and UTP hydrolysis to yield di- and monophosphonucleosides [46]. However, we failed to detect any significant action of 2 U/ml apyrase on MT of pressurized mesenteric arteries (data not shown). To explain these negative results, two hypotheses may be proposed. First, the diffusion limit caused by adventitial and endothelial cells protects released ATP and UTP from degradation by apyrase. Second, apyrase-mediated accumulation of diphosphonucleosides is sufficient to activate purinoceptors. Indeed, we demonstrated that UDP is a more potent activator of mesenteric artery contractions than UTP (Fig. 3B). Precise measurements of tri-, di-, and monophosphonucleosides released from pressurized vessels in space-limited compartments should be performed to further evaluate their relative impact on MT maintenance.