OUP user menu

Continuous inhalation of carbon monoxide attenuates hypoxic pulmonary hypertension development presumably through activation of BKCa channels

Eric Dubuis, Marie Potier, Rui Wang, Christophe Vandier
DOI: http://dx.doi.org/10.1016/j.cardiores.2004.11.007 751-761 First published online: 15 February 2005


Objective: We tested the hypothesis that inhalation of a low concentration of exogenous carbon monoxide (CO) attenuates the development of hypoxic pulmonary artery hypertension by activation of large-conductance voltage and Ca2+-activated K+ channels (BKCa).

Methods: The BKCa activity was measured using whole-cell and inside-out patch clamp recordings in Wistar rat pulmonary artery (PA) myocytes. Pulmonary artery pressures were measured in vivo and membrane potentials were recorded in vitro in pressurized resistance arteries.

Results: Chronic CO inhalation slightly increases single-channel conductance of BKCa channels and induces a large increase in the sensitivity of BKCa channels to Ca2+ of PA myocytes from normoxic and chronic hypoxic rats. Consequently, BKCa currents are increased and play a more prominent role in controlling resting membrane potential of PA myocytes. Chronic CO inhalation also reduces hemodynamic changes induced by chronic hypoxia and attenuates hypoxic pulmonary artery hypertension.

Conclusion: Chronic inhalation of CO attenuates hypoxic pulmonary artery hypertension development presumably through activation of BKCa channels. These results highlight the potential use of CO as a novel avenue for research on the treatment of pulmonary artery hypertension (PAHT) in a similar manner to another gasotransmitter, nitric oxide.

  • Pulmonary artery hypertension
  • Calcium-activated potassium channel
  • Carbon monoxide
  • Chronic hypoxia

1. Introduction

It is known that prolonged exposure to alveolar hypoxia (chronic hypoxia) induces pulmonary vascular remodeling which results in the development of pulmonary artery hypertension (PAHT) [1]. A hallmark of PAHT is an increase in pulmonary vascular resistance with sustained pulmonary artery vasoconstriction via depolarization of pulmonary artery (PA) myocytes [2]. This hypoxia-induced membrane depolarization is mainly due to a decrease in expression of α-subunit proteins of voltage-activated K+-channels (Kv) [3,4] and of voltage and Ca2+-activated K+ (BKCa) channels activity [5]. Indeed, the resting membrane potential (Em) of PA myocytes is largely controlled by these K+-channels [6–8] of which the opening tends to hyperpolarize the membrane.

It has been suggested that therapies targeting K+-channels expression may have potential value for the treatment of vascular diseases [9]. PA myocytes have a high input resistance in the resting state and opening of even a few number of K+-channels can cause a substantial hyperpolarization/repolarization effect on the membrane, resulting in modification of arterial tone. BKCa channels, which are expressed in high density in PA myocytes, act to hyperpolarize membrane in response to an increase in intracellular Ca2+ concentration [10]. It has recently been shown that an increase in the expression of these channels could induce compensatory vasodilatory reactions in systemic hypertension [11] and that activation of BKCa channels could reverse acute hypoxic pulmonary vasoconstriction [12].

Carbon monoxide (CO) is an endogenously generated gas as well as an ubiquitous environmental pollutant. The predominant biological source of CO is from degradation of heme by heme oxygenase (HO) [13]. At least two isoforms of HO, an inducible HO-1 and a constitutive isoform HO-2, have been found in vascular myocytes [14]. In PA myocytes, HO-1 is transiently increased by chronic hypoxia [15] which prevents the development of hypoxic PAHT [16,17]. Most experiments investigating the effects of endogenously generated CO on vascular tone, under physiological or pathological conditions, used activators or blockers of HO to deduce the role of endogenous CO [18]. Zhang et al. [19] demonstrated that an IbTx-sensitive current was decreased after inhibition of HO, suggesting that endogenous CO could activate vascular BKCa currents. Recently, we have observed that chronic exogenous CO exposure could directly decrease the tone of coronary and PA [20,21]. Furthermore, we demonstrated that in small PA myocytes chronic exogenous CO also activates BKCa currents [21]. Consistent with these observations, we demonstrated that chronic CO exposure of hypoxic rats increased the relaxation induced by acute CO of PA smooth muscle, a mechanism linked to an increase in K+-channels activity [22].

Since sustained activation or overexpression of HO-1 prevents the development of hypoxic PAHT and that chronic exogenous CO activates BKCa channels which could induce compensatory vasodilation in hypertension, we have examined in this study the possibility of chronic treatment with low concentrations of CO to reduce chronic hypoxic PAHT by activating BKCa channels in PA myocytes.

2. Methods

2.1. Exposure to chronic hypoxia and carbon monoxide

Adult male Wistar rats (8–10 weeks, 220–270 g) were randomized into four groups. Two groups were housed in room air at a normal atmospheric pressure (101 kPa) without (normoxic rats) or with 50 ppm of CO (CO rats) during 21 days. Two other groups were subjected for 21 days to chronic hypoxia condition without (hypoxic group; at pressure of 50.5 kPa) or with 50 ppm CO (hypoxic-CO group) as previously described [22].

2.2. Hemodynamic measurements

Heart rate was recorded. PAHT was assessed by measuring the PA pressure, the ratio of right ventricle (RV) to left ventricle plus septum (LV+S) weights, and hematocrit as previously described [23]. PA pressures are expressed as calculated mean PA pressure (MPAP). The investigations were carried out in agreement with the Guide for the Care and Use of Laboratory Animals published by the US National Institute of Health (NIH Publications No. 85-23, revised 1996) and European Directives (86/609/CEE).

2.3. Membrane potential in pressurized resistance arteries

Resistance PA with internal diameter less than 150 μm were isolated and prepared for membrane potential recording as previously described [21]. Endothelium was removed by air bubble passing inside the artery. Intraluminal pressure was increased to 14, 14, 36, and 25 mm Hg in resistance PA isolated, respectively, from normoxic, CO, hypoxic, and hypoxic-CO rats. These pressure settings were in line with the mean pulmonary arterial pressure calculated in the corresponding animal group. Membrane potentials were recorded in situ using glass sharp microelectrodes filled with KCl 3 M solution. PSS at 37 °C was used for both continuously superfusing (5 ml/min) and for the perfusion of the arteries. After 10 min of perfusion, the stopcock was closed and the intraluminal pressure was slowly increased. The different levels of pressure were maintained during a 60-min equilibration period before experiments. Response to iberiotoxin (IbTx, 100 nM), a selective blocker of BKCa [24], was recorded by changing three times the volume of the vessel chamber with IbTx-containing solution and then after stopping the superfusion.

2.4. Electrophysiology

Enzymatic isolation of intrapulmonary resistance arteries with internal diameter less than 100 μm was performed according to published method [25]. Electrophysiological recordings were obtained using the whole-cell patch-clamp and the inside-out configurations of Hamill et al. [26].

Whole-cell macroscopic K+ currents were generated by stepwise 10 mV depolarizing pulses (400-ms duration; 5-s intervals) from a constant holding potential of −80 mV up to +60 mV. Signals were filtered at 1 kHz and digitized at 5 kHz. The steady state current elicited at chosen membrane potential was calculated as the average of the current recorded during the latest 50 ms of the pulse. The IbTx-sensitive current was defined as the difference between outward currents recorded in drug-free bath solution and after superfusion with 100 nM IbTx. Em was measured in current clamp mode (I=0) just after the disruption of the patch membrane. The input resistance (RIN) was estimated as the slope of the IV curves between −80 and −60 mV where no dynamic currents were activated. The input capacitance (CIN) was determined by dividing the integral of the capacitive current by amplitude of a 10 mV voltage step from −80 mV [21].

The pipette solution contained (in mM): glutamic acid, 125; KCl, 20; Na2ATP, 1; CaCl2, 0.37; MgCl2, 1; HEPES, 10; EGTA, 1. pH was adjusted to 7.2 using KOH. pCa∼7 was calculated by a computer program developed by Godt and Lindley [27]. Physiological Salt Solution (PSS) contained (in mM): NaCl, 138.6; KCl, 5.4; CaCl2, 1.8; MgCl2, 1.2; NaH2PO4, 0.33; HEPES, 10 and glucose, 11. pH was adjusted to 7.4 using NaOH.

Unitary BKCa currents were obtained using pipettes with a resistance of 10–15 MΩ in inside-out configuration with symmetrical 140 mM KCl at different membrane potentials ranging from −60 to +150 mV. Currents were filtered at 5 kHz and sampled at 50 kHz and the activity of BKCa channels were recorded for periods of time ranging from 1 to 3–5 min according to the membrane potential and the pCa.

The intrapipette solution contained (in mM): 140 KCl, 10 HEPES, 5 EGTA, 1 MgCl2 and pH was adjusted at 7.4 using KOH. The bath solution (internal surface) contained (in mM): 140 KCl, 10 HEPES, 5 EGTA, pH was adjusted at 7.2 using KOH and various concentrations of CaCl2 and MgCl2 were added in order to obtain calculated free Ca2+ solutions of 10 nM, 100 nM, and 1 μM [27]. In some experiments, pipette tips were loaded with drug-free pipette solution and then back-filled with pipette solution containing 100 nM IbTx and unitary currents were recorded at the same membrane potential for 5 to 10 min to permit drug diffusion to the outside patch surface.

The number of channels in each patch was assessed by switching the pCa from 8 to 6 for a Vm of +150 mV. All patches containing more than five channels have been withdrawn from the analysis. The unit amplitude of BKCa channels was determined from all point histograms fitted with Gaussian curves.

The open probability (Po) was calculated from the ratio between the open time and the total time, and when channel activity was produced by more than one channel in the patch the current levels were calculated by fitting the histograms with multiple Gaussian curves and the Po was calculated according to the equation of Selyanko and Brown [28] which assumes that all channels in the patch have the same Po and that the gating of each channel occurs independently of each other NPo=(Σ(j=1; N)tjj)/T [28]. tj is the time spent at each current level (j=0, 1, 2,…, N), T is the duration of the recording and N is the number of channels in the patch. To confirm these results we also performed “threshold analysis” of single-channel events on records using Fetchan software (Axon instrument) to ensure that our estimates of Po were not in error. PoVm relations were fitted by a Boltzmann function (Po=Pmax/[1+exp(VmV1/2)/k]), where Pmax is the maximum Po obtained at strongly depolarized membrane potentials, Vm is the membrane potential, Po is the open channel probability, and V1/2 is the half-activation potential. k represents the slope factor, which is an indicator of single-channel voltage sensitivity. The V1/2 value have been estimated using the experimental data corresponding to the exponential growth part of the sigmoidal function for very low Ca2+ concentrations which do not allow to obtain high Po levels in normoxic and hypoxic groups even for depolarized membrane potentials.

Voltage-dependent activation relationships at pCa other than 8, 7, and 6 have been calculated according to Carl et al. [29]. Voltage and Ca2+ dependent parameters were determined: V1/2, half-maximal Po for a given [Ca2+]i; ΔV1/2, V1/2 shift for 10-fold increase in [Ca2+]i; C1/2 [Ca2+]i for half-maximal Po at a given voltage; Ca0 or set-point, [Ca2+]i for half-maximal Po at voltage V=0 mV and Hill coefficient (h). A linear relationship between pCa and V1/2 (V1/2V1/2(log10[Ca2+]i−log10[Ca0])) allow to calculate V1/2 and C1/2 (C1/2=Ca010(V/V1/2)). Using the relations Po=1/{1+exp[(V−ΔV1/2log10(Ca2+/Ca0))/k]} and Po=1/[1+(Ca2+/C1/2)h] (Hill equation), h was calculated as hV1/2/(kln10). Defining ΔVe as the shift in V1/2 for an e-fold increase in [Ca2+]i then ΔVe=Kh. This last equation is used to calculated Po as a function of voltage and Ca2+ as follow: Po=1/{1+exp[(V/k)−(hln(Ca2+/Ca0))]}. This equation assumes a linear relationship between V1/2 and pCa. It has been reported in some studies a deviation from linearity for very high or low [Ca2+]i which would need, if confirmed, to be added to these equations [30,31].

Voltage clamp protocols were generated and the data were captured with a computer using a Digidata 1200 interface, Axopatch 200 amplifier and pClamp 8 software (Axon Instruments, USA). The analysis was carried out using Clampfit 8, Fetchan 8, PSTAT 8, and Origin 6 softwares (Microcal Software, Northampton, USA).

2.5. Solutions and drugs

Stock solutions of IbTx were prepared in distilled water and then diluted to an appropriate concentration. All chemicals were from Sigma (St. Quentin Fallavier, France).

2.6. Statistics

Results are expressed as mean ± standard error (S.E.M.). Statistical analyses were made with the unpaired Student's t-test or the Mann–Whitney test when normality test (Anderson-Darling) failed. For comparison between more than two means, we used analysis of variance followed by Dunnett's test or Mood-Median test when normality test failed. Differences were considered significant when p<0.05. Statistical analysis was realized using Minitab software (Minitab). N (number of animals) and n (number of cells) values were, respectively, used for in vivo (Hemodynamic measurements) and in vitro (electrophysiology) statistical testing.

3. Results

3.1. Chronic CO treatment attenuates the hemodynamic changes induced by chronic hypoxia

MPAP was higher in rats after 3 weeks of exposure to hypoxia but, in contrast, the hypoxic animals exposed to CO developed less PAHT (Table 1). Right ventricular hypertrophy, a hallmark of PAHT resulting from right ventricle pressure overload, was also measured. Whereas RV/(LV+S) was significantly increased in hypoxic rats, this increase was significantly attenuated in hypoxic rats exposed to chronic CO (Table 1). The increase in RV/(LV+S) ratio was entirely due to the increased RV weight because (LV+S) weight was not different between groups (Table 1). When animals were exposed to CO alone, no significant changes in heart rate, pulmonary pressure, and right ventricular weight were observed (Table 1). All these data demonstrate that chronic CO treatment attenuates the hemodynamic changes induced by chronic hypoxia while CO treatment alone did not alter the hemodynamics of the rats.

View this table:
Table 1

Hemodynamics parameters measured in normoxic, hypoxic, CO and hypoxic-CO rats

HR (bpm)332 ± 28 (N=14)337 ± 30 (N=9)330 ± 30 (N=7)405 ± 25 (N=7)
MPAP (mm Hg)12.7 ± 1.2# (N=14)36.3 ± 3.3*& (N=9)14.0 ± 1.6 # (N=7)24.8 ± 1.6*#& (N=7)
RV/(LV+S) (%)32.3 ± 0.6# (N=29)61.6 ± 2.0*& (N=13)29.5 ± 2.2# (N=13)52.2 ± 1.3*#& (N=20)
(LV+S)/BD (‰)2.07 ± 0.04 (N=29)2.36 ± 0.02* (N=13)2.03 ± 0.06 (N=13)2.17 ± 0.07 (N=20)
RV/BD (‰)0.66 ± 0.01# (N=29)1.44 ± 0.10*& (N=13)0.58 ± 0.03# (N=13)1.14 ± 0.06*# & (N=20)
Hct (%)41.5 ± 1.5# (N=29)71.3 ± 3.4*& (N=13)51.2 ± 1.1*# (N=13)63.4 ± 1.3*# & (N=20)
  • HR: heart rate; MPAP: mean pulmonary artery pressure; RV: right ventricle weight; LV+S: left ventricle plus septum weight; BD: body weight; Hct: hematocrit. p<0.05 vs. normoxic group*; vs. hypoxic group# and vs. CO group&.

3.2. Chronic CO treatment prevents the depolarization induced by chronic hypoxia

Membrane potential recording demonstrated that chronic hypoxia significantly depolarized pressurized resistance PA myocytes and after chronic CO exposure this depolarization was reversed to a level comparable to that of normoxic rats (Fig. 1). Exposure to CO alone lead to hyperpolarization of the membranes of myocytes (Fig. 1A and B). As shown in Fig. 1A, IbTx had no effect on Em of normoxic and hypoxic arteries whereas it depolarized membranes of PA myocytes from hypoxic-CO and CO rats. Fig. 1C revealed no significant difference between Em recorded before and after IbTx in normoxic and chronic hypoxic arteries whereas IbTx significantly depolarized the membrane (p<0.05, post hoc Dunnett's test) in rats treated with chronic CO.

Fig. 1

Effect of IbTx on resting membrane potential of normoxic, hypoxic, CO and hypoxic-CO pressurized artery's myocytes. (A) Example of resting membrane potential recording in a pressurized artery before and after action of IbTx. (B) Histograms showing mean values of resting membrane potentials recorded from different groups of rats. (C) Effect of 100 nM IbTx on resting membrane potential. Data are expressed as mean ± S.E.M.; * indicates significant difference compared to normoxic rats and # to hypoxic rats (p<0.05, post hoc Dunnett's test).

3.3. Chronic CO treatment prevents the effect of hypoxia on passive membrane properties of PA smooth muscle

No statistically significant differences were observed between mean CIN values (Table 2) of cells isolated from all groups of rats. While hypoxia significantly depolarized (p<0.05, post hoc Dunnett's test) the membranes of isolated PA myocytes, cells from hypoxic-CO rats showed no significant changes in Em compared to values obtained in normoxic rats. Furthermore, control rats exposed to chronic CO exhibited a hyperpolarization of their membranes. In the same way, whereas RIN was significantly higher (p<0.05, ANOVA and post Dunnett's test) in cells from hypoxic rats compared to normoxic rats, Em was not significantly different between cells from normoxic rats and hypoxic-CO rats (Table 2).

View this table:
Table 2

Passive membrane properties measured in normoxic, hypoxic, CO and hypoxic-CO PA myocytes

n, N23, 1213, 711, 716, 7
CIN (pF)10.8 ± 0.98.2 ± 1.09.0 ± 0.78.1 ± 0.9
Em (mV)−44.3 ± 2.0#−32.5 ± 2.5*−52.1 ± 3.0*#−39.6 ± 2.1#
RIN (GΩ)4.3 ± 0.3#6.3 ± 0.5*3.2 ± 0.6*#4.7 ± 0.4#
  • CIN: input capacitance; RIN: input resistance; Em: resting membrane potential. p<0.05 vs. normoxic group* and vs. hypoxic group#.

3.4. Chronic CO treatment prevents the hypoxic decrease of outward potassium currents by increasing an IbTx-sensitive current

Fig. 2A shows typical examples of whole-cell outward currents recorded in normoxic, hypoxic, CO and hypoxic-CO PA myocytes before application of 100 nM IbTx and the 100 nM IbTx-sensitive currents. Compared with normoxic PA myocytes, hypoxia decreased the net whole-cell K+ current (Fig. 2A). In contrast, when rats were exposed to chronic CO, hypoxia did not decrease the net outward current and K+ current exhibited a higher sensitivity to 100 nM of IbTx (Fig. 2A). Fig. 2B shows that the current density in hypoxic cells was significantly decreased (p<0.05, ANOVA, post hoc Dunnett's test) compared with cells isolated from normoxic animals. When rats were exposed to CO, with or without chronic hypoxia, current densities of cells were significantly increased (p<0.05, ANOVA, post hoc Dunnett's test). Further analysis revealed that hypoxia caused a significant reduction of IbTx-sensitive current (p<0.05, Mood's Median test) (Fig. 2C). In contrast, in hypoxic-CO and CO cells IbTx-sensitive currents were markedly increased (p<0.05, Mood's Median test) (Fig. 2C).

Fig. 2

Effect of IbTx on whole-cell currents in PA myocytes from normoxic, hypoxic, CO-treated, and hypoxic-CO treated rats. (A) Family of currents elicited by incremental 10-mV depolarizing steps from −80 to +60 mV in all population of cells. IbTx (100 nM) blocked a small component of outward current in normoxic and hypoxic myocytes. In contrast, the IbTx-sensitive component was increased in hypoxic-CO and CO cells. (B) Current density–voltage relation of net outward current amplitudes in myocytes dissociated from the four different groups of rats. (C) Current density–voltage relationships showing that IbTx-sensitive current was significantly smaller in hypoxic cells but greater in hypoxic-CO and CO cells compare to normoxic cells. Results represent the mean ± S.E.M. *p<0.05 indicates significant difference compared to normoxic cells.

3.5. Chronic CO treatment increases BKCa channel activity in PA myocytes of hypoxic rats by increasing single-channel conductance and Ca2+ sensitivity

Whether the chronic CO treatment increased IbTx-sensitivity of membrane current in hypoxic rats was due to an increased BKCa channels activity was investigated using single-channel experiments. Fig. 3A illustrates that the unitary amplitudes of single-channels current were higher in all cells exposed to chronic CO. The resulting current–voltage relation (Fig. 3B) demonstrates that chronic hypoxia induced no significant changes in single-channel conductances of BKCa channel, 226 ± 3 pS in normoxic cells (n=8, N=4) and 228 ± 6 in hypoxic cells (n=8, N=4). The single-channel conductances were significantly increased in hypoxic-CO cells (n=9, N=4) to 292 ± 7 pS and in CO cells to 266 ± 5 pS (n=9, N=4) (p<0.05, ANOVA, post hoc Dunnett's test, Fig. 3B). Single-channel conductances of BKCa channels in all groups were not modified by changing pCa (data not shown). In all groups, the high amplitude current was blocked by 100 nM IbTx, confirming the nature of the current being BKCa (Fig. 3C). The voltage-activation curves of BKCa current at different [Ca2+]i (Fig. 3D) clearly shows a parallel shift along the voltage axis to more negative values of the Po of BKCa after chronic CO exposure which is attained without modification of the slope. This shows that chronic CO exposure does not modify the effective valence of the voltage gating process of the BKCa. Fig. 4 illustrates the relation between normalized Po, Vm and [Ca2+]i for physiological membrane potentials.

Fig. 4

Voltage and Ca2+ dependence of BKCa channel activation. The voltage-dependent activations of BKCa at pCa 8, 7 and 6 were determined for membrane potentials ranging from −60 up to +150 mV in myocytes isolated from normoxic (top left), CO (top right), hypoxic (bottom left) and hypoxic-CO (bottom right) rats. These relations were fitted using the Boltzman relation to determine the slope factor (k) and V1/2. In order to better show the modification of the behavior of BKCa channels in all populations of rats, we choose to only show data obtained from −40 up to +40 mV. The voltage-dependent activations at pCa other than 8, 7 and 6 were calculated (open diamonds) according to Carl et al. [29] (see Methods for the calculations). All the experimental values are derived from four to five animals and eight to nine cells.

Fig. 3

Single-channel current recording of BKCa current in normoxic, hypoxic, CO and hypoxic-CO cells. (A) Family of currents elicited at incremental membrane potentials in all population of cells. C indicates the close state of the channel. (B) Unitary current amplitude–voltage relation showing variation of channel conductance in myocytes dissociated from CO and hypoxic-CO compared to normoxic and hypoxic rats. A linear fit was performed in the four populations of cells in order to calculate single-channel conductance. (C) IbTx (100 nM) blocked BKCa unitary currents in all groups, elicited at +60 mV membrane potential with 100 nM [Ca2+]i. Results represent the mean ± S.E.M. (D) Single channel open probability of BKCa for the four populations for calcium concentrations of 1, 0.1 and 0.01 μM.

While hypoxia with or without CO inhalation has no significant effect on k (14.0 ± 0.9 mV in hypoxic cells, 13.9 ± 3.2 mV in hypoxic-CO cells, and 12.9 ± 1.1 mV in CO cells) compared to normoxic cells (k=12.6 ± 1.0 mV), the Ca2+ sensitivity of K+-channels are significantly modified (Table 3). While hypoxia tends to decrease the ΔV1/2 and h, these parameters are increased in hypoxic-CO and CO cells. Furthermore, BKCa in PA cells exposed to chronic CO, with or without hypoxia, showed significantly increased ΔV1/2 and h compared to normoxic rats. While the set point was increased in hypoxic cells, it was decreased in hypoxic-CO and CO cells and this decrease bring it to values significantly lower than those observed in normoxia (Table 3). These results indicate that while the voltage sensitivity was not altered by chronic hypoxia or CO exposures the Ca2+ sensitivity was decreased after chronic hypoxia without CO but increased if CO was added at the same time.

View this table:
Table 3

Ca2+-dependent parameters of BKCa calculated in normoxic, hypoxic, CO and hypoxic-CO PA myocytes

n, N8, 48, 49, 49, 4
ΔV1/2 (mV)−54.9 ± 4.0−44.9 ± 4.1−63.2 ± 2.8#−80.0 ± 2.7*#
h1.89 ± 0.261.39 ± 0.262.33 ± 0.14*#2.49 ± 0.27*#
Set point (μM)11.11 ± 2.767.49 ± 16.3 *1.42 ± 0.35*#0.68 ± 0.05*#&
  • ΔV1/2: leftward shift of V1/2 for increase of one pCa in physiological range; h: Hill coefficient; “Set point”: [Ca2+]i inducing 50% activity of the channel at membrane potential of 0 mV. See Methods for the calculation of the parameters. p<0.05 vs. normoxic group*, vs. hypoxic group# and vs. CO group&.

4. Discussion

This study demonstrates that chronic inhalation of low concentration of CO attenuated the development of PAHT and increased BKCa channels activity of hypoxic PA myocytes. The concentration of CO used (50 ppm) is physiologically relevant to external chronic CO exposure. Levels ranged from 10 to 50 ppm were found for pedestrians and street workers in Toronto [32] and a cigarette smoker may be exposed to 400–500 ppm CO [33].

Different mechanisms could explain the attenuation of PAHT by chronic exogenous CO exposure. These mechanisms include a decrease of hypoxic cardiac output and/or of hypoxic vasoconstriction. Nevertheless, we observed no change of cardiac output for rats exposed to CO during 3–10 weeks at 50 ppm [34] and a slight increase for rats exposed to hypoxia at the same time than CO. Numerous publications present evidences that acute CO [18] and chronic CO [20,35–37] are general vasodilators acting directly on vascular myocytes. In pulmonary circulation, acute exogenous CO was shown to decrease vascular resistance [38] and to inhibit hypoxic pulmonary vasoconstriction [39,40]. Almost no experiments were performed using chronic CO on PA. Recently, we demonstrated that acute CO application relaxed endothelium-free PA rings [22]. The sensitivity to CO of PA obtained from hypoxic rats was decreased [22] while it was increased if 50 ppm chronic exogenous CO was added simultaneously to hypoxia.

The mechanisms of CO-induced vasodilation include soluble guanylyl cyclase (sGC)-dependent and sGC-independent mechanisms [18]. Among the sGC-independent mechanisms, acute CO could directly activate BKCa in vascular myocytes [41] by a specific interaction with α-subunit [42] or by increasing the effective coupling between Ca2+ sparks and the channels [43]. We also demonstrated that chronic CO activates an IbTx-sensitive current, probably BKCa [21]. Pulmonary and systemic hypertension are associated with a decreased BKCa channel activity in myocytes [5,44,45]. Thus, a mechanism which could increase the activity of BKCa channels and/or prevent the decrease of BKCa activity in hypertension may contribute to correcting vascular PA dysfunction. We demonstrated that chronic exogenous CO slightly increased single-channel conductance of BKCa and induced a large increase of the sensitivity of BKCa channels to Ca2+ of PA myocytes from normoxic and hypoxic rats. This increase of BKCa activity had no effect on MPAP of normoxic rats but decreased the PAHT. This absence of effect of chronic CO on normoxic rats could be explained by other and opposite effects of CO which could induce mainly an increase of mean PA pressure. For example, CO could inhibit endothelial NO synthase activity [14] favoring contraction. Then, using endothelium-free pressurized PA vessels and isolated PA myocytes, chronic CO hyperpolarized normoxic PA myocytes membranes and prevented the hypoxic depolarization of PA myocytes membranes. We demonstrated that IbTx depolarized membranes of PA myocytes from hypoxic-CO and CO rats but not those from normoxic and hypoxic rats. Thus, BKCa channels played a much more important role in controlling Em of PA myocytes after exposure to CO. Nevertheless, in this preparation the main currents regulating the membrane potential are IKv and IKN currents, this latter could be due to a background activity of a two-pore domain acid sensitive potassium channel TASK-1. Chronic hypoxia, which decreases the amplitude of both currents, induces a depolarization [6,46,47]. This could explain why the Em of PA myocytes from CO rats is more negative than in hypoxic-CO rats. Furthermore, we reported that chronic hypoxia decreased BKCa activity by decreasing Ca2+ sensitivity (increase of the set point), but not voltage sensitivity, of BKCa channels. A previous study reported the decreased BKCa activity in PA myocytes after chronic hypoxia [5]. In agreement with Peng et al. [5], we found that chronic hypoxia had no effect on single-channel conductance of BKCa, suggesting no change in the permeation of this channel. In contrast, when animals were exposed to CO the single-channel conductance of BKCa was slightly increased. Thus, chronic CO treatment might induce a change in some amino acid residues of α-subunit involved in the permeation properties of BKCa [48]. Nevertheless, this increase of single-channel conductance would not be functionally important and the main effect of chronic CO was to increase the sensitivity of BKCa channels to Ca2+ compared to the normoxic group. Since the CO-induced increase of Ca2+ sensitivity of BKCa channels was greater than the hypoxia-induced decrease in BKCa activity, we observed a large increase of IbTx-sensitive currents responsible of the prevention of the hypoxic depolarization of membrane of pressurized PA vessels. The IbTx-sensitive current amplitude is smaller in hypoxic-CO PA myocytes than in CO PA myocytes. Since this current showed an inactivation in hypoxic-CO group but not in the CO group, this can explain part of the difference. Indeed, only the steady current measured at the end of the pulse were used to build the current density–voltage relation. This inactivation of IbTx-sensitive current in hypoxic-CO group but not in CO group could be explained by a differential co-expression of BKCa channel α-subunits with β-subunits. Indeed, it was demonstrated that the expression of β3b-subunits induces a partial inactivation of BKCa currents [49] and a possible involvement of this subunit may be envisaged as β3b mRNA has been found in the lung but remains to be confirmed in the future [49]. The activity of BKCa channels are regulated by pH, phosphorylation, divalent cations and polyamines too [50,51]. Thus, the observed activity of BKCa channels which is larger in hypoxic-CO PA compared to CO PA is not really surprising if we consider that in the inside-out configuration all the regulating compounds were lost. Furthermore, the effect of chronic CO is even more complicated if we consider that β3b-subunits is regulated by phosphorylation too [52].

Chronic CO treatment appears to have different effects from acute CO treatment in which there is no change in unitary current amplitude [41]. This difference could be explained by different CO application periods (acute vs. chronic) and the involvement of BKCa channel expression. On the other hand, no matter with acute or chronic CO treatment the gating and Ca2+ sensitivity of BKCa channels were consistently changed in vascular myocytes. Interestingly, the hill coefficient was also increased in both acute and chronic CO experiments to values around 3 [41] suggesting a conformational change of the channel protein. The increased BKCa activity in hypoxia-CO compared to normoxia may be explained by CO-induced prevention of hypoxic effect on channel as well as an increased permeation property and Ca2+ sensitivity of BKCa channels. The synchronization of these effects may explain the beneficial effect of chronic CO treatment on hypoxic PA myocytes. The possibility that a differential co-expression of BKCa channel α-subunits with β-subunits as well as an increased level of expression of the α-subunits might also be induced by chronic CO exposure [42,53] should be entertained in the future.

Although the link between the chronic-CO-induced BKCa channels increase in activity and the reduction of the hypoxia-induced pulmonary hypertension after CO exposure has not been directly assessed using “in vivo” recordings of pulmonary artery pressure while blocking the BKCa channels, our results may have a functional and therapeutical impact concerning the use of very low concentration of CO inhalation in the treatment of PAHT. Exogenous application of low concentration of CO has been shown to have benefic effects on hypoxic PA structural remodeling [37], ischemic lung injury [54], and arteriosclerotic lesions [55]. Our present study outlines an innovative notion that CO may be a novel therapeutic agent for PAHT in a similar manner to NO [56].


This work was funded by «Agence De l'Environnement Et de la Maîtrise d'Energie » (ADEME; 98 93 029) and «la Fondation Simone et Cino Del Duca ».

We thank Nadine Gaudin for excellent secretarial support and Jean-Pierre Moisan, Stephanie Martin and Valerie Schubnel for technical assistance. Eric Dubuis holds a doctoral fellowship from ADEME and Conseil Regional du Centre.


  • 1 Present address: INSERM E-0211 Nutrition, Croissance, Cancer, Faculté de Médecine, 2 bis Boulevard Tonnellé, 37032 Tours, France.

  • Time for primary review 27 days


  1. [1]
  2. [2]
  3. [3]
  4. [4]
  5. [5]
  6. [6]
  7. [7]
  8. [8]
  9. [9]
  10. [10]
  11. [11]
  12. [12]
  13. [13]
  14. [14]
  15. [15]
  16. [16]
  17. [17]
  18. [18]
  19. [19]
  20. [20]
  21. [21]
  22. [22]
  23. [23]
  24. [24]
  25. [25]
  26. [26]
  27. [27]
  28. [28]
  29. [29]
  30. [30]
  31. [31]
  32. [32]
  33. [33]
  34. [34]
  35. [35]
  36. [36]
  37. [37]
  38. [38]
  39. [39]
  40. [40]
  41. [41]
  42. [42]
  43. [43]
  44. [44]
  45. [45]
  46. [46]
  47. [47]
  48. [48]
  49. [49]
  50. [50]
  51. [51]
  52. [52]
  53. [53]
  54. [54]
  55. [55]
  56. [56]
View Abstract