SEARCH

SEARCH BY CITATION

Keywords:

  • action potential;
  • atrial myocytes;
  • inducible NOS (iNOS);
  • ion channels;
  • NO;
  • polyphenol;
  • sepsis

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of interest
  9. References

BACKGROUND AND PURPOSE Supraventricular tachyarrhythmias, including atrial fibrillation, are occasionally observed in patients suffering from sepsis. Modulation of cardiac ion channel function and expression by sepsis may have a role in the genesis of tachyarrhythmias.

EXPERIMENTAL APPROACH Sepsis was induced by LPS (i.p.; 300 µg·kg−1) in guinea pigs. Membrane potentials and ionic currents were measured in atrial myocytes isolated from guinea pigs 10 h after LPS, using whole cell patch-clamp methods.

KEY RESULTS In atrial cells from LPS-treated animals, action potential duration (APD) was significantly shortened. It was associated with a reduced L-type Ca2+ current and an increased delayed rectifier K+ current. These electrophysiological changes were eliminated when NG-nitro-l-arginine methyl ester (l-NAME) or S-ethylisothiourea was given together with LPS. In atrial tissues from LPS-treated animals, Ca2+ channel subunits (Cav1.2 and Cav1.3) decreased and delayed rectifier K+ channel subunits (Kv11.1 and Kv7.1) increased. However, L-NAME treatment did not substantially reverse such changes in atrial expression in LPS-treated animals, with the exception that Kv11.1 subunits returned to control levels. After LPS injection, inducible NOS in atrial tissues was up-regulated, and atrial NO production clearly increased.

CONCLUSIONS AND IMPLICATIONS In atrial myocytes from guinea pigs with sepsis, APD was significantly shortened. This may reflect nitration of the ion channels which would alter channel functions, rather than changes in atrial expression of the channels. Shortening of APD could serve as one of the mechanisms underlying atrial tachyarrhythmia in sepsis.


Abbreviations
APD

action potential duration

DAF-2

DA, 4,5-diaminofluorescein diacetate

EIT

S-ethylisothiourea

eNOS

endothelial NOS

ICa

inward Ca2+ current

ICU

intensive care unit

IK

delayed rectifier K+ current

INa

inward Na+ current

iNOS

inducible NOS

l-NAME

NG-nitro-l-arginine methyl ester

LPS

lipopolysaccharide

Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of interest
  9. References

Supraventricular tachyarrhythmias, including atrial fibrillation and flutter, are frequently encountered in surgical and non-surgical intensive care unit (ICU) patients (Artucio and Perrier, 1990). These arrhythmias would result in a longer and more expensive stays in the ICU and in hospital (Ommen et al., 1997; Maisel et al., 2001; Reinelt et al., 2001). Moreover, their new onset in critically ill patients has been strikingly associated with increased morbidity and mortality (Brathwaite and Weissman, 1998).

Growing evidence suggests that the presence of sepsis is by far the most important risk factor for the development of supraventricular tachyarrhythmias in the ICU. More than three decades ago, Ledingham and McArdle,(1978) found development of atrial fibrillation was not unusual in patients with septic shock. In a retrospective study, sepsis was found to be a significant risk factor associated with new-onset tachyarrhythmias after cardiac surgery (Mayr et al., 2001) and in a prospective case control study performed in a surgical ICU (Knotzer et al., 2000). Furthermore, Bender (1996) showed that 32% of surgical ICU patients who developed supraventricular tachyarrhythmias had an association with sepsis and indicated that the presence of shock, notably septic shock, was an independent risk factor of atrial fibrillation. These clinical reports are consistent with the idea that sepsis is closely associated with the development of new-onset supraventricular tachyarrhythmias and its greater mortality (Goodman et al., 2007).

The mechanisms involved in the occurrence of supraventricular tachyarrhythmias during sepsis remain to be clarified. Generally, the principal mechanisms responsible for tachyarrhythmias are considered to be those of enhanced automaticity, triggered activity or re-entry (Kadish and Passman, 1999). Alterations in structure and function of atrial ion channels associated with the pathophysiology of sepsis may play a role in causing these electrophysiological abnormalities. However, such electrophysiological disturbances in atrial cells under conditions of sepsis are poorly understood.

In the present study, the experiments were undertaken to investigate the changes in function and expression of the membrane ion channels responsible for triggering and maintaining the action potentials in atrial cells from guinea pigs after induction of sepsis with lipopolysaccharide (LPS). During sepsis, inducible NOS (iNOS) can be induced in a variety of cells, including macrophages, vascular smooth muscle cells, cardiac myocytes and endothelial cells, leading to overproduction of NO (Morris and Billiar, 1994). As NO has been shown to modulate a number of cardiac ion channels (Fischmeister et al., 2005; Gonzalez et al., 2009; Tamargo et al., 2010), the regulatory role of excessive NO was also examined in animals treated with the NOS inhibitor NG-nitro-l-arginine methyl ester (l-NAME) or the selective iNOS inhibitor S-ethylisothiourea (EIT).

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of interest
  9. References

Animals and treatments

All animal care and experimental procedures were in accordance with the National Institutes of Health guidelines and were approved by the Animal Use Committee of the University of Toyama. Male Hartley guinea pigs, weighing 200–350 g, were injected with LPS (i.p.; 300 µg·kg−1; Escherichia coli 055; List Biological Laboratories, Campbell, CA) or an equivalent volume of sterile saline, as previously described (Kamiyama et al., 2007). This treatment of guinea pigs with LPS mimicked many of the clinical features of sepsis. Heart rate and respiratory rate were temporally elevated from baseline in the LPS-treated guinea pigs, reaching a maximum at 10 h (from 223 ± 3 to 374 ± 2 beats min−1 and from 36 ± 1 to 106 ± 2 min−1, n = 4) and declining thereafter. Rectal temperature also showed a peak fall at 10 h after LPS (from 37.4 ± 0.1 to 34.3 ± 0.2°C) and returned to baseline values at 24 h. Furthermore, LPS treatment resulted in a progressive increase in the number of leukocytes from 3820 ± 112 µL−1 at 0 h to 10120 ± 199/µL at 10 h remaining high (10620 ± 465 µL−1) at 24 h. These findings prompted us to choose 10 h after LPS as the relevant time point for our measurements. Thus, after 10 h from LPS or saline injection, the animals were killed by an overdose of pentobarbital. In some experiments, i.p. injections of l-NAME (200 mg·kg−1) or EIT (5 mg·kg−1) were given at the same time as the LPS or saline.

Measurements of action potentials and membrane currents

Single atrial myocytes of the guinea pig were obtained by enzymatic dispersion (Hatakeyama et al., 2002). The hearts were excised and the ascending aorta was cannulated with a dull-edged needle. To wash out the blood, S-MEM solution (Gibco, Carlsbad, CA) containing high Ca2+ (1.0 mM) was circulated for 10 min via the coronary artery, and then normal S-MEM solution was circulated for 10 min to remove excessive Ca2+. Subsequently, S-MEM solution containing collagenase (type 1A; 0.02%) and trypsin (0.002%) was circulated for 10 min. The atrial tissue was cut into small pieces in S-MEM solution containing BSA, transferred into KB solution (composition (in mM) 70 glutamic acid, 15 taurine, 30 KCl, 10 KH2PO4, 0.5 MgCl2, 11 glucose, 10 HEPES, 0.5 EGTA; pH 7.3), and stored in a refrigerator (5°C) for later use.

Electrophysiological recordings were performed by the whole-cell patch-clamp technique, using glass patch electrodes with a tip diameter 1 µm and a resistance of 3–5 MΩ (Hattori et al., 2000; Hatakeyama et al., 2009). Glass patch electrodes were fabricated with a horizontal microelectrode puller (P-87; Sutter, Novato, CA) from glass microtubes (G-1.5; Narishige, Tokyo, Japan) and heat polished with a microforge (MF-83; Narishige). In order to record membrane currents under whole-cell voltage clamp and action potentials under the current-clamp version of patch clamping, a patch clamp amplifier (EPC-9; Heka, Lambrecht, Germany) was used. Acquisition and analysis of the data were performed by Pulse and Pulsefit software (Heka). The compositions of internal and external solutions were essentially the same as those used in our previous studies (Hatakeyama et al., 2002; 2009). In general, the L-type Ca2+ current (ICa) was elicited by a 200 ms depolarizing test pulse to +10 mV from a holding potential (VH) of −30 mV in order to avoid the Na+ current (INa) and the T-type Ca2+ current. We confirmed that this current was completely blocked by the dihydropyridine compound nifedipine at 1 µM. The amplitude of ICa was thus measured as the difference between the peak of the inward current and the current at the end of the test pulse. Meanwhile, delayed rectifier K+ current (IK) was measured; external solution contained 1 µM nifedipine and 100 nM CaCl2 to eliminate the involvement of ICa in the whole-cell membrane currents. All experiments were carried out at room temperature (21–23°C).

RNA extraction and real-time quantitative PCR

Total RNA was extracted from atrial tissues with the use of an RNeasy Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer's instructions. RNA was reverse-transcribed to cDNA using a PrimeScript RT reagent kit (Takara Bio, Ohtsu, Japan), and PCR was performed with a Takara RNA PCR kit (Takara Bio) as described in the manufacturer's manual. Quantitative PCR assay was performed in 20 µL volumes using optical PCR tubes, and the PCR mixture included SYBR Premix Ex Taq (Takara Bio) and 0.2 pmol·µL−1 of each pair of oligonucleotide primers. The reactions were run on an Mx3000P™ Real-Time PCR System (Stratagene, La Jolla, CA). Each experiment was repeated at least twice. The primer pairs were designed for amplifying the following target genomic sequences: Nav1.5 (channel nomenclature follows Alexander et al., 2011) (forward, 5′-CTCCTGCAGCGGCCTAAGAAGCC-3′; reverse, 5′-GCTGTGCTCCCTTCAGAGTAACTGTCCTCG-3′), Cav1.2 (forward, 5′-GGACACCCGGCCCTCTCCTCG-3′; reverse, 5′-ATGCAAGGGTAGGACTGTC-3′), Kv11.1 (forward, 5′-AAGATCCATCGGGATGACCTGCTGG-3′; reverse, 5′-GTTGCCGATTGAAGCCGCCCTC-3′), Kv7.1 (forward, 5′-ATCAGGCGCATGCAGTACTTTG-3′; reverse, 5′-TGACATCCCGCACGTCATAAG-3′) and iNOS (forward, 5′-ATGGCCTGCCCCTGGAATTTCC-3′; reverse, 5′-CTGTAGAGATGACTGGCTTTTCCCACG-3′). β-Actin (forward, 5′-CATCCGTAAGGACCTCTATGCCAAC-3′; reverse, 5′-ATAGAGCCGCCGATCCACA-3′) was used as a reference gene. Data analysis was carried out with Stratagene software as described by Stratagene.

Western blot analysis

After being removed and rinsed in sterilized PBS on ice, atrial tissues were homogenized in RIPA buffer (Thermo Fisher Scientific, Waltham, MA) and then centrifuged at 600×g for 15 min to pellet any insoluble material. When required, the membrane fractions were prepared as described previously (Matsuda et al., 1999; 2006). Thus, the supernatant was then centrifuged at 100 000×g for 60 min at 4°C. The membrane pellet was re-suspended in RIPA buffer and saved. Samples (10–20 µg) were run on 7.5% SDS-polyacrylamide gel and electrotransferred to PVDF filter membranes. To reduce non-specific binding, the membrane was blocked for 60 min at room temperature in Odyssey blocking buffer (LI-COR, Lincoln, NE). Thereafter, the membrane was incubated overnight at 4°C with primary antibodies for Cav1.2 (Millipore, Billerica, MA), Cav1.3 (Millipore), Kv11.1 (Millipore), Kv7.1 (Millipore), iNOS (Millipore), endothelial NOS (eNOS; Abcam, Cambridge, UK), α-adaptin (Thermo) and β-actin (Cell Signaling, Danvers, MA) (1:200–1000 dilution) in Odyssey blocking buffer. The membrane was washed six times with PBS with 0.1% Tween 20 and incubated with goat anti-rabbit IRDye 680 or goat anti-mouse IRDye 800 CW (LI-COR) diluted in 1:1500 in Odyssey blocking buffer for 60 min at room temperature. After being washed six times in PBS with 0.1% Tween 20, the blots were visualized using the Odyssey Infrared imaging system from LI-COR.

Imaging of intracellular NO in atrial cells

To detect NO production, we loaded atrial cells with 4,5-diaminofluorescein diacetate (DAF-2 DA; Sekisui Chemical CO., Osaka, Japan). This membrane-permeable dye is hydrolysed intracellularly by cytosolic esterases releasing DAF-2, which is converted in the presence of NO into a fluorescent product, DAF-2 triazole (Kojima et al., 1998; Nakatsubo et al., 1998). The DAF-2 DA was loaded at a concentration of 1 µM in Krebs–Ringer's phosphate (KRP) buffer [composition (mM): 120 NaCl, 4.8 KCl, 0.54 CaCl2, 1.2 MgSO4, 15.9 NaH2PO3 and 11 glucose; pH 7.2] in the dark at 37°C for 60 min. The nucleus was counterstained with Hoechst 33258 (Nacalai Tesque, Kyoto, Japan). Immunofluorescence images were traced by means of confocal laser scanning microscopy.

Immunofluorescence and confocal analysis

After atrial tissues had been removed and rinsed in sterilized PBS in ice, the tissues were fixed with 4% buffered formalin solution, immersed in sucrose solution, dipped into OCT compound (Sakura Finetechnical, Tokyo, Japan) and frozen at −20°C. The embedded tissues were sectioned at a thickness of 30 µm and air-dried. For immunofluorescence, rehydrated sections were incubated with the primary antibody, anti-Nav1.5 (Abcam), overnight at 4°C, followed by extensive washes with PBS and incubation with the secondary antibody conjugated to high quality flurophores, Alexa 488 or Alexa 568, overnight at 4°C. Immunofluorescence images were observed under Leica TCS-SP-5 confocal system (Leica, Wetzlar, Germany).

Statistical analysis

Data are expressed as means ± SEM. Statistical assessment of the data was performed using Student's t-test or a repeated-measure one-way anova followed by Bonferroni's multiple-comparison test when appropriate. A P-value of <0.05 was considered to be statistically significant.

Materials

Pentobarbital was supplied by Abbott Laboratories, Abbott Park, IL and nifedipine, E4031 and Chromanol 29B by Sigma-Aldrich, St. Louis, MO.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of interest
  9. References

Changes in action potentials and membrane currents

When action potentials in atrial myocytes were elicited by current injection at a rate of 2 Hz, action potential duration (APD) was shorter in myocytes from guinea pigs given i.p. injection of 300 µg·kg−1 LPS (Figure 1). The APD shortening was more marked at the 30% repolarization level (APD30) than at the 90% repolarization level (APD90). Thus, in the LPS treatment group, APD30 and APD90 were shortened by 51% and 23% respectively. The resting membrane potential and the maximum rate of rise of action potential (Vmax) remained unchanged in atrial myocytes from LPS-treated animals. In atrial myocytes from the animals treated with l-NAME co-administered with LPS, APD30 and APD90 were not significantly different from those of control myocytes. Treatment with the selective iNOS inhibitor EIT also significantly reversed the APD shortening seen in myocytes from LPS-treated animals.

image

Figure 1. Action potentials in atrial myocytes from control and LPS-challenged guinea pigs. Action potentials were elicited by current injection at a rate of 2 Hz. Animals treated with l-NAME or EIT received i.p. injections of l-NAME or EIT, at the same time as LPS. (A) Typical traces of action potentials are shown. (B, C) Bar graphs comparing APD30 and APD90 in atrial myocytes from control, LPS-treated, LPS/l-NAME-treated and LPS/EIT-treated animals. Data shown are means ± SEM of 10 cells from at least three different guinea pigs. *P < 0.05 significantly different from control. #P < 0.05 significantly different from LPS alone.

Download figure to PowerPoint

In each voltage-clamp experiment, membrane capacitance was measured immediately after disruption of the membrane patch. Membrane capacitance of atrial myocytes from LPS-treated animals (25.1 ± 2.2 pF, n = 20) did not differ from that of controls (30.9 ± 1.9 pF, n = 20). Typical tracings of ICa elicited by a depolarizing pulse from a holding potential of −30 to +10 mV in atrial myocytes from control, LPS-treated, LPS/l-NAME-treated and LPS/EIT-treated animals are shown in Figure 2A. The net ICa obtained in myocytes from LPS-treated animals was much smaller than in controls (Figure 2B). The estimated density of ICa was significantly greater in myocytes from control animals (4.69 ± 0.48 pA/pF) than those from LPS-treated animals (2.24 ± 0.32 pA/pF, n = 10 for each group).

image

Figure 2. Basal and isoprenaline-stimulated ICa in atrial myocytes from control and LPS-challenged guinea pigs with and without l-NAME or EIT treatment. (A) Current traces before and 5 min after exposure to 100 nM isoprenaline are superimposed. ICa was elicited by a 200 ms depolarizing test pulse to +10 mV from a holding potential of 30 mV. (B) Bar graph comparing basal and isoprenaline-stimulated ICa elicited by a depolarizing pulse from 30 to +10 mV in atrial myocytes from control, l-NAME-treated, LPS-treated and LPS/l-NAME-treated animals. *P < 0.05 significantly different from control. #P < 0.05 significantly different from LPS alone. (C) Current–voltage relations for control and septic atrial myocytes before and after exposure to 100 nM isoprenaline. There were significant differences between control and septic current–voltage relation curves in both the absence and presence of isoprenaline (P < 0.05). Data shown are means ± SEM of 10 cells from at least three different guinea pigs.

Download figure to PowerPoint

The addition of 100 nM isoprenaline resulted in a large increase in ICa in control myocytes (2.7-fold), which was less pronounced in the LPS-treated group (2.1-fold; Figure 2A,B). As illustrated in Figure 2C, the current–voltage relationships showed that isoprenaline produced a larger increase in ICa at each test pulse in control myocytes than in those from LPS-treated animals. Treatment with l-NAME or EIT had no effect on the response to isoprenaline on ICa in the control group but significantly normalized basal and isoprenaline-stimulated ICa in the LPS-treated group (Figure 2).

Typical tracings of IK elicited by a 2 s depolarizing pulse from a holding membrane potential of −30 to +60 mV in atrial myocytes from control, LPS-treated, LPS/l-NAME-treated and LPS/EIT-treated animals are shown in Figure 3A. The activating current of IK was clearly larger in the LPS-treated group than in the control group. The current–voltage relationships in 10 different cells for each group showed that the outward current at each test pulse was significantly increased in the LPS-treated group compared with the control group (Figure 3B). The increase in IK was not observed in myocytes from the animals when l-NAME or EIT was given together with LPS (Figure 3).

image

Figure 3. Effect of treatment with l-NAME or EIT on the change in IK in atrial myocytes from LPS-challenged guinea pigs. (A) Typical tracings of IK elicited by a 2 s depolarizing test pulse to +60 mV from a holding potential of 30 mV. (B) Current–voltage relations in atrial myocytes from control, LPS-treated and LPS/L-NAME-treated animals. In this current–voltage relation curves, significant differences were found between control and LPS-treated animals (P < 0.05), and also between LPS-treated and LPS/l-NAME-treated animals (P < 0.05). Data shown are means ± SEM of 10 cells from at least three different guinea pigs.

Download figure to PowerPoint

After the rapid component of IK (IKr) was blocked by its specific blocker E-4031 (5 µM), the residual major part of the current was eliminated by further administration of 30 µM chromanol 293B, an inhibitor of the slow component of IK (IKs) in both atrial myocytes from control and LPS-treated animals (Figure 4A). Conversely, in the presence of chromanol 293B, further addition of E-4031 had little effect on the activating current of IK in the two different myocytes (Figure 4B). These observations are indicative of a predominant increase in IKs in atrial myocytes from LPS-treated animals.

image

Figure 4. Changes in IK in the presence of the IKr inhibitor E-4031 and/or the IKs inhibitor chromanol 293B in atrial myocytes from control and LPS-challenged guinea pigs. (A) Cells were initially given 5 µM E-4031 followed by 30 µM chromanol 293B. (B) Cells were initially given 30 µM chromanol 293B followed by 5 µM E-4031.

Download figure to PowerPoint

Changes in atrial expression of channel subunits

Cav1.2 and Cav1.3 phenotypes encode dihydropyridine-sensitive L-type Ca2+ channels in atrial myocytes (Zhang et al., 2005), but the dominant high-voltage activated Ca2+ channels are Cav1.2 (Bohn et al., 2000). Western blot analysis showed significantly lowered expression of Cav1.2 (Figure 5A) and Cav1.3 (data not shown) protein in atrial tissues from LPS-treated guinea pigs. l-NAME treatment did not significantly affect the changes in protein levels of Cav1.2 and Cav1.3 in atrial tissues from LPS-treated animals. Quantitative real-time PCR revealed that Cav1.2 mRNA was less abundantly expressed in atrial tissues from LPS-treated animals compared with controls (Figure 5B). This reduced expression of Cav1.2 mRNA was unaffected by l-NAME, given together with LPS.

image

Figure 5. Changes in atrial expression of the Ca2+ channel subunit Cav1.2 in LPS-challenged guinea pigs. (A) Atrial expression of Cav1.2 protein levels in control and LPS-challenged guinea pigs was compared with the relative level using the endocytic protein α-adaptin. Typical Western blots of Cav1.2 and α-adaptin are shown in the top trace. Data shown are means ± SEM of four separate experiments. *P < 0.05 significantly different from control. (B) Real-time PCR analysis of Cav1.2 mRNA expression in atrial tissues from control, LPS-treated, and LPS/l-NAME-treated animals. The Cav1.2 mRNA levels are normalized to β-actin mRNA levels. Data shown are means ± SEM of six separate experiments. *P < 0.05 significantly different from control.

Download figure to PowerPoint

The Kv11.1 subunit is the principal component of the rapid delayed rectifier K+ channel, while Kv7.1 forms the slow component of the delayed rectifier K+ channel (Mitcheson and Sanguinetti, 1999; Sanguinetti, 2010). When we determined Kv11.1 and Kv7.1 protein expression in atrial tissues by Western blotting, induction of sepsis by LPS moderately, but significantly, increased these K+ channel proteins (Figure 6A). L-NAME treatment did not alter Kv11.1 protein expression but resulted in a significant return of Kv7.1 protein expression to control levels in atrial tissues from LPS-treated animals. In contrast to protein expression, the mRNA expression levels of Kv11.1 and Kv7.1 remained substantially unchanged in atrial tissues from LPS-injected guinea pigs regardless of whether l-NAME was co-administered (Figure 6B).

image

Figure 6. Changes in atrial expression of delayed rectifier K+ channel subunits (Kv11.1 and Kv7.1) in LPS-challenged guinea pigs. (A) Atrial expression of delayed rectifier K+ channel subunit protein levels in control and LPS-challenged guinea pigs was compared with the endocytic protein α-adaptin. Typical Western blots of Kv11.1, Kv7.1, and α-adaptin are shown in the top trace. Data shown are means ± SEM of six separate experiments. *P < 0.05 significantly different from control. #P < 0.05 significantly different from LPS alone. (B) Real-time PCR analysis of delayed rectifier K+ channel subunit mRNA expression in atrial tissues from control, LPS-treated and LPS/l-NAME-treated guinea pigs. The delayed rectifier K+ channel subunit mRNA levels are normalized to β-actin mRNA levels. Data shown are means ± SEM of six separate experiments. No significant differences in the atrial expression of delayed rectifier K+ channel subunit mRNA among groups was noted.

Download figure to PowerPoint

We also assessed mRNA and protein expression of the cardiac Na+ channel isoform Nav1.5 in atrial tissues using real-time PCR and immunofluorescence assay respectively. No significant difference was observed between the control and LPS-treated groups (data not shown).

Changes in iNOS and eNOS expression and NO production

On Western blots, atrial tissues from control guinea pigs showed a low level of 130 kDa iNOS protein, which doubled in those from LPS-treated animals (Figure 7A). Higher expression of iNOS in atrial tissues from LPS-treated animals was also found at the mRNA level (Figure 7B). No significant difference in atrial eNOS expression levels was found between the control and LPS-treated groups (Figure 7C).

image

Figure 7. Changes in atrial expression of iNOS and eNOS in LPS-challenged guinea pigs. (A) Atrial expression of iNOS protein levels in control and LPS-challenged guinea pigs was compared with the housekeeping protein β-actin. Typical Western blots of iNOS and β-actin are shown in the top trace. Data shown are means ± SEM of three separate experiments. *P < 0.05 significantly different from control. (B) Real-time PCR analysis of iNOS mRNA expression in atrial tissues from control and LPS-challenged guinea pigs. The iNOS mRNA levels are normalized to β-actin mRNA levels. Data shown are means ± SEM of six separate experiments. *P < 0.05 significantly different from control. (C) Atrial expression of eNOS protein levels in control and LPS-challenged guinea pigs was compared with β-actin. Typical Western blots of eNOS and β-actin are shown in the top trace. Data shown are means ± SEM of four separate experiments.

Download figure to PowerPoint

We assessed whether NO production actually increases in atrial cells from LPS-treated guinea pigs, using the fluorescent dye DAF-2 DA (Figure 8). A strong intensity of staining was observed in atrial cells from LPS-treated animals. When l-NAME or EIT was given together with LPS, the intensity of staining declined to the control level.

image

Figure 8. NO production by atrial cells from control and LPS-challenged guinea pigs. When the animals were treated with l-NAME or EIT, i.p. injection of l-NAME or EIT was given together with LPS. Intracellular NO is visualized with the use of the NO-sensitive dye DAF-2 DA (green). Nuclei were stained with Hoechst 33258 (blue). Representative images from two separate experiments are shown.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of interest
  9. References

Earlier studies reported a significant reduction in ICa in ventricular myocytes isolated from LPS-injected guinea pigs and rats (Zhong et al., 1997; Abi-Gerges et al., 1999). Thus, in vivo administration of LPS has been shown to result in reduced Ca2+ influx through L-type Ca2+ channels in isolated ventricular myocytes. Although whole-cell voltage-clamp experiments may have technical limitations inherent to measurement of ICa, we revealed a considerable reduction in ICa in atrial myocytes isolated from LPS-injected guinea pigs. This was clear in our assays of ICa elicited by a depolarizing pulse from −30 to +10 mV. A reduction in ICa of atrial myocytes from LPS-treated guinea pigs did not result from alterations in the voltage-dependent properties of the L-type Ca2+ channel, because currents of both atrial myocytes from control and LPS-treated animals had similar voltage dependence, as indicated by the similar current–voltage relationships. Abnormal temperature, heart rate, respiratory rate and white blood cell count observed in LPS-treated guinea pigs suggest that this animal model fulfils several criteria of clinical sepsis. Therefore, the host response to sepsis involved a reduction in ICa of atrial myocytes.

In sepsis, pro-inflammatory stimuli lead to over-expression of iNOS in many organs. Excessive NO generated by iNOS is implicated in the pathophysiology of sepsis (Titheradge, 1999). In this study, we also observed a significant increment of gene and protein expression of iNOS and a clear overproduction of NO in atrial cells from our animal model of sepsis. Importantly, when l-NAME or EIT was given together with LPS to inhibit iNOS activity, the reduction in ICa of atrial myocytes was abolished. Recent studies suggest that reactive nitrogen species, including NO, peroxynitrite (ONOO-) and nitrogen dioxide (NO2), can modulate ion channel structure and function (Matalon et al., 2003). These reactive nitrogen species are known to mediate protein tyrosine nitration (Radi, 2004). Indeed, our immunofluorescence study showed that sections from LPS-induced septic guinea pigs exhibited significant immunostaining with the polyclonal antibody to nitrotyrosine (unpublished observations). Inflammatory colitis results in the nitration of Y1837 and Y2134 within the C-terminus of human smooth muscle Ca2+ channel, Cav1.2b, and thereby impairs the ability of the tyrosine kinase, c-src kinase, to phosphorylate the Ca2+ channel, leading to diminished Ca2+ current and muscle contraction during colonic inflammation (Ross et al., 2007). We thus assume that iNOS-induced over-productive NO during sepsis may act as a means of increasing tyrosine nitration of L-type Ca2+ channels in atrial myocytes, leading to impaired function of the channels. Interestingly, the Cav1.2 subunit of the L-type Ca2+ channel was hypernitrosylated in left atrial tissues from patients with atrial fibrillation (Carnes et al., 2007).

In atrial myocytes from septic guinea pigs, the APD was significantly shortened, without changes in the resting membrane potential and the maximum rate of rise of action potential. Similar to its effect on ICa, l-NAME or EIT treatment abolished the APD shortening seen in atrial myocytes from septic animals. This suggests that the shortening of APD was associated with the iNOS-mediated increases in NO. Alternatively, the sepsis-related atrial APD shortening might have arisen from a reduction in ICa due to tyrosine nitration of L-type Ca2+ channels. A decrease in ICa, which shortens the APD, is one of electrophysiological alterations associated with the genesis of atrial tachyarrhythmias, including atrial fibrillation (Skasa et al., 2001; Van Wagoner, 2008).

The present study showed that the increasing effect of β-adrenoceptor stimulation on ICa was markedly depressed in atrial myocytes from septic animals. This finding is in marked contrast with a previous report that β-adrenoceptor stimulation reversed the reduction in ICa in ventricular myocytes 4 h after LPS injection (Zhong et al., 1997). However, the β-adrenoceptor control of ICa may be modified in a time-dependent manner after LPS challenge. Abi-Gerges et al. (1999) found that the decrease of ICa in ventricular myocytes from LPS-treated rat was compensated by an early potentiation of the β-adrenoceptor response, but this stimulatory effect on ICa declined at the late stage of sepsis. In other words, the regulatory influence of sepsis on the β-adrenoceptor response on ICa may be different between atrial and ventricular myocytes. When l-NAME or EIT was co-administered with LPS, the ICa response to β-adrenoceptor stimulation in atrial myocytes was nearly normalized, suggesting that tyrosine nitration can markedly affect β-adrenoceptor-mediated, L-type Ca2+ channel function in atrial myocytes.

L-type Ca2+ channels are heteromultimeric complexes of a pore-forming, transmembrane-spanning α1-subunit, a disulphide-linked complex of α2- and δ-subunits and an intracellular β- and γ-subunits (Catterall, 1995; Walker and De Waard, 1998). The α1-subunit is the largest and incorporates the conduction pore, the voltage sensor, gating apparatus and the known sites of channel regulation by second messengers, drugs and toxins. Previous data suggest that cardiac tissues express mainly L-type Ca2+ channels containing the α1C subunit (Cav1.2), but those containing the α1D subunit (Cav1.3) are found in atrial tissues (Bohn et al., 2000; Zhang et al., 2005). In this study, we found that both Cav1.2 and Cav1.3 subunit expression were significantly down-regulated in atrial tissues after sepsis. However, this is likely to be independent of iNOS-related changes, because l-NAME treatment did not substantially reverse the changes in protein levels of Cav1.2 and Cav1.3 in atrial tissues from septic animals. In light of the fact that the sepsis-induced reduction in atrial ICa was prevented by l-NAME treatment, down-regulation of α1 subunits did not appear to contribute to the reduced ICa in atrial myocytes from septic animals.

In addition to the change in ICa, atrial myocytes from septic guinea pigs exhibited a significant increase in IK. Since IK is one of the major components that determine the timing of repolarization of cardiomyocytes, the increase in IK could also contribute to the shortening of APD in atrial myocytes from septic guinea pigs. Treatment with l-NAME reversed the IK increase seen in atrial myocytes from septic animals, implying involvement of iNOS-mediated over-production of NO in this electrophysiological change. It has been reported that nitrosylation increases IKs in a manner dependent on eNOS in guinea pig ventricular myocytes, resulting in a shortening of the APD (Bai et al., 2004; 2005). Furthermore, over-expression of CAPON (carboxy-terminal PDZ ligand of neuronal NOS), a binding protein for neuronal NOS, has been demonstrated to result in increased IKs and shortened APD in guinea pig ventricular myocytes (Chang et al., 2008). Our experiments, with the use of E-4031 and chromanol 293B, which selectively block IKr and IKs, respectively, suggest a predominant increase in IKs in atrial myocytes from septic guinea pigs. We observed that protein expression of Kv11.1 and Kv7.1, which underlie the rapid component and the slow component of the delayed rectifier K+ channel, respectively (Mitcheson and Sanguinetti, 1999; Sanguinetti, 2010), were increased in atrial tissues from septic animals. Interestingly, l-NAME treatment resulted in a significant return of protein expression of Kv7.1, which forms the slow component of the delayed rectifier K+ channel, to control levels in atrial tissues from septic animals However, the sepsis-induced increase in protein expression of Kv11.1 and Kv7.1 was significant but not striking, and their gene expression remained substantially unchanged in sepsis, regardless of treatment with l-NAME.

In conclusion, the data presented here indicate that atrial myocytes from septic guinea pigs display a shortening of the APD, resulting from a decrease in ICa and an increase in IK. These electrophysiological alterations appeared to be attributable to the nitration of the ion channels rather than to changes in atrial expression of the ion channels. Given the importance of APD shortening in the induction of arrhythmias, the present results may be relevant to understanding the mechanisms responsible for the occurrence of atrial tachyarrhythmias, including atrial fibrillation, in sepsis.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of interest
  9. References

This work was supported in part by a Grant-in-Aid for Scientific Research from the Ministry of Education, Culture, Sports, Science and Technology of Japan. We thank Toshio Fujimori for his excellent technical assistance.

Conflict of interest

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of interest
  9. References

All of the authors have no conflict of interest on this manuscript.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. Conflict of interest
  9. References