Restoration of the response of the middle cerebral artery of the rat to acidosis in hyposmotic hyponatremia by the opener of large-conductance calcium sensitive potassium channels (BK_Ca)

Abstract

Hyposmotic hyponatremia (the decrease of extracellular concentration of sodium ions from 145 to 121 mM and the decrease of hyposmolality from 300 to 250 mOsm/kg H₂O) impairs response of the middle cerebral artery (MCA) to acetylcholine and NO donor (S-nitroso-N-acetyl-DL-penicillamine). Since acidosis activates a similar intracellular signaling pathway, the present study was designed to verify the hypothesis that the response of the MCA to acidosis is impaired during acute hyposmotic hyponatremia due to abnormal NO-related signal transduction in vascular smooth muscle cells. Studies performed on isolated, cannulated, and pressurized rat MCA revealed that hyposmotic hyponatremia impaired the response of the MCA to acidosis and this was associated with hyposmolality rather than with decreased sodium ion concentration. Response to acidosis was restored by the BK_Ca but not by the K_ATP channel activator. Patch-clamp electrophysiology performed on myocytes freshly isolated from MCAs, demonstrated that hyposmotic hyponatremia does not affect BK_Ca currents but decreases the voltage-dependency of the activation of the BK_Ca channels in the presence of a specific opener of these channels. Our study suggests that reduced sensitivity of BK_Ca channels in the MCA to agonists results in the lack of response of this artery to acidosis during acute hyposmotic hyponatremia.

Introduction

Hyponatremia is a relatively common disturbance of the water-electrolyte balance observed in medical practice. It is defined as a decrease in plasma sodium concentration below 135 mM.^1,2 This disorder occurs in many diseases such as: hypothyroidism, liver cirrhosis, renal and heart failure, cancer and central nervous system disorders.³ However, it is most dangerous for neurosurgical patients, in whom hyponatremia is often diagnosed after: subarachnoid hemorrhage, head trauma, intracerebral hemorrhage, and encephalomeningitis.^1,4 Hyponatremia may occur with low, normal, or high plasma osmolality.¹ In neurological disorders, hyponatremia is usually associated with hyposmolality and is a result of either inappropriate water retention or excessive sodium excretion. The resulting brain swelling is considered to be the main consequence of hyposmotic hyponatremia in those patients.¹

Accordingly, previous studies on the effect of hyponatremia on the brain mainly focused on the issues related to hyposmotic brain edema and the complications associated with the treatment of hyponatremia.^5,6 There is, however, little known about the effect of hyponatremia on the function of cerebral blood vessels. In a few studies, the hyposmotic challenge was used as a tool to activate mechanosensitive channels in endothelium-denuded cerebral blood vessels or in isolated smooth muscle cells.^7–9 In other studies concerning the effect of reduced sodium ion concentration on the myogenic tone of cerebral blood vessels, the extracellular concentration of sodium ions was reduced below 100 mM or sodium ions were eliminated.^10–12 These results are not applicable to clinically relevant hyponatremia. Patients who suffer from a decreased concentration of sodium ions in the plasma below 100 mM are in a critical state and often do not survive such low levels of natremia.¹

Neurological symptoms of hyponatremia depend on the level of natremia as well as the rate of its development. Moderate to severe hyponatremia (Na⁺<120 mM), developing in less than 48 h, is defined as acute and is characterized by headache, vomiting, lethargy, seizure, and coma which are typical signs of brain edema.¹³

Our previous study performed on the isolated, intact (i.e. with endothelium) middle cerebral artery (MCA) of the rat, subjected to acute moderate to severe hyposmotic hyponatremia in vitro, showed a lack of the normal response of this vessel to acetylcholine and nitric oxide (NO) donor which speaks in favor of the abnormality of NO-related signal transduction in vascular smooth muscle in hyponatremic conditions.¹⁴ The more so because administration of a NO synthase inhibitor caused similar vasoconstriction of the MCA in normonatremia and acute hyponatremia, indicating that endogenous NO production might not be disordered in acute hyposmotic hyponatremia.¹⁴

It is well known that relaxation of vascular smooth muscle by NO depends on: (i) stimulation of cytosolic guanylyl cyclase and the rise of the cGMP concentration; and (ii) activation of large-conductance Ca²⁺ sensitive K⁺ channels (BK_Ca) independently of cGMP.¹⁵ BK_Ca channels act as a buffer limiting vasoconstriction due to: stretching of the vessel wall, smooth muscle cell depolarization, increase in the intracellular concentration of Ca²⁺ ions and in response to increased perfusion pressure.^16,17 BK_Ca channels are also essential for the dilation of cerebral blood vessels during acidosis as reported by Lindauer et al.¹⁸ who have shown that relaxation of the rat MCA in response to the increase of hydrogen ion concentration requires activation of both BK_Ca and K_ATP channels in vascular smooth muscle.

The present study aimed to test the hypothesis that the response of the MCA to acidosis is impaired during acute moderate to severe hyposmotic hyponatremia due to abnormal NO-related signal transduction in vascular smooth muscle cells.

To this end, the reactivity of isolated intact rat MCA with endothelium to:

1. increased H⁺ concentration;

2. membrane permeable analog of cGMP, 8Br-cGMP;

3. activator of K_ATP channels and increased H⁺ ion concentration with K_ATP channels activator in the background;

4. inhibitor of BK_Ca channels;

5. activator of BK_Ca channels and increased H⁺ ion concentration with BK_Ca channels activator in the background;

6. increasing concentrations of BK_Ca channels activator was studied in normonatremia and in acute hyposmotic hyponatremia of severity compatible with clinically relevant disturbances of sodium ion homeostasis. In addition, the effect of equalization of the osmolality and chloride ion concentration on the response of the MCA to acidosis and NS1619 was tested in hyponatremia.

Moreover, the effect of acute hyposmotic hyponatremia on the electrophysiological characteristics of BK_Ca currents in smooth muscle freshly isolated from MCAs was studied using the patch-clamp method.

Materials and methods

Animals

Male Wistar rats (body weight 250–300 g, n = 83 animals) used in this study were supplied by the Animal House of Mossakowski Medical Research Centre, Warsaw, Poland. All animal experiments were performed in accordance with Law and Regulations on Animal Protection in Poland (Dz.U 2015/266) and were approved by the Extramural First Committee for the Care and Use of Laboratory Animals for Experimental Procedures, National Medicines Institute in Warsaw. The study complies with the ARRIVE guidelines for reporting animal research.

Isolated cerebral artery studies

The rats were anesthetized with 5% isoflurane in 70% NO₂/30% O₂ and decapitated. The brain was removed from the cranium and placed in a cold (4℃, pH = 7.4) physiological saline buffered with MOPS (3-(N-morpholino) propanesulfonic acid) (MOPS-PSS) containing: 3.0 mM MOPS, 144.0 mM NaCl, 3.0 mM KCl, 2.5 mM CaCl₂, 1.5 mM MgSO₄, 1.21 mM NaH₂PO₄, 0.02 mM EDTA, 2.0 mM sodium puryvate, 5.0 mM glucose and 1% dialyzed bovine serum albumin (BSA). Middle cerebral arteries were isolated and cleaned of adhering connective tissue. The arteries were transferred to the organ chamber filled with MOPS-PSS with 1% BSA, cannulated with a glass micropipette containing PSS with 1% BSA and secured with 10–0 nylon suture. The organ chamber was placed on the stage of an inverted microscope (CKX41, Olympus, Germany) equipped with a video camera and a monitor. Extraluminal fluid was replaced with MOPS-PSS without BSA, slowly heated to 37℃ and exchanged at a rate of 20 mL/min with the help of a peristaltic pump (Masterflex, Cole-Parmer, USA). The pressure-servo system connected to the inlet of the MCA was set to maintain transmural pressure at 80 mmHg. The vessel was continuously perfused at a slow rate of 100 µL/min with MOPS-PSS containing 1% BSA.

The vessels were allowed to equilibrate for 1 h to develop myogenic tone, i.e. a contraction by about 30–40% of the diameter measured after pressurizing. Smooth muscle cell function was tested by increasing extraluminal K⁺ ion concentration from 3 mM to 20 mM. Vessels, which did not dilate at least 20% in response to KCl, were discarded from further studies. Acute moderate to severe hyponatremia was induced in vitro by lowering the Na⁺ ion concentration from 145 mM to 121 mM in the intra- and extravascular fluid. The control group consisted of vessels placed in a buffer containing 145 mM Na⁺. All tested compounds were added to the extraluminal bath. NS1619 [1,3-Dihydro-1-[2-hydroxy-5-(trifluoromethyl)phenyl]-5-(trifluoromethyl)-2H-benzimidazol-2-one], pinacidil and paxilline were dissolved in ethanol. The final concentration of ethanol in the buffer was 0.1%. At this concentration, ethanol does not affect the diameter of the MCA as tested in a pilot experiment.

The effects of the tested compounds on the inner MCA diameter were assessed either 15 min (NS1619 and pinacidil) or 30 min (paxilline and 8Br-cGMP) following their administration. Acidosis was induced for 15 min by adding 2.5% HCl to the extravascular buffer which resulted in the decrease of pH from 7.4 to 7.0 but did not significantly affect extravascular concentration of chloride ions. Only one treatment protocol was carried out on the same vessel. At the end of the experiment, maximal dilation of the vessel was obtained in calcium-free MOPS-PSS containing EGTA (3 mM). The percentage dilation during vasodilatory tests was calculated from the equation: (D_active − D_baseline)/(D_maximum – D_baseline) × 100%, where D_active is the diameter obtained after drug administration, D_baseline is the diameter prior to drug administration, and D_maximum is the diameter in calcium-free MOPS-PSS with EGTA. Changes in the diameter of the MCAs in response to vasoconstrictor paxilline were calculated using the equation: (D_active/D_baseline) × 100%, where D_active is the diameter after paxilline administration and D_baseline is the diameter prior to drug administration.

Since induction of hyponatremia by decreasing NaCl concentration by 24 mM caused also a decrease in osmolality from 300 to 250 ± 2 mOsm/kg H₂O and a decrease in Cl⁻ ion concentration by 24 mM, two series of experiments, most relevant to the subject of the study, were repeated in isosmotic hyponatremia with standard concentration of Cl⁻ ions. To achieve equalization of buffer osmolality and Cl⁻ ions concentration, 24 mM HCl and 29 mM Tris were added to the hyponatremic buffer.

Seven series of experiments were performed:

Series I covered the response of the MCA to a decrease in pH from 7.4 to 7.0 in normonatremia and in both hypo- and isosmotic hyponatremia.

Series II included studies of the response of the MCA to a membrane permeable analogue of cGMP, 8Br-cGMP (10 µM) in normonatremia and in hyposmotic hyponatremia.

Series III covered response of the artery to K_ATP channel activator pinacidil (10 µM) and to a decrease in pH from 7.4 to 7.0 with pinacidil (10 µM) in the background in normonatremia and in hyposmotic hyponatremia.

In series IV, the participation of BK_Ca channels in the maintenance of basal MCA diameter was studied using an inhibitor of BK_Ca channels paxilline (1 µM) in normonatremia and in hyposmotic hyponatremia. To test whether the specific BK_Ca channel inhibitor affects the response of MCA to acidosis, combined administration of paxilline (1 µM) and MOPS with pH 7.0 was used in normonatremia and in hyposmotic hyponatremia.

Series V comprised studies of the response of the artery to BK_Ca channel opener NS1619 (10 µM) and to a decrease in pH from 7.4 to 7.0 with NS1619 (10 µM) in the background in normonatremia and in hyposmotic hyponatremia.

Series VI included studies of the response of the MCA to increasing concentrations of NS1619 (1 µM; 10 µM; 2 × 10 µM; 3 × 10 µM) in normonatremia and in hyposmotic hyponatremia.

In Series VII, effect of isosmotic hyponatremia on the response of the MCA to NS1619 (10 µM) was studied.

Isolated arterial smooth muscle studies

Isolated MCA myocytes were obtained according to the Holland et al.¹⁹ method. The brain was placed in a cold buffer containing: 136 mM NaCl, 10 mM HEPES, 5.6 mM KCl, 4.17 mM NaHCO₃, 2.6 mM CaCl₂, 1 mM MgCl₂, 0.44 mM NaH₂PO₄, 0.42 mM NaHPO₄, pH 7.4. Isolated MCAs were placed for 10 min in a buffer identical in composition to the one described above except for the Ca²⁺ ion concentration which was reduced to 0.1 mM. Vessels were subjected to a two-step digestion in a low calcium buffer at 37℃. In the first step, vessels were placed for 20 min in a solution composed of papain (1.5 mg/mL), dithioerythritol (2 mg/mL), and albumin (1 mg/mL). In the second step, artery segments were placed for 10 min in fresh buffer containing collagenase type F (1.5 mg/mL), hyaluronidase type I-S (1 mg/mL) and albumin (1 mg/mL). The digested artery was washed twice in isolation solution and triturated with a Pasteur pipette. The cells were stored on ice and used for experiments within 8 h.

Conventional patch-clamp technique was used to measure whole cell BK_Ca currents in isolated smooth muscle cells. Patch pipettes were pulled from borosilicate capillaries (O.D. 1.2 mm, Warner Instruments) with a micropipette puller and heat-polished in a microforge (Narishige, Japan). Pipette resistances were in the range of 3–7 MΩ. Whole cell currents were measured using an Axopatch 200B amplifier (Axon Instruments, USA). Data from the amplifier were filtered with a low-pass filter of 1 kHz and sent via an analog-to-digital converter (Digidata 1200 Series, Axon Instruments) directly to a computer. pClamp 7 software (Axon Instruments) was used to record and analyze membrane currents. After the onset of whole cell configuration, the holding potential was set at −20 mV. BK_Ca currents were induced by voltage steps in 20 mV increments from −40 mV to +60 mV of 1.2 s duration. To confirm that the observed currents were BK_Ca channels-dependent, a specific BK_Ca channel inhibitor paxilline (1 µM) was used. The pipette solution contained: 140 mM KCl, 10 mM glucose, 10 mM HEPES, 0.5 mM MgCl₂, 5 mM EGTA (adjusted to pH 7.2 with KOH). The bathing (external) control solution contained: 144 mM NaCl, 10 mM glucose, 10 mM HEPES, 6 mM KCl, 1 mM MgCl₂, 1.8 mM CaC1₂ (adjusted to pH 7.4 with NaOH). The bathing hyponatremic solution had NaCl concentration reduced to 121 mM which resulted in hyposmolality of 250 ± 2 mOsm/kg H₂O. Freshly isolated arterial smooth muscle cells were suspended in a control or a hyposmotic hyponatremic solution in poly-L-lysine-coated dishes (BD Biosciences, USA). The cells were incubated in the hyponatremic solution for 1 h at 4℃. After the onset of the whole cell configuration, the cell was perfused with the control or the hyposmotic hyponatremic solution stream using the perfusion pipette and a Rapid Solution Changer (Bio-Logic, France) and the control currents were recorded. Then the cell was perfused with normonatremic or hyposmotic hyponatremic solution with BK_Ca channel activator NS1619 (10 µM) accordingly. The recording was performed after 1 min and 5 min of perfusion in activating solution. All experiments were performed at room temperature. All reagents used in these experiments except for 8Br-cGMP (Fluka) were purchased from Sigma-Aldrich.

The concentration of the chemicals used in this study was chosen based on the available literature.^18–21

The osmolality of solutions was determined by freezing-point depression using a semi-micro osmometer (Osmomat 030, Gonotec, Germany).

Statistics

The statistical analyses were performed using Statistica 10 software. Results are expressed as means ± S.E.M. The data were analyzed using one-way ANOVA followed by Tukey’s multiple comparison tests, or paired or unpaired Student’s t-test, as appropriate. Differences with p < 0.05 were considered statistically significant.

Results

Effect of acidosis on MCA diameter

Reduction of extravascular pH from 7.4 to 7.0 led to a statistically significant dilation of the MCA in normonatremia by 33 ± 3% (p < 0.001). In contrast, in hyposmotic hyponatremia, there was no response of the MCA to acidosis (Figure 1). However, in isosmotic hyponatremia with standard Cl⁻ions concentration, the MCA response to acidosis normalized i.e. dilation by 45 ± 9 % (p < 0.05) was observed with no significant difference from the response in normonatremia.

Figure 1.

Response of the MCA to the reduction of pH from 7.4 to 7.0 in extravascular fluid in normonatremia (n = 6), in hyposmotic (n = 6) and in isosmotic (n = 4) hyponatremia. Dilation is expressed as a percentage of maximum diameter (0 Ca²⁺, EGTA 3 mM). Data are expressed as means ± S.E.M. **p < 0.001 between the groups.

Effect of 8Br-cGMP on MCA diameter

Extraluminal administration of 8Br-cGMP (10 µM) elicited a comparable vasodilation of the MCA in normonatremia (43 ± 3%, p < 0.05) and in hyposmotic hyponatremia (39 ± 6%, p < 0.05) demonstrating that cGMP-dependent transduction pathway is not affected by hyponatremia (Figure 2).

Figure 2.

Response of the MCA to analog of cGMP, 8Br-cGMP (10 µM) in normonatremia (n = 5) and in hyposmotic hyponatremia (n = 5). Dilation is expressed as a percentage of maximum diameter (0 Ca²⁺, EGTA 3 mM). Data are expressed as means ± S.E.M.

Contribution of K_ATP channels to the response of MCA to acidosis

The K_ATP channel activator, pinacidil (10 µM), significantly dilated MCAs by 54 ± 7% (p < 0.05) in normonatremia. Pinacidil also caused a relaxation of the MCA in hyposmotic hyponatremia (24 ± 5%, p < 0.05); however, this response was statistically significantly smaller (p < 0.05) than in normonatremia. Combined extraluminal application of pinacidil and acidosis caused dilation of the MCA by 75 ± 5% (p < 0.05) in normonatremia; however, this combination did not improve the response to acidosis in hyposmotic hyponatremia (Figure 3(a)).

Figure 3.

Response of the MCA to a reduction of pH to 7.0 (a) with K_ATP channel activator, pinacidil (10 µM) and BK_Ca channel activator, NS1619 (10 µM) in the background, in normonatremia (n = 9) and in hyposmotic hyponatremia (n = 10); (b) with inhibitor of BK_Ca channel, paxilline (1 µM) in the background, in normonatremia (n = 5) and in hyposmotic hyponatremia (n = 4). Dilation is expressed as a percentage of maximum diameter (0 Ca²⁺, EGTA 3 mM). Data are expressed as means ± S.E.M. *p < 0.05 between the groups.

Effect of BK_Ca channel inhibitor on MCA diameter

Administration of a specific inhibitor of BK_Ca channels, paxilline (1 µM), led to a reduction in vessel diameter from 135 ± 6 µm to 116 ± 5 µm (14 ± 5%, p < 0.05) in normonatremia. A similar effect was observed in hyposmotic hyponatremia, i.e. constriction of the MCA by 13 ± 2% (p < 0.05).

Contribution of BK_Ca channels to the response of MCA to acidosis

When acidosis was induced during normonatremia with paxilline (1 µM) in the background, MCA dilated statistically significantly (p < 0.5) but only by 7 ± 2%. This was in contrast to the response observed during acidosis alone (Figure 1 vs. Figure 3(b)). The presence of paxilline during acidosis in hyposmotic hyponatremia, in which the response of the MCA to acidosis was abolished, remained without influence (Figure 3(b)).

The BK_Ca channel activator, NS1619 administered in the standard dose of 10 µM, caused dilation of the MCA by 21 ± 5% (p < 0.05) in normonatremia, whereas it did not change vessel diameter in hyponatremia (Figure 4). It restored, however, the response of the MCA to acidosis. During combined extraluminal application of 10 µM NS1619 and acidosis in hyponatremia, the MCA dilated by 28 ± 5% (p < 0.001), (Figure 3(a)) which was not significantly different from the response to acidosis observed in normonatremia.

Figure 4.

Response of the MCA to: (a) increasing concentration of NS1619 (1 µM, 10 µM, 2 × 10 µM, 3 × 10 µM) in normonatremia (n = 18) and in hyposmotic hyponatremia (n = 17); (b) NS1619 (10 µM) in isosmotic hyponatremia (n = 4). Dilation is expressed as a percentage of maximum diameter (0 Ca²⁺, EGTA 3 mM). Data are expressed as means ± S.E.M. *p < 0.05 between the groups.

Effect of NS1619 on MCA diameter, impact of osmolality

Administration of NS1619 in increasing concentrations resulted in a progressive dilation of the MCA in normonatremia which started from the threshold concentration of 10 µM. The response of the MCA to NS1619 in hyposmotic hyponatremia was shifted to the right, i.e. threshold concentration which resulted in statistically significant increase of the MCA diameter by 51 ± 9%, (p < 0.05) amounted to 3 × 10 µM (Figure 4(a)). However, in isosmotic hyponatremia with standard Cl⁻ ions concentration, the MCA response to 10 µM NS1619 did not differ from that observed in normonatremia (Figure 4(b)).

Effect of acute hyponatremia on BK_Ca channel conductance in MCA smooth muscle cells

For the experiments, 13 elongated cells with a length of about 75 µm were used. Their cell membrane capacitance was 9 ± 0.8 pF. Depolarization of the cell membrane above −40 mV activated the outward potassium current. Activation potential and strong fluctuations of current suggested opening of the channels with large single channel conductance which is typical for BK_Ca channels. The identification of the current was confirmed by its reduction with specific BK_Ca channel inhibitor paxilline (Figure 5).

Figure 5.

Effect of paxilline (1 µM) on BK_Ca currents in normonatremia (145 mM Na⁺, 300 mosmol/kg H₂O, n = 3). (a) Representative recordings of BK_Ca currents before and (b) after paxilline administration, (c) Average I–V relationships in both cases. Data are expressed as means ± S.E.M. *p < 0.05 between the groups.

Lowering the concentration of Na⁺ to 121 mM for 1 h did not change the BK_Ca channel conductance at any of the tested membrane potentials (Figure 6).

Figure 6.

Effect of decrease of Na⁺ ion concentration to 121 mM on BK_Ca currents, n = 18. (a) Representative recordings of BK_Ca currents in normonatremia (145 mM Na⁺, 300 mOsm/kg H₂O) and (b) in hyponatremia (121 mM Na⁺, 250 mOsm/kg H₂O), (c) Average I–V relationships in normonatremia and in hyponatremia. Data are expressed as means ± S.E.M.

Effect of NS1619 on BK_Ca channel conductance in normonatremia and in hyponatremia

In this part of the study, we used 10 µM NS1619 which was the threshold concentration at which MCA dilated in response to this compound in normonatremia. The currents of the BK_Ca channel were measured after 1 and 5 min following administration of NS1619. For further analysis, the currents measured after 1 min were chosen due to the lack of time-dependent differences in the activation of the currents. NS1619 increased the activation of BK_Ca currents induced by depolarization to 40 mV and 60 mV in normonatremia (Figure 7(a) to (c)). The activation of BK_Ca currents by NS1619 in hyponatremia required higher potential (60 mV) than in normonatremic conditions (40 mV) (Figure 7(d) to (f)) demonstrating decreased voltage-dependency of BK_Ca channels in hyponatremia in the presence of specific activator of these channels.

Figure 7.

Effect of NS1619 (10 µM) on BK_Ca currents in normonatremia (145 mM Na⁺, 300 mOsm/kg H₂O, n = 5) and in hyponatremia (121 mM Na⁺, 250 mOsm/kg H₂O, n = 5). (a) Representative recordings of BK_Ca currents before and (b) after NS1619 administration in normonatremia, (c) Average I–V relationships in normonatremia in both cases, (d) Representative recordings of BK_Ca currents before and (e) after NS1619 administration in hyponatremia, (f) Average I–V relationships in hyponatremia in both cases. Potassium currents shown in curves were normalized to the maximum current measured in the cell prior to NS1619 administration. Data are expressed as means ± S.E.M. *p < 0.05 between the groups.

Discussion

There are three essential findings in the present study concerning the effect of acute, hyposmotic hyponatremia on the regulation of the MCA. Firstly, acute hyposmotic hyponatremia results in the elimination of the response of the MCA to extraluminal acidosis which can be restored by the BK_Ca channel activator, secondly, acute hyposmotic hyponatremia decreases the voltage-dependency of the activation of the BK_Ca channels in the presence of a specific opener of these channels. Thirdly, impaired response of the MCA to acidosis depends on hyposmolality rather than on decreased sodium ion concentration.

In this study, we used the model of cannulated, pressurized, and perfused isolated cerebral artery, which enables the investigation of vascular responses without parenchymal influences.^18,22,23 Slow perfusion of the vessel and pressurization to the physiological level of intravascular pressure simulates in vivo conditions, ensures proper myogenic tone and shear stress.^23,24 Luminal perfusion was adjusted to 100 µL/min which is consistent with a normal physiological shear stress (20 dyn/cm²).²⁵ The blood vessels are studied in such a system either in physiological saline solution/HEPES aerated with CO₂ in which the pH is adjusted by a bicarbonate buffer or in a MOPS buffer.^18,26–35 Both systems seem to be equivalent.²³ In the case of the aerated buffer, acidosis is achieved by increasing CO₂ content or decreasing NaHCO₃ concentration in the luminal perfusate or extraluminal solution,^27,32 whereas, in the case of a MOPS buffer, acidosis is induced by increasing H⁺ concentration in the buffer.^18,33,36 It has been demonstrated that in either a bicarbonate-buffered saline solution or in artificial buffer solutions, the reduction of vascular tone in response to the decrease in extracellular pH is similar.³⁷

Moreover, hypercapnic relaxation of cerebral vessels is associated with a fall of extracellular pH, and extracellular pH rather than pCO₂ results in cerebral blood vessels dilation.²⁷ Accordingly, acidic solutions infused ventriculocisternally, in the absence of changes in pCO₂ caused a rise in total and/or regional cerebral blood flow (CBF)^34,35,38 and, superfused in a cranial window, increased CBF³⁹ and dilated pial arterioles in the rat.⁴⁰ Also, in vitro perfusion with acidic isocapnic solution dilated pressurized segments of cerebral arterioles^33,41,42 and the rat MCA.³⁶ Additionally, the response of the cerebral vessels to hypercapnic and normocapnic acidosis was similar in studies conducted in vivo in the cranial window in the cat⁴³ and in vitro on isolated cerebral vessels.^28–32

In the present study, lowering of pH in the extravascular space to 7.0 caused dilation of the MCA in normonatremia and this result is consistent with other studies performed on isolated and pressurized rat MCA.³⁶ However, in hyposmotic hyponatremia, we did not observe such a response. To induce hyponatremia, we withdrew 24 mM NaCl from the MOPS-PSS, which produced a decrease in Na⁺ and Cl⁻ concentrations to 121 mM and 126 mM, respectively, and a reduction of osmolality to 250 mOsm/kg H₂O. The equalization of osmolality and Cl⁻ concentration resulted in the restoration of the MCA response to acidosis in spite of hyponatremia. These results indicate that the disordered response of the MCA to acidosis in hyposmotic hyponatremia is associated with hyposmolality and not with decreased Na⁺ concentration. We do not take into consideration the decreased Cl⁻ concentration as it did not fall below the physiological range. Besides, it should be mentioned that according to the available literature, hyposmotic challenge activates Cl⁻ currents in the smooth muscle cells of the cerebral blood vessels secondary to the opening of non-selective cation channels.^7,44

Discussing the possible mechanism behind the lack of dilation of the MCA to acidosis in hyposmotic conditions, we should at first focus on the factors which are responsible for the response to acidosis in normonatremia. It is known that protons act through multiple mechanisms including the NO pathway^36,41 and activation of the K_ATP^26,41,45 or K_Ca/BK_Ca channels.^32,46,47 According to Lindauer et al.,³⁶ inhibition of NO synthase abolished the response of the MCA to the lowering of the extravascular pH to 7.0. NO synthase activated during acidosis was, however, of neuronal origin as removal of the endothelium had no effect on the response of cerebral blood vessels to respiratory and metabolic acidosis, while denervation of them abolished this response.^27,28,36

Thus, the possible causes of the elimination of the MCA response to acidosis in hyposmotic hyponatremia could be the impairment of NO-dependent signal transduction or dysfunction of the respective K⁺ channels in vascular smooth muscle cells.

It is known that smooth muscle response to NO in the rat MCA depends on two mechanisms: (i) stimulation of cytosolic guanylyl cyclase and a rise of the cGMP concentration, and (ii) activation of BK_Ca potassium channels which are involved in the response to NO independently of cGMP.¹⁵ Increase of the cGMP concentration relaxes the MCA vascular smooth muscle due to a decrease of sensitivity of the contractile proteins to Ca²⁺.¹⁵ To study the possibility that cGMP-related signal transduction is impaired in the MCA in hyponatremia, a cell membrane permeable analog of cGMP, 8-Br-cGMP, was administered. The dilation of the MCA in response to 8-Br-cGMP in hyponatremia did not differ from the one observed in normonatremia. Therefore, it does not seem that impairment of the cGMP pathway is responsible for the elimination of the MCA response to acidosis in hyposmotic hyponatremia.

Considering potassium channels modulated by both NO and H⁺ ions, it has been shown that dilation of MCA in response to acidosis is associated with the opening of K_ATP and K_Ca channels,^18,46,47 the latter being in fact the BK_Ca channels, because only this subtype of calcium sensitive potassium channel is present in smooth muscles in cerebral circulation. Other types of K_Ca channels: SK_Ca and IK_Ca are present in the endothelium.^48,49 Although it has been suggested that the SK_Ca and IK_Ca channels might also be present in smooth muscle cells,⁵⁰ this was not confirmed functionally.⁵¹ Similarly, functional BK_Ca channels are absent in the endothelium.⁵¹ Thus, the preserved response of the endothelium-denuded MCA to acidosis indicates that the SK_Ca and IK_Ca channels do not participate in this response.

Another K⁺ channel which is present in vascular smooth muscle cell membrane and theoretically can participate in the vasodilation during acidosis is the K_IR channel – inward rectifier involved in neurovascular coupling.⁵² This channel is sensitive to small changes in the extracellular concentration of K⁺ and cooperates with the BK_Ca channels in intercellular K⁺ channel- to -K⁺ channel signaling.⁵² Although we are not aware of any publication on H⁺ sensitivity of the vascular K_IR channels, we can hypothesize that the outward K⁺ current associated with the opening of BK_Ca channels during acidosis may result in the activation of K_IR channels. However, we can exclude the participation of K_IR channels in the dilation of the MCA during acidosis based on the observation by Lindauer et al.¹⁸ that an inhibitor of these channels did not affect the response of the MCA to acidosis.

According to the same study, K_Ca channel inhibition significantly reduces, while inhibition of the K_Ca and K_ATP channels completely abolishes the MCA response to acidosis. Similar results were obtained in studies performed on newborn pigs in which blockade of both K_ATP and K_Ca channels completely prevents dilation of the cerebral vessels to hypercapnia.²⁹ Our results also show that inhibition of the BK_Ca channel with paxilline significantly reduces, but does not completely abolish, vasorelaxation in response to acidosis in normonatremia leaving the participation of the K_ATP channel in this response open.

Concerning the contribution of the impairment of K_ATP channels to the elimination of the MCA response to acidosis during hyposmotic hyponatremia, we did not find evidence that the function of this channel was markedly influenced by hyponatremia. Pinacidil, the K_ATP channel activator evoked vasodilatation of the MCA in normonatremia and hyponatremia. Moreover, it did not restore the response of the MCA to acidosis in hyponatremia. Therefore, our results indicate that the impaired response of MCA to acidosis in hyposmotic hyponatremia is not associated with the attenuation of the K_ATP channel.

The results of the present study support, however, an alternative hypothesis that BK_Ca channels are affected by hyponatremia. To examine the possible contribution of the BK_Ca to the disordered response of the MCA to acidosis in hyponatremia, we used paxilline, and NS1619 (BK_Ca channel blocker^29,32 and opener,^18,19 respectively) in concentrations typically used in BK_Ca channel studies.^18,20,53,54 Paxilline is known to constrict cerebral blood vessels under physiological conditions, which indicates that the BK_Ca channels participate in the maintenance of myogenic tone in cerebral circulation.^20,53,54 In the present study, this inhibitor had a similar constrictor effect on MCA, both in normonatremia and in hyponatremia. These results indicate that the BK_Ca channels are present and participate in the maintenance of the basal diameter of the MCA both in iso- and hyposmotic conditions as well as in normo- and hyponatremia. These observations are in agreement with our previous findings that autoregulation, in which BK_Ca channels are involved,¹⁷ is also unchanged in acute hyposmotic hyponatremia.¹⁴

On the other hand, NS1619 caused relaxation of the MCA in normonatremia but did not elicit such an effect in hyponatremia. However, NS1619 administered together with increased H⁺ ion concentration in the extravascular space restored dilation of the MCA to acidosis in hyponatremia. Patch clamp experiments on smooth muscle cells originating from MCA showed no difference in whole cell currents elicited by depolarizing potentials between cells conditioned in normonatremic or hyponatremic bath solution indicating that channel function is not disturbed in acute hyponatremia; 10 µM NS1619 increased whole cell current, but, in hyponatremic conditions, the activation by NS1619 required a higher potential (60 mV) than in normonatremic conditions (40 mV), suggesting a depolarizing shift of channel activation in hyponatremia.

At this point, the question arises of how specific is NS1619 as an activator of BK_Ca channels. From a literature overview, it seems that the specificity of NS1619 depends on the vessel and species. According to Holland et al.¹⁹ in smooth muscle cells isolated from the rat basilar artery, NS1619 not only directly activates the BK_Ca channel but also inhibits K_V currents as well as the L-type Ca²⁺ channel. However, some other studies indicated that vasoconstriction of pial arteries of the piglet in response to Ca²⁺ channel agonist was unchanged in the presence of NS1619.⁵⁵ According to Lindauer et al.,¹⁸ in the rat MCA, NS1619 acts as a selective activator of the BK_Ca channels rather than the calcium channel blocker. In the present study, this activator evoked vasodilation of the MCA in control conditions comparable to the results reported by Lindauer et al.¹⁸

In arterial smooth muscle, BK_Ca channels are activated by micromolar intracellular Ca²⁺ release (Ca²⁺ sparks) from sarcoplasmic reticulum (SR) through ryanodine channels.^17,56 Such a mechanism was also described in human cerebral arteries.⁵⁷ Most Ca²⁺ sparks occur in close proximity to the plasma membrane which facilitate the activation of BK_Ca channels.¹⁷ It is worth to mention that NS1619, apart from direct activation of BK_Ca, can also act indirectly by increasing Ca²⁺ release from intracellular stores.¹⁷ Interestingly, it has been recently found that reducing external pH from 7.4 to 7.0 changed the mode of releasing Ca²⁺ from SR in brain parenchymal arterioles: it reduced Ca²⁺ wave activity, and drastically elevated Ca²⁺ spark activity.³² Therefore, apparently impaired function of the BK_Ca channel in hyponatremia could be associated with decreased Ca²⁺ spark-related activation of BK_Ca channel, similarly as reported during hypertension and diabetes^58–60 or with a reduction in Ca²⁺ spark activity, as described after subarachnoid hemorrhage.²⁰ Decreased Ca²⁺ spark activation of BK_Ca channel resulted in a smaller amplitude of BK currents and an impaired response of the cerebral arteries to a specific channel inhibitor.^59,60 With regard to our findings, it can be assumed that the release of Ca²⁺ from internal stores in response to 10 µM NS1619 or acidosis is attenuated in hyponatremia. In accordance with such an assumption, the stronger stimulus represented by the combined administration of 10 µM NS1619 and acidosis or by the application of 30 µM NS1619 results in the increase of Ca²⁺ sparks activity enough to restore dilatation of the MCA to acidosis and NS1619, respectively. At this point, it is worth to mention that the decrease of Ca²⁺ spark activity during hyposmotic hyponatremia has been reported in the rat isolated ventricular myocytes.⁶¹ Restoration of the MCA response to acidosis and NS1619 in isosmotic hyponatremia in our study may speak in favor of a similar mechanism in cerebral vascular myocytes. However, we could not exclude reduced sensitivity of the BK_Ca channels to Ca²⁺ sparks or even reduced sensitivity of BK_Ca channels to the direct action of NS1619 in hyponatremia.

In conclusion, acute hyponatremia impaired the response of the MCA to acidosis and decreased the voltage-dependency of BK_Ca channels during their stimulation by the specific activator. The interesting observation that specific activator of BK_Ca channels restored the response of the MCA to increased H⁺ ion concentration during hyponatremia requires further investigations. Taking into consideration that many patients with hyponatremia also developed acidosis,⁶² our data may contribute to a better understanding of this disorder and the treatment of patients who demonstrate brain symptoms connected with hyponatremia.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by a grant of the Polish Ministry of Science and Higher Education [N401 19032/3924] and by the Mossakowski Medical Research Centre PAS.

Declaration of conflict of interest

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Authors’ contributions

MA performed research, analyzed data, and wrote the article; BD performed research; KD analyzed data; EK designed research, analyzed data, and drafted the article.