Author + information
- Received October 23, 2003
- Revision received April 5, 2004
- Accepted May 4, 2004
- Published online August 18, 2004.
- Noriko Inoue, MD*,
- Tomoko Ohkusa, MD, PhD*,
- Tomoko Nao, MD, PhD*,
- Jong-Kook Lee, MD, PhD†,
- Tomo Matsumoto, MD*,
- Yuji Hisamatsu, MD, PhD*,
- Takashi Satoh, MD*,
- Masafumi Yano, MD, PhD*,
- Kenji Yasui, MD, PhD†,
- Itsuo Kodama, MD, PhD† and
- Masunori Matsuzaki, MD, PhD, FACC*,* ()
- ↵*Reprint requests and correspondence:
Dr. Masunori Matsuzaki, Division of Cardiovascular Medicine, Department of Medical Bioregulation, Yamaguchi University Graduate School of Medicine, 1-1-1, Minami-kogushi, Ube, Yamaguchi 755-8505, Japan
Objectives The aim of this study was to investigate the effects of rapid electrical stimulation (RES) of contraction on the expression of connexin (Cx)43 gap junction in neonatal rat cultured ventricular myocytes and the consequent changes of conduction properties.
Background The expression and distribution of gap junctions in cardiac muscle can be changed readily under a variety of pathological conditions because of dynamic turnover of Cxs. The effects of RES of contraction on gap junction remodeling are not well understood.
Methods Neonatal rat ventricular myocytes cultured for five days were subjected to RES (field stimulation) at 3.0 Hz for up to 120 min.
Results Rapid electrical stimulation resulted in a significant upregulation of Cx43 (by ∼1.5-fold in protein and by ∼1.9-fold in messenger ribonucleic acid at 60 min). Immunoreactive signal of Cx43 was also increased. Angiotensin II (AngII) content was increased significantly by RES >15 min. Phosphorylated forms of extracellular signal-regulated protein kinase (ERK), c-Jun NH2-terminal kinases, and p38 mitogen-activated protein kinases (MAPKs) were all increased dramatically by RES with peaks at 5 ∼ 60 min. Propagation of excitation was visualized by extracellular potential mapping by using a multiple electrode array system. Conduction velocity was increased significantly by RES for 60 to 90 min (25% ∼ 27% increase). Treatment of myocytes with losartan (100 nmol/l) prevented most of these effects of RES; RES-induced upregulation of Cx43 was also prevented by specific inhibitors for ERK and p38 MAPKs.
Conclusions A short-term RES causes upregulation of Cx43 in cardiomyocytes and a concomitant increase of conduction velocity, mainly through an autocrine action of AngII to activate ERK and p38 MAPKs.
Cardiac myocytes are connected with each other electrically and metabolically through gap junction channels composed of connexins. Many animal and human studies have described alterations in the abundance or subcellular localization of connexin proteins under a variety of pathological conditions, including hypertrophy, heart failure, ischemic heart diseases, and atrial fibrillation (1,2). Many connexins, including connexin43 (Cx43), which is expressed most abundantly in the heart, have short half-lives of 1 ∼ 5 h, reflecting their rapid turnover through synthesis and degradation (3–5).
Zhuang et al. (6) demonstrated in monolayers of rat cultured ventricular myocytes that application of linear pulsatile stretch (10% resting length at 3 Hz) only for 1 h caused an increase of Cx43 expression by ∼2-fold, and the change was associated with a significant increase of conduction velocity. If the mechanical stretch on the myocytes plays a primary role in the upregulation of Cx43, an increase of twitch contraction per unit of time resulting from a sudden increase of heart rate is expected to cause a comparable mechanical stimuli to affect the expression of gap junctions. The present study was designed to test this hypothesis.
We investigated the effects of rapid electrical stimulation (RES) on the expression of Cx43 gap junction in neonatal rat cultured ventricular myocytes and the consequent changes of conduction properties. As to the signaling cascade, we focused on the role of angiotensin II (AngII) and mitogen-activated protein kinase (MAPK) family because AngII is known to be secreted from cardiac myocytes that have been subjected to mechanical stretch and to play a crucial role in autocrine action in stretch-induced hypertrophic responses through activation of multiple kinases, including MAPKs (7–9). The results have first revealed that a short-term RES of contraction causes a prominent increase of Cx43 expression (messenger ribonucleic acid [mRNA] and protein) in cardiac myocytes, mainly through autocrine action of AngII to activate MAPKs, and these observations are associated with substantial alterations of conduction properties.
All the animals were handled according to the guideline of the Animal Care Committee of Yamaguchi University. This investigation complied with the Guide for the Care and Use of Laboratory Animals published by the National Institutes of Health (NIH Publication No. 85-23, revised 1996).
Primary cultures of one- to two-day-old neonatal rat ventricular myocytes were prepared as reported previously (10). The myocytes dissociated enzymatically were seeded on culture trays at a concentration of 3 × 105cells/cm2, and then incubated in Leibovitz L-15 Medium (Worthington Biochemical Co., Lakewood, New Jersey) supplemented with 5% fetal bovine serum at 37°C. The cell suspension was replated to reduce the fibroblast content. Cells were grown in a random orientation (isotropic growth) to make a confluent layer, and by day 5 they showed synchronized spontaneous beating with a regular frequency. The relative population of myocytes over nonmyocytes in the respective culture, that was estimated by immunolabeling against antiactin antibody, was >97%.
RES of contraction
Cultured myocytes on day 5 were transferred to serum-free medium, and subjected to RES at 3.0 Hz for up to 120 min; RES was applied by field stimulation via platinum wire electrodes (0.5 mm in diameter) with an interpolar distance of 6.5 cm (for experiments of Cx43 expression analysis) or of 0.5 cm (for potential mapping experiments). The stimulation intensity was adjusted to a level twice the threshold for synchronous contraction. The pulse polarity was alternated one after another to minimize the electrolysis and possible generation of oxidative species. Ascorbic acid (250 μmol/l), a scavenger of oxidative species, was added to the culture medium (11). In experiments to elucidate the signaling cascade, we used a specific AngII type-1 receptor antagonist, losartan (gift from Merck & Co. Inc., Rahway, New Jersey), and specific MAPK inhibitors (PD98059, SP600125, and SB203580) (Calbiochem, San Diego, California). These compounds at appropriate concentrations were applied to the culture medium 30 min before the initiation of RES.
Extracellular potential mapping of propagation
Conduction properties of the cultured ventricular myocytes were examined by extracellular potential mapping by using a specifically designed multielectrode array system (MED-P545A and MED64, Panasonic, Tokyo, Japan) (12,13). The 64 planar microelectrodes, each having a size of 50 × 50 μm, were arranged in an 8 by 8 pattern with an interelectrode distance at 450 μm to cover a 3,150 × 3,150 μm area beneath the cultured myocytes. All of the 64 unipolar signals were recorded simultaneously with a common reference from four distant electrodes, and amplified with high gain (60 dB) and at a low frequency (1.0 ∼ 100 Hz) filter setting. Constant stimuli at 3.0 Hz were applied via four lateral electrodes of the recording array to elicit one-way propagation of excitation in the observation area before and after RES. Biphasic pulses of 0.1 ms duration were employed for the stimulation with an intensity of 100 to 200 μA (twice the diastolic threshold). Activation time at each recording site was identified by the sharp negative deflection. The conduction velocity was measured by plotting activation time against distance from the stimulation site along a midline from left to right.
Electrophoresis and immunoblotting
Collected cells were lysed, and the cellular extracts were electrophoresed on SDS-polyacrylamide gels. The following primary antibodies were used for immunodetection: rabbit anti-connexin43 antibody (Zymed, South San Francisco, California); goat anti-angiotensin I/II antibody (Santa Cruz, Santa Cruz, California); rabbit anti-extracellular signal-regulated protein kinases (ERK)1/2 antibody; rabbit anti-c-Jun NH2-terminal kinases (JNK)1 antibody; rabbit anti-p38 antibody (Santa Cruz, California); rabbit anti-active MAPK pAb; rabbit anti-active JNK pAb; rabbit anti-active p38 pAb (Promega, Madison, Wisconsin). The amount of protein recognized by the antibodies was quantified by means of an ECL immunoblotting detection system (Amersham, Buckinghamshire, England).
Reverse transcription and polymerase chain reaction amplification
Total cellular ribonucleic acid was isolated from each frozen cell sample by the acid guanidinium thiocyanate/phenol/chloroform extraction method (14). Complementary deoxyribonucleic acid was prepared using a Takara RNA PCR Kit (Takara, Tokyo, Japan) (15). The primers for the amplification of Cx43 and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were designed from published sequences. For Cx43, they were based on rat sequences (GenBank, accession no. P08050) (16): at positions 367 to 390 (sense primer, 5′-GTG GAG ATG CAC CTG AAG CAG ATT-3′) and 1,038 to 1,058 (antisense primer, 5′-AGT TGC TGC TGG ACA TGA ACT-3′) (predicted length of the PCR product, 671 base pairs [bp]). For GAPDH, they were based on rat sequences: at positions 102 to 125 (sense primer, 5′-CTT CAT TGA CCT CAA CTA CAT GGT-3′) and 805 to 828 (antisense primer, 5′-CTC AGT GTA GCC CAG GAT GCC CTT-3′) (predicted length of the PCR product, 726 bp). The GAPDH gene product was used as an internal control.
Cell size measurement and immunohistochemistry
Cardiomyocytes on culture slides were fixed and stained with hematoxylin-eosin for measurements of their dimension. The cell area of myocytes was measured in 10 random fields (50 cells/field) with the aid of NIH Image (Bethesda, Maryland).
For immunodetection of gap junctions, a polyclonal rabbit anti-Cx43 antibody (Zymed, California) was used. After permeabilization (0.1% Triton X-100), quenching and blocking (10% goat serum), samples were incubated with the antibody (1:100 diluted in PBS) overnight at room temperature. Primary antibody-bound Cx43 was visualized by FITC-conjugated anti-rabbit IgG, and examined using a confocal microscope (LSM500, Zeiss, Germany). Samples processed without primary antibody served as negative control. The proportion of Cx43 immunoreactive signal was defined as the number of high signal intensity (>90% of the maximum level) pixels divided by total number of pixels occupied by the myocytes (6). The 10 individual measurements at different fields were averaged to yield a single value for each culture slide.
Measurement of AngII concentration
Angiotensin II contents in culture media under the respective condition were measured by radioimmunoassay as previously described (17).
All data are presented as the mean ± SD. Differences in continuously distributed variables between predetermined experimental groups were analyzed using one-way analysis of variance followed by Dunnett's multiple comparison method with SPSS software (SPSS Inc., Chicago, Illinois). We did not model experimental design as repeated measures. Differences were taken to be significant at p < 0.05.
Effects of RES on cell size and spontaneous beating rate of cultured cardiomyocytes
Cell size was measured in each eight culture slides. Cell area at the baseline was 610 ± 141 μm2. The values at 15, 30, 60, 90, and 120 min after RES were 620 ± 134, 570 ± 119, 594 ± 124, 626 ± 138, and 547 ± 91 μm2, respectively, and they were not significantly different from the baseline. There were no appreciable differences in the cell density among the groups. The baseline spontaneous beating rate was 130 ± 10 beats/min (n = 8). Rapid electrical stimulation for 15 to 120 min had no significant effects on the spontaneous beating rate.
Effects of RES on Cx43 expression
In the immunoblotting (Fig. 1A),the Cx43 antibody recognized three bands migrating between 40 and 43 kDa. The amount of Cx43 protein normalized to the value for control protein was increased gradually by RES, and the change reached a statistical significance at 60 min (by 1.53 ± 0.11-fold, n = 8, p < 0.01). The amount of Cx43 mRNA was also increased by RES up to 60 min (by 1.90 ± 0.73-fold, n = 8, p < 0.01) (Fig. 1B). The increase of Cx43 protein reached a peak at 90 min, and then began to decline. The mRNA returned to the baseline already at 90 min.
Figure 2Ashows confocal images of Cx43 immunolabeling. Connexin43-containing gap junctions were visualized as aggregates of bright punctate fluorescent domains around the cell perimeter at the abutment with neighboring cells. Rapid electrical stimulation for 60 to 90 min resulted in a homogeneous increase of the Cx43 immunoreactive signals without affecting their distribution pattern. Quantitative analysis revealed a significant increase in the Cx43-positive area by RES (by 1.53 ± 0.18-fold at 60 min, and 1.71 ± 0.55-fold at 90 min) (Fig. 2B).
Effects of RES on AngII production and secretion by cardiomyocytes
We examined the effect of RES on AngII content in the culture media and cardiomyocytes. The AngII content in the culture media was increased remarkably by RES for 15 min, and the elevation was maintained thereafter (Fig. 3A).Figure 3B shows the protein amount of AngII in cardiomyocytes estimated by immunoblotting. Angiotensin II was increased significantly after RES for 15 to 60 min (by 1.6 ∼ 1.9-fold). Further prolongation of RES was associated with a gradual return of the AngII amount toward the baseline. These observations indicate rapid increases in production and secretion of AngII by cardiomyocytes subjected to RES.
Effects of RES on ERK, JNK, and p38 MAPK
We next examined three major MAPK signaling pathways downstream of AngII. The total ERK level was increased significantly by RES for 5 min (by 1.6-fold) (Fig. 4A).Phosphorylated ERK (p-ERK) showed a much greater increase in response to RES with dual peaks at 5 min (by 2.1 ± 1.1-fold, p < 0.05) and at 60 min (by 3.2 ± 0.8-fold, p < 0.01) (Fig. 4A). Although the total JNK level was unaffected, phosphorylated form of JNK (p-JNK) was increased dramatically in response to RES with dual peaks at 15 min (by 3.5 ± 1.6-fold, p < 0.01) and at 60 min (by 4.1 ± 1.3-fold, p < 0.01) (Fig. 4B). The total p38 MAPKs (total p38) was unchanged, but phosphorylated form of p38 MAPKs (p-p38) showed the most pronounced and rapid increase in response to RES with a single peak at 5 min (by 6.5 ± 1.0-fold, p < 0.01) (Fig. 4C). These results suggest that MAPK signaling pathways must be activated by RES.
Prevention of RES-induced Cx43 upregulation and MAPK activation by losartan
If RES-induced Cx43 upregulation is the result of autocrine action of AngII to activate MAPKs, the increase of Cx43 expression and the level of activated forms of MAPKs should be prevented by pharmacologic blockade of AngII type 1 receptor. Figure 5shows the effects of RES on the Cx43 expression in the presence of losartan. The baseline protein and mRNA levels of Cx43 were unaffected by treatment of the cardiomyocytes with losartan (100 nmol/l). In the presence of losartan, unlike in the absence of compound, RES for 60 or 90 min did not cause significant increases in protein (Fig. 5A) and mRNA (Fig. 5B) of Cx43. In the immunohistochemistry, RES for 30 to 90 min in the presence of losartan did not cause significant changes in the Cx43-positive area (Fig. 5C).
Figure 6shows the effects of RES on the activated forms of MAPKs. In the presence of losartan, RES for 60 to 90 min did not cause significant increase of p-ERK and p-JNK. The level of p-p38 was increased by RES for 5 min even in the presence of losartan (by 2.9 ± 1.1-fold, p < 0.05), but the increases were much less than that in the absence of losartan. The p-p38 response to RES for 15 min was abolished by losartan. These results indicate effective prevention of RES-induced Cx43 upregulation and MAPK activation by AngII type 1 receptor blockade.
To examine the relative contribution of different MAPKs, RES was applied to cardiomyocytes pretreated with specific inhibitors for each branch of the MAPKs (Fig. 7).The baseline protein level of Cx43 was unaffected by either PD98059 (50 μmol/l), SP600125 (50 μmol/l), or SB203580 (10 μmol/l). In the presence of PD98059, an inhibitor of MAPK/ERK, the RES-induced increase of Cx43 protein was almost abolished. Pretreatment with SP600125, a selective inhibitor of JNK pathway, in contrast, did not affect the RES-induced Cx43 upregulation. Pretreatment with SB203580, an inhibitor of p38 MAPKs and JNK2 pathway caused a partial prevention of the Cx43 upregulation. These observations suggest that the RES-induced Cx43 upregulation is mediated mainly by an activation of ERK and p38 MAPK pathways.
Effects of RES on the propagation of excitation
We investigated functional manifestation by multielectrode extracellular potential mapping. The measurements were made before and after RES. A clear local electrogram composed of a positive deflection followed by negative deflection was recorded from most (>90%) of the 64 electrodes (Fig. 8A).Isochrone maps of activation time showed almost uniform propagation of excitation from left to right both before (control) (Fig. 8B) and after RES for 90 min (Fig. 8C). The activation times after RES at the respective recording sites were much shorter than those of control, indicating the RES-induced acceleration of propagation. Such changes in propagation after RES were prevented by pretreatment with losartan (100 nmol/l) (Fig. 8D). Figure 9shows the pooled data of conduction velocity. In the absence of losartan, the conduction velocity was increased by RES >30 min; the values at 60 min and at 90 min after RES were significantly higher than control. In the presence of losartan, RES did not affect the conduction velocity.
In the present study, we showed that a short-term RES of contraction caused a prominent increase of Cx43 expression in cardiac myocytes. This was associated with an increase of AngII production and secretion by the myocytes as well as upregulation of activated forms of MAPKs. Parallel elevation of mRNA and protein suggest that an increase of synthesis may play an important role in the upregulation of Cx43. Extracellular potential mapping revealed an acceleration of propagation by RES. Most of these effects of RES were prevented by losartan, an AT1 blocker.
Signaling pathway for the RES-induced Cx43 upregulation
In cultured neonatal rat cardiomyocytes, several MAPK family members were shown to be activated in response to pulsatile stretch (9). Extracellular signal-regulated protein kinases, JNK, and p38 are activated by G-protein-coupled receptor agonists such as AngII or endothelin-1 (18). Extracellular signal-regulated protein kinases, JNK, and p38 can also be activated by a variety of cellular stress such as hyperosmotic shock, protein synthesis inhibitors, hypoxia/reoxygenation, and reactive oxygen species (18). Our results have revealed that RES does cause prompt and dramatic increases of the phosphorylated forms of ERK, JNK, and p38. In addition, most of the RES-induced activation of these MAPKs was prevented by losartan. The RES-induced upregulation of Cx43 protein was also prevented by the inhibitors of ERK and p38 MAPKs (PD98059 and SB203580), whereas it was unaffected by the inhibitor of JNK (SP600125). These observations seem to suggest pivotal roles of ERK and p38 MAPKs in the downstream of AngII signaling pathways. Polontchouk et al. (19) have demonstrated that application of exogenous AngII to cultured rat cardiomyoctes causes a specific upregulation of Cx43 within 24 h through signaling pathways including activation of ERK1/2 and p38. The longer time required for Cx43 upregulation might be due to less efficiency of exogenous AngII than its autocrine action in stimulating the downstream pathways.
We cannot exclude mediators other than AngII for the upregulation of Cx43, because the early activation of p38 by RES was not completely prevented by losartan (Fig. 6C). Those might include cAMP (3), Wnt-1/β-catenine (20), and vascular endothelial growth factor (21). More extensive experimental studies will be required to elucidate the exact signaling pathway for the RES-induced Cx43 upregulation.
Functional manifestation of the RES-induced Cx43 upregulation
The present data demonstrate that RES for 60 to 90 min causes a significant increase of the conduction velocity. If it is assumed that the increase in Cx43 immunoreactive signal (1.53- to 1.71-fold after 60 to 90 min RES) is proportional to the increase in gap junctional conductance, the increase of conduction velocity (25% to 27%) is attributed most likely to an upregulation of gap junction channels.
Several other possible mechanisms, however, should also be taken into account for the altered conduction properties after RES. Newly synthesized Cx43 gap junctions could have different functions than those of basal condition. Yao et al. (22) reported recently that remodeled gap junction channels in a border zone of healing infarcts had lower conductance and altered voltage-dependence. Potential upregulation of other connexins including Cx45 (5) remains to be investigated. Post-translational changes in Cx43 may also affect the electrical coupling through gap junctions. Connexin43 can be phosphorylated by protein kinase A, C, G, and mitogen-activated protein kinase, and the sarcoma protein kinase (5,23). Actions of these kinases usually lead to diminished or increased single-channel or whole-cell conductances, but the responsible mechanisms are poorly defined (5,23). Phosphorylation is also involved in the connexin stability to form gap junction channels. Increasing evidence has focused attention on this process in remodeling of gap junctions in diseased hearts (5,23). Recently, Beardslee et al. (24) showed in a model of acute ischemia that uncoupling of cardiac tissue is associated with dephosphorylation of Cx43 and accumulation of unphosphorylated Cx43 in the gap junction plaques. Rapid electrical stimulation of contraction might affect the conduction properties through a modulation of Cx43 phosphorylation, and it will be an interesting issue to be studied in the future.
Impulse propagation in cardiac tissue is generally assumed to be impaired by a reduction of intercellular electrical coupling. If the tissue contains cellular discontinuities, however, the modification is more complex. Rohr et al. (25) have demonstrated in patterned growth myocyte culture that partial cellular uncoupling causes paradoxical improvement of impulse conduction at the site of abrupt geometric expansion. Accordingly, an increase in electrical cell-to-cell coupling by RES could jeopardize conduction at the corresponding sites in the heart, leading to unidirectional conduction block and reentrant arrhythmias. Proarrhythmic propensity after tachycardia might be attributed, in part, to such altered conduction properties.
The present study has first revealed that a short-term RES of contraction causes a prominent increase of Cx43 expression, which is mediated mainly by AngII-MAPKs pathways, and it is accompanied by alterations of conduction properties.
The authors thank Ms. Keiko Miwa (Nagoya University) and Dr. Ken Shimono (Alpha MED Sciences Co., Ltd., Tokyo) for their technical assistance.
Supported, in part, by a grant-in-aid for scientific research in Japan (C12670671 and C15590753) from the Ministry of Education, Vehicle Racing Commemorative Foundation, Japan, and Japan Foundation of Cardiovascular Research. Drs. Inoue and Ohkusa contributed equally to this work.
- Abbreviations and acronyms
- angiotensin II
- extracellular signal-regulated protein kinases
- c-Jun NH2-terminal kinases
- mitogen-activated protein kinase
- messenger ribonucleic acid
- rapid electrical stimulation
- Received October 23, 2003.
- Revision received April 5, 2004.
- Accepted May 4, 2004.
- American College of Cardiology Foundation
- Peters N.S.,
- Green C.R.,
- Poole-Wilson P.A.,
- Severs N.J.
- Jongsma H.J.,
- Wilders R.
- Darrow B.J.,
- Fast V.G.,
- Kleber A.G.,
- Beyer E.C.,
- Saffitz J.E.
- Dodge S.M.,
- Beardslee M.A.,
- Darrow B.J.,
- Green K.G.,
- Beyer E.C.,
- Saffitz J.E.
- Saffitz J.E.,
- Laing J.G.,
- Yamada K.A.
- Zhuang J.,
- Yamada K.A.,
- Saffitz J.E.,
- Kleber A.G.
- Malhotra R.,
- Sadoshima J.,
- Brosius F.,
- Izumo S.
- Yonemochi H.,
- Yasunaga S.,
- Teshima Y,
- et al.
- Shimono K.,
- Brucher F.,
- Granger R.,
- Lynch G.,
- Taketani M.
- Mullis K.,
- Faloona F.,
- Scharf S.,
- Saiki R.K.,
- Horn G.,
- Erlich H.
- Ohkusa T.,
- Ueyama T.,
- Yamada J,
- et al.
- Beyer E.C.,
- Paul D.L.,
- Goodenough D.A.
- Sugden P.H.,
- Clerk A.
- Polontchouk L.,
- Ebelt B.,
- Jackels M.,
- Dhein S.
- Pimental R.C.,
- Yamada K.A.,
- Kleber A.G.,
- Saffitz J.E.
- Yao J.-A.,
- Hussain W.,
- Patel P.,
- Peters N.S.,
- Boyden P.A.,
- Wit A.L.
- van Veen T.A.B.,
- van Rijen H.V.M.,
- Opthof T.
- Beardslee M.A.,
- Lermer D.L.,
- Tadros P.N,
- et al.
- Rohr S.,
- Kucera J.P.,
- Fast V.G.,
- Kleber A.G.