
Departments of
* Physiology and
Medicine, Pulmonary Research Group, University of Alberta, Edmonton, Canada
Correspondence: Dr. Marek Duszyk, Department of Physiology, University of Alberta, 7-46 Medical Sciences Building, Edmonton, Alberta T6G 2H7, Canada. E-mail: marek.duszyk{at}ualberta.ca
|
|
|---|
Key Words: cGMP diazoxide KATP RT-PCR
|
|
|---|
Nitric oxide (NO) is known to regulate ion channels in many cell types. NO activates KCa and adenosine 5'-triphosphate (ATP)-sensitive K+ channels (KATP) in smooth muscle [6 ], neuronal [7 ], endothelial [8 ], and colonic epithelial cells [9 ] via cGMP-dependent and -independent mechanisms [10 , 11 ]. In addition, NO activates cystic fibrosis transmembrane conductance regulator (CFTR) [12 ] and non-CFTR Cl- channels in lung epithelial cells [13 ]. In human airways, several cell types produce NO, including neutrophils [14 ], macrophages [15 ], epithelial cells [16 ], endothelial cells [17 ], and to a lesser extent, eosinophils [18 ]. An increased amount of NO was found in the exhaled air of asthmatic patients and was shown to correlate with lung eosinophilia [19 ]. NO may also be protective to airways reducing neutrophil recruitment [20 ] and O2- production [21 ]. However, the intracellular mechanisms that enable NO to exert such a wide range of effects in airways are not understood completely.
The aim of the present study was to investigate the effect of NO on ion channel function in human peripheral blood eosinophils. We hypothesized that endogenous and exogenous NO affects ion channel function via cGMP-dependent and/or -independent mechanisms. We found that endogenous NO did not affect channel function, but NO donors activated whole-cell currents via a cGMP-dependent pathway. Ion replacement studies indicated that the NO-activated current was carried by K+ ions. It is interesting that NO affected only KATP channel function, suggesting that these channels may represent a novel target to modulate eosinophil activation in asthma and related allergic diseases.
|
|
|---|
Patch-clamp experiments in human peripheral blood eosinophils
Peripheral blood eosinophils were purified to homogeneity (<98%) from atopic asthmatic volunteers who had given their informed consent, as described previously [22
, 23
]. Whole-cell recordings were obtained using the amphotericin B-perforated patch-clamp technique. In preliminary experiments, baseline whole-cell currents were recorded for up to 45 min. These experiments showed that the current stabilized after approximately 10 min, and all recordings were obtained under stable baseline conditions. Patch pipettes were pulled from borosilicate glass (A-M Systems, Carlsbourg, WA) using a Narishige puller (Tokyo, Japan).
Experiments were performed in the following bath solution (mM): 137 NaCl, 6.4 KCl, 4.3 Na2HPO4, 1.4 KH2PO4, 1.2 MgCl2, 0.5 CaCl2, 5 glucose. The pipette tip was dipped into pipette solution (mM): 137 KCl, 6.4 NaCl, 4.3 NaHPO2, 1.4 KH2PO2, 1.2 MgCl2, 0.5 CaCl2, 1 ethyleneglycol-bis(ß-aminoethylether)-N,N'-tetraacetic acid, 4 glucose. The pipette was then back-filled with the same solution containing amphotericin B (240 µg/mL, Sigma-Aldrich, St. Louis, MO). When the contribution of K+ channels to the whole-cell current was measured, 137 mM Cs+ replaced equimolar K+ in the pipette solution. Similarly, to study the contribution of Cl- channels, 134 mM gluconate replaced equimolar Cl- in the bath and pipette solutions. All buffers were supplemented with 0.1% bovine serum albumin, and the pH was adjusted to 7.4. Experiments were performed at room temperature.
Liquid junction potentials (LJPs) develop during a patch-clamp recording whenever the pipette-filling solution is different from the bath solution [24
]. Because LJPs could be misinterpreted as cellular effects, data were corrected for LJPs when they exceeded ±1 mV. An LJP was defined as the potential of the bath solution with respect to the pipette solution, and the membrane potential VM was calculated from the reading provided by the patch-clamp amplifier V, as VM = -V + LJP. LJPs were calculated using the generalized Henderson equation for N polyvalent ions:
![]() |
![]() |
Pipette resistances were between 3 and 8 M
, and recordings were obtained using a patch-clamp amplifier (EPC-7; List Medical, Germany) in the voltage-clamp mode. The holding potential was -60 mV, and 20 mV steps, ranging from -80 to +80 mV, were applied every 200 ms. The cells were kept in the bath solution for 1 h before starting the experiments. After baseline currents stabilized, stimuli or blockers were added to the bath solution, and currents were recorded for up to 40 min. Data were analyzed using custom-written patch-clamp software (kindly provided by Dr. A. S. French, Dalhousie University, Canada). Statistical analysis of whole-cell currents was performed at +80 mV.
Reverse-transcriptase polymerase chain reaction (RT-PCR)
Total RNA was isolated from 2 x 106 eosinophils using the Qiagen RNeasy Mini kit (Qiagen, Chatsworth, CA). The average amount of RNA obtained from 2 x 106 eosinophils was 300 ng. One-third of the RNA was reverse-transcribed using superscript II RT (Gibco-BRL, Grand Island, NY) and random hexamers (50 A260 units; Boehringer Mannheim, Mannheim, Germany) as primers. Thereafter, PCR was performed in 20 µL reactions. The expression of K+ channels was studied using the primer pairs described in Table 1
. As a positive control for K+ channels, IK, BK, and Kv, mRNA from the human airway epithelial cell line A549 was used, and for TWIK-1 and TASK-2 expression, mRNA from Calu-3 cells was used. One-tenth of the cDNA was used in PCR experiments. DNA amplification was obtained by annealing for 45 s at 64°C for IK and BK, 56°C for Kv, 60°C for TWIK-1, and 51°C for TASK-2. This was followed by an elongation step at 72°C for 1 min. DNA sequences were amplified during 30 cycles. The sizes of the expected amplified products are shown in Table 1
.
|
View this table: [in a new window] |
Table 1. List of the K+ Channels Tested, the Positions of the Primers Relative to the Published Data Sequences, the Expected Size of the RT-PCR Products, and the Base Sequence of the Primers in the 5' 3' Direction
|
|
|
|---|
![]() View larger version (28K): [in a new window] |
Figure 1. (A) A representative recording showing activation of the whole-cell current by GSNO (100 µM). (B) I-V relationship of the recording shown in panel A. GSNO shifted the reversal potential from -21 mV to -30 mV. (C) A representative recording showing activation of the whole-cell current by SNAP (100 µM). (D) I-V relationship of the recording shown in panel C. SNAP shifted the reversal potential from -18 mV to -31 mV.
|
NO stimulates the whole-cell current via a cGMP-dependent pathway
To investigate whether NO was acting on eosinophil ion channels via a cGMP-dependent or -independent mechanism, cells were incubated with 8-Br-cGMP, the soluble guanylyl cyclase inhibitor ODQ, or a protein kinase G inhibitor H89 before stimulation with SNAP (Fig. 2
). ODQ (10 µM) and H89 (1 µM) had no effect on the baseline current and prevented subsequent activation of the current by SNAP (n=5 for both drugs). In other experiments, addition of 8-Br-cGMP increased the whole-cell current significantly, indicating that cGMP is involved in the activation of the observed current. The addition of 100 µM and 500 µM 8-Br-cGMP increased whole-cell currents by 27 ± 8% (n=4) and 124 ± 25% (n=4), respectively. The increase in whole-cell current caused by 500 µM 8-Br-cGMP was similar to that induced by SNAP (139±12%). All of these results suggest that exogenous NO activates the whole-cell current via a cGMP-dependent pathway.
![]() View larger version (39K): [in a new window] |
Figure 2. The NO/cGMP pathway is involved in the whole-cell current activation by SNAP. Baseline currents are expressed as 100%. ODQ (10 µM) had no effect on the baseline current (P>0.05; n=5) but inhibited the SNAP-induced increase (n=5). Similarly, H89 (1 µM) did not affect the baseline current (P>0.05; n=5) but inhibited stimulation of the whole-cell current by SNAP (n=5). 8-Br-cGMP (0.1 mM and 0.5 mM) increased the whole-cell current in a dose-dependent manner by 27 ± 8% and 124 ± 25% (n=4; P<0.05 in each series), respectively. (*P<0.05).
|
![]() View larger version (26K): [in a new window] |
Figure 3. SNAP activates K+ but not Cl- channels. (A) A representative trace of a whole-cell current using Cs+-containing pipette and bath solution, showing that the addition of SNAP had no effect on the whole-cell current. (B) I-V relationship of the recording shown in panel A. (C) A representative trace of a whole-cell current in the presence of gluconate in the pipette and bath solution, showing that SNAP increased the whole-cell current. (D) I-V relationship of the recording shown in panel C.
|
![]() View larger version (57K): [in a new window] |
Figure 4. Summary of the effects of K+ channel openers and blockers on the eosinophil whole-cell current. Baseline currents are expressed as 100%. Diazoxide alone increased the current by 73 ± 19% (100 µM; n=10) and in combination with SNAP by 102 ± 10% (n=5). Glibenclamide (5 µM) had no effect on the whole-cell current but inhibited subsequent current activation by SNAP (n=6 and n=4, respectively). 1-EBIO increased the current by 64 ± 20% (500 µM; n=5), whereas clotrimazole had no effect on the baseline current (10 µM; n=6) and did not affect subsequent current activation by SNAP (n=4). 4-AP inhibited the baseline current by 41 ± 5% (2 mM; n=7) and blocked subsequent current activation by SNAP (n=4). *, P < 0.05.
|
Gene expression of K+ channels in eosinophils
Because KATP channels have been described in human eosinophils previously [4
, 25
], we investigated the expression of IK and large conductance (BK) KCa channels, Kv, and two members of the K2P family, TWIK-1 and TASK-2 (Table 1)
. Figure 5
shows representative RT-PCR experiments. Eosinophils express mRNA for IK, Kv, and TWIK-1 but not for BK and TASK-2 channels (n=6). The identity of the PCR products was confirmed by sequencing and comparison with the corresponding Genbank sequences.
![]() View larger version (38K): [in a new window] |
Figure 5. Expression of K+ channels in human eosinophils using RT-PCR. Eosinophils express mRNA for IK, Kv, and TWIK-1 but not BK and TASK-2. For IK, BK, and Kv, A549 cells were used as a positive control. For TWIK-1 and TASK-2, Calu-3 cells were used as a positive control. Data are representative of six different experiments. M, DNA standard (a 100-bp ladder).
|
|
|
|---|
Activation of soluble guanylyl cyclase (sGC) and generation of cGMP are responsible for many biological effects of NO [13 , 27 ]. The results of this study show that the NO/cGMP-dependent pathway is also involved in the regulation of the whole-cell current in eosinophils because the NO effects could be eliminated by pretreatment of cells with a selective inhibitor of sGC, ODQ; activation of the whole-cell current by NO was abolished in the presence of a PKG inhibitor, H89; and application of membrane-permeable 8-Br-cGMP produced an effect similar to that of NO. Although these results are consistent with the regulation of eosinophil whole-cell current via the NO/cGMP-dependent pathway, they do not exclude the involvement of a cGMP-independent pathway in this process.
The baseline and NO-stimulated whole-cell currents showed a nearly linear I-V relationship and were neither time- nor voltage-dependent (Fig. 1)
. This suggests that under our experimental conditions, a contribution of H+ channels to the whole-cell current was insignificant, because these channels show time-dependent activation at depolarizing voltages [28
]. Our data indicate that Cl- and K+ channels dominate the whole-cell current in eosinophils. This conclusion is based on ion substitution studies and the use of ion channel blockers. Although Cl- channels account for
50% of the whole-cell current in unstimulated eosinophils [5
], substitution of Cl- with gluconate did not affect current activation by NO, indicating that NO donors activated K+ but not Cl- channels. It is important to notice that human peripheral blood eosinophils represent a heterogeneous population of cells, often with different sizes, or even different activation status. This heterogeneity contributes to the variability in baseline whole-cell currents observed in the present and previous studies [5
].
RT-PCR studies showed that eosinophils express mRNA for the ß subunit of Kv channels, Ca2+-dependent K+ channels (IK), and a member of the K2P family, TWIK-1. The functional role of Kv channels in eosinophils remains unclear, because the nearly linear I-V relationship suggests that these channels do not make a significant contribution to the baseline or SNAP-induced current. Similarly, the TWIK-1 channels do not contribute to baseline or NO-stimulated currents, because these channels conduct an inwardly rectifying current. Studies from other laboratories suggested expression of KATP channels in human eosinophils [4 , 25 ].
The KATP channel is a heteromultimeric complex of a K+-selective pore and a sulfonylurea receptor (SUR), which are structurally unrelated to each other. It is composed of four inwardly rectifying K+ channel subunits, Kir6.1 or Kir6.2, and four SUR subunits, which belong to the family of ATP-binding cassette (ABC) transporters. Different combinations of the Kir6.2 or Kir6.1 subunits and the SUR1 and SUR2 account, in part, for the molecular and functional diversity of KATP channels. Depending on the combination of Kir6.x and SURx subunits, the single-channel conductance of KATP channels has been estimated between 10 and 80 pS in the presence of 140 mM K+ on both sides of the membrane [29 30 31 ].
Glibenclamide is an agent that binds with high specificity to sulfonylurea receptors, which belong to the ABC-transporter family (subfamily ABCC). This family also includes the CFTR Cl- channels. Although glibenclamide at the concentrations used in our experiments (5100 µM) is known to affect CFTR function, previous studies have shown that eosinophils do not express CFTR mRNA [5 ]. Therefore, the effect of glibenclamide on the whole-cell current cannot be attributed to its effects on CFTR.
An opener of KATP channels, diazoxide, and NO donors activated a whole-cell current of
50 pA, corresponding to a whole-cell conductance of 625 pS at +80 mV. The number of ion channels (N) activated by NO could be evaluated from the following relationship: G =
N P, where G- is the whole-cell conductance,
- is the single-channel conductance, and P- is the channel open probability. If we assume that the KATP channel conductance is 50 pS and that its open probability is P = 0.5, then the number of KATP channels activated by NO is equal to 25.
After activation of KATP channels by diazoxide, the subsequent addition of SNAP increased the whole-cell current further. A similar effect has been shown for mitochondrial KATP channels, where NO has been shown to activate KATP channels and potentiate the effect of diazoxide [32 ]. It is possible that diazoxide opens KATP channels partially, and NO is further increasing the channel open probability. Alternatively, eosinophils may possess different subtypes of KATP channels, and NO might open a different subpopulation than diazoxide.
In summary, eosinophils possess several types of K+ channels, but NO affects only KATP channel function. NO induces an outflow of K+ ions in eosinophils and thus causes membrane hyperpolarization. This reduces the cytoplasmic Ca2+ concentration; as a result, the eosinophils could be less excitable, and stronger stimuli are needed for depolarization. Thus, NO at concentrations likely to be encountered in vivo could perform a protective role, preventing eosinophil activation.
Received October 22, 2001; revised January 22, 2002; accepted January 23, 2002.
|
|
|---|
and Herpes simplex virus/tumor necrosis factor-
synergistically induce nitric oxide synthase 2 in macrophages through cooperative action of nuclear factor-
B and IFN regulatory factor-1 Eur. Cytokine Netw. 12,297-308[Medline]
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||