Background
Gap junctions are clusters of intercellular channels resulting from the connection of two hexameric assembly of membrane proteins termed connexins (Cx) [
1]. Each hexameric assembly is also known as a hemichannel or connexon, localized on the membrane of two adjacent cells and arranged with identical Cx (homomeric connexon) or different Cx (heteromeric connexon) with various possible combinations [
2]. Such process has functional consequences and provides an efficient cellular strategy to finely regulate cell-to-cell communication. In the vascular wall, the most common connexins are Cx37, Cx40 and Cx43 in endothelial and smooth muscle cells [
3]. Gap junctions allow cell-to-cell coupling in between vascular cells of the same type, namely endothelial or smooth muscle cells but they are also present in between endothelial and smooth muscle cells (myoendothelial gap junctions). Gap junctions allow direct diffusion of ions and small molecules between adjacent cells in almost all animal tissues. As a consequence, gap junctions are vital components in the coordination of vascular response and are therefore essential for the control of vascular functions including vasoreactivity and cell proliferation [
3]. There is now accumulating evidence indicating that Cx may play a role in a variety of vascular diseases including systemic arterial hypertension [
4]. For instance, elevated pressure has been shown to increase the expression of Cx 43 in cultured cells from aorta [
5]. However, the role of gap junctions in pulmonary hypertension (PH) remains largely unknown.
PH is a multifactorial disease characterized by a progressive increase in pulmonary vascular resistance caused by vasoconstriction, vascular cell proliferation and obliteration of pulmonary microvasculature. These processes lead to right heart failure and ultimately to death [
6]. PH occurs in a variety of clinical situations and is associated with a broad spectrum of histological patterns and molecular abnormalities. Because of this diversity, early diagnosis is difficult and efficient treatments are still lacking. The recent revision of the classification of PH distinguishes five groups [
7]. Among these groups, the category 1 PH also known as pulmonary arterial hypertension (PAH) includes idiopathic PAH, familial PAH and acquired PAH, the latter of which being associated with other diseases such as HIV or connective tissue diseases. The non-category 1 PH previously known as secondary PH includes the category 3 which is a widely distributed PH secondary to alveolar hypoxia as a result of lung disease such as chronic obstructive pulmonary disorder (COPD). Although, PH has progressively evolved from a fatal to a chronic disease, none of the currently available therapies is curative [
8]. Despite intensive research, PH remains an important medical challenge and a better knowledge of the underlying molecular and cellular mechanisms remains crucial for the development of new or additional innovative therapies.
To comprehensively address the issue of the role of gap junction in PH, we have used two different rat models, the hypoxia and monocrotaline-induced models that share pathophysiological characteristics with category 3 and category 1 PH, respectively. Like category 1 and 3 PH patients, monocrotaline- and the chronic hypoxia-treated rats (MCT and CH rats respectively) exhibit high circulating concentrations of serotonin (5-HT), endothelin-1 (ET-1) and norepinephrine (an adrenoceptor agonist) [
9‐
14]. These increased concentrations of 5-HT, ET-1 and norepinephrine participate to the increase in pulmonary vascular tone [
14‐
16]. Moreover, reduced expression of a variety of potassium channels and mainly voltage-gated and calcium-activated potassium channels leads to membrane depolarisation, voltage-gated calcium channel opening and cytosolic calcium increase in rat and patients with PH [
15,
17]. The resulting intracellular calcium increase participates to the high pulmonary arterial vascular tone observed in PH.
In the normal pulmonary arterial wall, we have recently demonstrated a functional role of gap junctions in contractile and calcium responses to 5-HT [
18]. The aim of the present study was thus to examine the role of gap junctions in the abnormal pulmonary arterial wall in PH. We have demonstrated that Cx 37, 40 and 43 are expressed in intrapulmonary arteries (IPA) from the normoxic (N), chronic hypoxia (CH) and MCT rats and that Cx 43 is overexpressed in CH rats. By using Cx-mimetic peptides as blockers of gap junctions [
18,
19], we also highlight the role of Cx in the contractile responses to stimuli, already known to be involved in PH, namely 5-HT, ET-1, phenylephrine (an α1-adrenoceptor agonist, Phe) and depolarising solutions (high potassium solutions).
Discussion
Underlying cellular mechanisms as well as targets, governing the pathogenesis of PH, still remain to be fully elucidated. This may be due, at least in part, to the fact that the pathogenesis of PH differs depending on the type of aetiology of this vascular disorder. For this reason, we conducted the present study in two different PH models: MCT induced PH (non hypoxic model) and CH-induced PH (hypoxic model).
Two major excitation-contraction coupling mechanisms are involved in regulating pulmonary vascular tone: electromechanical and pharmacomechanical coupling. We thus adressed agonist-mediated pulmonary contraction (pharmacomechanical coupling induced by either 5-HT, ET-1 or Phe) and depolarisation-mediated contraction (electromechanical coupling induced by high potassium solution). In the present study, IPA from CH rats exhibited hyperreactivity to 5-HT (Figure
3A), which is consistent with previous studies [
22,
24,
25]. Regarding ET-1, vasoconstriction to this agent was only slightly decreased in hypoxic PH whereas it was dramatically decreased in MCT-induced PH. In CH rats, pulmonary arterial contractile responses to ET-1 have been shown to be increased in resistant IPA but decreased in extra pulmonary arteries [
16]. Consequently, the slight decrease in the contraction to ET-1 observed in CH rats may be explained by the intermediate size of the pulmonary artery (PA) used for the present contractile experiments (namely PA of the first order or IPA1). In MCT rats, a recent study has shown that although the contraction to ET-1 was not altered, ET-B receptors were decreased in resistance PA [
26]. Since the contraction to ET-1 is linked to ET-A receptors in extra-PA and to ET-B receptors in IPA [
26,
27], the present decreased ET-1 contraction may result from the reported decrease in ET-B receptors in MCT rats. Regarding Phe and in contrast to ET-1, we found that Phe-induced contraction was increased in PA from MCT rats. A recent report from Mam
et al. (2010) showed a reduced contraction to Phe in CH and MCT rats [
28]. This difference may be explained by different animal strain (Sprague Dawley
vs Wistar rats) and/or different vascular preparation (extra-PA
vs IPA) and/or different experimental conditions (normobaric
vs hypobaric hypoxia). Finally, regarding the receptor independent depolarizing agent KCl, pulmonary vascular reactivity to membrane depolarization was increased in CH rats, whereas it was reduced in MCT rats. These results are consistent with previous findings from our laboratory indicating that PA smooth muscle cells (PASMC) from CH rats exhibit a higher resting membrane potential and a higher basal intracellular calcium concentration [
22,
29]. Moreover, the decreased expression of potassium channels previously observed in CH rats [
17] could also contribute to increase the sensitivity to depolarisation.
We have previously shown that Cx 37, 40 and 43, three gap junction proteins commonly found in the vasculature, are expressed in rat PA endothelial cells, while Cx 37 and 40 only are found in PASMC [
18]. In the current study, Cx 37, 40 and 43 were also expressed in pulmonary arterial wall from both hypoxic and non-hypoxic PH rats. However, PASMC from CH rats no longer expressed Cx 37 and 40 isoforms, whereas they were sparsely expressed in MCT rats (Figure
2). Interestingly, Cx 43 expression was upregulated in CH rats but not in MCT rats (Figure
1). We cannot exclude that Cx 43 is expressed at the myoendothelial junctions and possibly on the smooth muscle side of these heterocellular structures. Cowan et al. have indeed demonstrated that hypoxia (2.2% O
2 during 6 h) increased Cx 43 expression in cultured smooth muscle cells from rat thoracic aorta [
5]. Consequently, we can hypothesize that the increase in Cx 43 in the IPA from CH rat could be due to the effect of hypoxia on Cx 43 localized on the smooth muscle side of the myoendothelial gap junctions. Interestingly, we have previously demonstrated that Cx 43 present between PASMC and endothelial cells plays an important role in the vasoreactivity to 5-HT in IPA from N rats [
18].
Since (i) the three Cx 37, 40 and 43 are expressed in CH and MCT rats (Figure
1 and
2) and (ii) Cx 37 and 43 are involved in the contractile and calcium responses to 5-HT in IPA from N rats [
18], we addressed the role of these Cx in the contractile response to various agonists known to contribute to PH (5-HT, ET-1 and Phe) in IPA from CH and MCT rats. In the present study, the effect of Gap27 varied according to the agonists used and the rat model considered. Since we have previously demonstrated that the role of the gap junctions in the contraction depended on the amount of superoxide anion (O
2
●
) level in smooth muscle, it can be hypothesized that the modulation of O
2
●
level in IPA from hypertensive rat models may differ according to the agonist. In this respect, we can suggest that, unlike 5-HT, Phe increases O
2
●
levels in pulmonary hypertensive models and not in N rats although this hypothesis would require further experimental investigation.
Moreover, 5-HT, ET-1 and Phe are pulmonary arterial vasoconstrictors with different modes of actions. Smooth muscle contraction is well known to be both dependent and/or independent on cytosolic calcium increase depending on the vasoconstrictor. When smooth muscle contraction is calcium independant, the contraction is due to calcium sensitization of the contractile proteins. On the one hand, ET-1 and 5-HT increase intracellular calcium by acting on receptors localised in both smooth muscle and endothelial cells (namely 5-HT
2A, 5-HT
1B/D, ET
A and ET
B in PASMC and 5-HT
2B and ET
B in endothelial cells) whereas Phe acts on α
1-adrenoceptors on PASMC only [
14,
21,
30,
31]. On the other hand, although 5-HT, ET-1 and Phe activate inositol 1,4,5-triphosphate-induced intracellular calcium release, 5-HT and Phe also stimulate voltage-independent calcium permeable channels (namely receptor- and/or store-operated channels) whereas ET-1 rather stimulates voltage-dependent calcium channels [
32‐
34]. Finally, contraction to ET-1 and Phe are strongly dependent on calcium sensitization of contractile proteins whereas contraction to 5-HT is only slightly dependent on this process [
21,
35‐
37]. Altogether, such differences in the contractile mechanisms might explain why the effects of the gap junction blockers differ according to the agonist used.
Regarding receptor-independent depolarisation, we have observed a significant decrease in the contraction to KCl in IPA from MCT rats but an increase in CH rats (Figure
3D). Moreover,
37-43Gap27 decreased the sensitivity to high potassium solutions in IPA from CH rats only, thus inducing a contraction similar to the one observed in the N rats (Figure
7C). These results suggest that Cx 37 and/or 43 were involved in the hypersensitivity to KCl in CH rats. This result is thus in good agreement with the expected role of Cx in the conduction of the electrical activity observed in various systemic vessels [
38].
Authors' contributions
MB, DD, RM, JPS and CG contributed to the conception and design of the study. MB, DD and CG performed experiments, evaluated results and interpreted data. MB, DD, RM, JPS and CG were involved in drafting and revising the manuscript. All authors read and approved the final manuscript.