Contribution of KChIP2 to the developmental increase in transient outward current of rat cardiomyocytes

Contribution of KChIP2 to the developmental increase in transient outward current of rat cardiomyocytes

Journal of Molecular and Cellular Cardiology 35 (2003) 1073–1082 www.elsevier.com/locate/yjmcc Original Article Contribution of KChIP2 to the develo...

217KB Sizes 0 Downloads 10 Views

Journal of Molecular and Cellular Cardiology 35 (2003) 1073–1082 www.elsevier.com/locate/yjmcc

Original Article

Contribution of KChIP2 to the developmental increase in transient outward current of rat cardiomyocytes Takeshi Kobayashi a,b,*, Yoichi Yamada a, Masato Nagashima a, Sumihiko Seki a, Masaaki Tsutsuura a, Yoshinori Ito c, Ichiro Sakuma d, Hirofumi Hamada c, Tomio Abe b, Noritsugu Tohse a a

Department of Cellular Physiology and Signal Transduction, Sapporo Medical University School of Medicine, South 1 West 17, Chuo-ku, Sapporo 060 8556, Japan b Department of Thoracic and Cardiovascular Surgery, Sapporo Medical University School of Medicine, Sapporo 060 8543, Japan c Department of Molecular Medicine, Sapporo Medical University School of Medicine, Sapporo 060 8556, Japan d Department of Cardiovascular Medicine, Hokkaido University School of Medicine, Sapporo 060 0815, Japan Received 11 February 2003; received in revised form 7 May 2003; accepted 20 May 2003

Abstract The Ca2+-independent, voltage-gated transient outward current (Ito) displays a marked increase during development of cardiomyocytes. However, the molecular mechanism remained unclear. In rat adult ventricular myocytes, Ito can be divided into a fast (Ito,f) and a slow (Ito,s) component by recovery process from inactivation. Voltage-gated K+ channel-interacting proteins 2 (KChIP2) has recently been shown to modify membrane expressions and current densities of Ito,f. Here we examined the developmental change of Ito and the putative molecular correlates of Ito,f (Kv4.2 and Kv4.3) and KChIP2 in rat ventricular myocytes. Even in rat embryonic day 12 (E12) myocytes, we detected Ito. However, Ito in E12 was solely composed of Ito,s. In postnatal day 10 (P10), we recorded much increased Ito composed of two components (Ito,f and Ito,s), and Ito,f was dominant. Thus, the developmental increase of Ito from E12 to P10 can be explained by the dramatic appearance of Ito,f. Real-time RT-PCR revealed that Kv4.2 and Kv4.3 mRNA levels were slightly changed. By contrast, KChIP2 mRNA level increased from E12 to P10 by 731-fold. Therefore, the huge increase of KChIP2 expression was likely to be the cause of the great increase of Ito,f. In order to confirm that KChIP2 is crucial to induce Ito,f, we used adenoviral gene transfer technique. When KChIP2 was over-expressed in E12 myocytes, a great amplitude of Ito,f appeared. Immunocytochemical experiments also demonstrated that KChIP2 enhanced the trafficking of Kv4.2 channels to cell surface. These results indicate that KChIP2 plays an important role in the generation of functional Ito,f channels during development. © 2003 Elsevier Ltd. All rights reserved. Keywords: Ion channels; Potassium; Genes; Cardiomyocytes; KChIP2

1. Introduction The Ca2+-independent, voltage-gated transient outward current (Ito) underlies the early phase of action potential repolarization in mammalian cardiomyocytes and a notch (dip) at phase 1 of action potential in human and canine cardiomyocytes [1,2]. In the mammalian ventricular myocytes, Ito is composed of at least two components; one is characterized by a relatively fast recovery from inactivation (Ito,f), and another by a slow recovery from inactivation (Ito,s) * Corresponding author. Tel.: +81-11-611-2111x2652; fax: +81-11-618-3280. E-mail address: [email protected] (T. Kobayashi). © 2003 Elsevier Ltd. All rights reserved. doi:10.1016/S0022-2828(03)00199-8

[3–5]. The molecular correlate of Ito,f is thought to be Kv4.2/4.3, and that of Ito,s is Kv1.4 [5,6]. Previous studies reported that Ito was not observed in rat embryonic myocytes [7] but that it increased dramatically after birth [8–11]. Contribution of Ito,f component in Ito is relatively smaller in rat neonatal myocytes [5,12,13] than that in adult myocytes [2,3,14]. Thus, the developmental increase in Ito is thought to be associated with the increase in Ito,f. However, the developmental change in Kv4 proteins reported previously was not enough to explain the increase in density of Ito,f [11]. The increase in Ito between days 5 and 30 in postnatal period (P5–P30) was much larger than that expected from Kv4.2 protein expression patterns [11]. These

1074

T. Kobayashi et al. / Journal of Molecular and Cellular Cardiology 35 (2003) 1073–1082

findings led us to speculate that another accessory subunit plays a role in the developmental change of Ito,f Recently, voltage-gated K+ channel-interacting proteins (KChIPs) were identified by a yeast two-hybrid system as proteins that interacted Kv4 channels [15]. KChIP2, a member of KChIP2 family, was preferentially expressed in hearts [15–17]. When co-expressed with Kv4 channels in mammalian cells, KChIPs increase Ito (8–55-fold) by enhancing the trafficking of channels from the endoplasmic reticulum to the plasma membrane [15,16]. In cardiomyocytes from KChIP2deficient mice, almost no Ito was observed [18]. Therefore, co-expression of Kv4 and KChIPs are thought to be necessary for functioning of Ito. In this study, we evaluated the relationship between KChIP2 and Ito,f during development. The present results indicated that KChIP2 is responsible for the appearance of Ito,f with development. 2. Materials and methods 2.1. Preparation of single cardiac ventricular myocytes in E12 and P10 Freshly isolated single myocytes were prepared from ventricles of embryonic day 12 (E12) and postnatal day 10 (P10) Wister rats. The pregnant rat (12 d after impregnation) was anesthetized with ether for a few minutes and then decapitated as described previously [19]. The abdomen was opened and 10–15 of the fetuses were removed from the uterus. The hearts were obtained from fetuses in oxygenated Tyrode’s solution under a stereomicroscope. The composition of the normal Tyrode’s solution was (mmol/l): NaCl 143; KCl 5.4; CaCl2 1.8; MgCl2 0.5; NaH2PO4 0.33; glucose 5.5; HEPES 5.0 (pH 7.4 by NaOH). P10 rats were also anesthetized with ether and then decapitated. The hearts were removed quickly into oxygenated Tyrode’s solution. Ventricles of E12 and P10 rats were removed from the hearts. These ventricular tissues were immersed in Ca2+-free Tyrode’s solution at room temperature (>20 min), to remove extracellular Ca2+. The tissues were digested by collagenase (5 mg/ml, Wako Pure Chemical Industries) for 60 min at 37 °C. The digested tissues were transported to Kraftbrühe (KB) solution, and pipetteing was gently performed to make a cell suspension. The composition of KB solution was (mmol/l): L-glutamic acid 50; KCl 40; taurine 20; KH2PO4 20; MgCl2 3; glucose 10; EGTA 0.5; HEPES 10 (pH 7.4 by KOH). 2.2. Electrophysiological recordings and data analysis Isolated cardiac myocytes were continually perfused with the external solutions at a rate of 10 ml/min on the stage of an inverted microscope. The whole-cell membrane currents were recorded by the patch-clamp method, using glass-patch electrodes with a tip resistance of 2–4 MΩ. In order to isolate Ito, fast Na+ current (INa) and Ca2+ current (Ica) were blocked by means of replacing external Na+ and Ca2+ with choline

and Co2+ [20] (mmol/l): choline-Cl 143; KCl 5.4; CoCl2 2; MgCl2 0.5; glucose 5.5; HEPES 5.0 (pH 7.4 by Tris). The composition of the internal pipette solution was (mmol/l): K-aspartate 110; KCl 20; MgCl2 1.0; ATP-K2 5.0; phosphcreatine-K2 5.0; EGTA 5; HEPES 5.0 (pH 7.4 by KOH). The temperature of the external solutions was kept at 36–37 °C. All data were corrected for a liquid junction potential of –7 mV. Analysis and voltage protocols were performed with the use of an Axopatch 1D amplifier/Digidata 1322A interface (Clampex software, pCLAMP 8.1, Axon Instruments Inc, Union City, CA, USA) [20,21]. Current signals were filtered at 2 kHz and digitized at 10 kHz. Current amplitudes of Ito were defined as the difference between the peak and the sustained current and they were normalized to cell capacitance (Cm) to correct for different cell sizes. 2.3. Cloning and generation of recombinant adenoviral vectors of rat-KChIP2 Adult Wister rats weighing 250–350 g were anesthetized and killed with an excessive amount of pentobarbital, and the cardiac ventricles were removed as quickly as possible and frozen in liquid nitrogen. Total RNA was extracted using TRIzol® reagent (Invitrogen) according to the manufacturer’s instructions. Total RNAs (2 µg) of rat hearts were reverse transcribed (RT) with Superscript II reverse transcriptase (Invitrogen), using oligo(dT) primers. PCR was performed by using 1 µl of each RT product as template DNA using Taq DNA polymerase (Invitrogen) as described previously [21]. The primers for rat-KChIP2b were designed on the basis of the published sequences of rat-KChIP2b (GenBank™ accession number AF269284) and are shown in Table 1. The specificity of the PCR products was confirmed by a DNA sequencer (373 DNA Sequencing System, Applied Biosystems, CA, USA) after they had been subcloned into pCR®-XL-TOPO vector (Invitrogen), which generated pCR-XL-TOPOKChIP2. To express functional KChIP2 in native rat myocytes, pCR-XL-TOPO-KChIP2 was digested by BamHI and NotI and subcloned into pIRES-hrGFP-1a (Stratagene), which generated pIRES-KChIP2-hrGFP. We chose pIREShrGFP-1a as an expression vector because this contained an IRES between the multiple cloning site and hrGFP, allowing the expression of our gene of interest, i.e. KChIP2 to be monitored at the single-cell level due to expression of hrGFP on the same transcript. We constructed recombinant adenoviral vectors by standard procedures as described previously [22,23]. Briefly, the pIRES-KChIP2-hrGFP was digested by NsiI and MluI and was blunted by T4 DNA polymerase (New England Bio Labs). This fragment was inserted into the SwaI site of the cosmid pAxcw (Takara) [22,23]. This cosmid was cotransfected with the genomic DNA–terminal protein complex of adenovirus type 5 (Ad5dlX), and the recombinant adenovirus (Ad-KChIP2-hrGFP) was generated as described

T. Kobayashi et al. / Journal of Molecular and Cellular Cardiology 35 (2003) 1073–1082 Table 1 DNA primers used for RT-PCR and TaqMan probes Gene

NIH PCR primer and TaqMan probe GenBank accession number rat-KChIP2b AF269284 Forward (full length) 5'-ATGCGGGGCCAGGGCCGCAAGG-3' Reverse (full length) 5'-CTAGATGACATTGTCAAAGAGCTGCATG -3' Forward (internal) 5'-GCCACTTTTCTCTTCAATGCCTT-3' Reverse (internal) 5'-AAACCAGCCACAAAGTCCTCAA-3' TaqMan probe 5'-ACACCAACCACGATGGCTCTGTCAGTT-3' Kv4.2 NM031730 Forward (internal) 5'-AAGTTCACCAGCATCCCTGCA-3' Reverse (internal) 5'-CCCTGCTATGGTTTTTGGTACCA-3' TaqMan probe 5'-ACACCATCGTCACCATGACAACACTGG-3 Forward (standard) 5'-TCGCTACGGTTATGTTCTACG-3' Reverse (standard) 5'-TCTTCTGTGCCCTTCGTTTGT-3' Kv4.3 U42975 Forward (internal) 5'-GAATCCGTGTGGCCAAAACAG-3' Reverse (internal) 5'-TTCATTGAGGAGTCCATTGCG-3' TaqMan probe 5'-AGCTCCAATGCCTACCTGCACAGCAA-3' Forward (standard) 5'-CAGAACCAGAGAGCAGATAAA-3' Reverse (standard) 5'-CTGCTCATCAATAAACTCGTG-3' Beta-actin NM031144 Forward (internal) 5'-AGAGCAAGAGAGGCATCCTGAC-3' Reverse (internal) 5'-TCTCCATATCGTCCCAGTTGGT-3' TaqMan probe 5'-TGAAGTACCCCATTGAACACGGCATTGT-3' Forward (standard) 5'-GCTCGTCGTCGACAACGGCTC-3' Reverse (standard) 5'-CAAACATGATCTGGGTCATCTTCTC-3' NIH indicates National Institutes of Health.

previously [23]. The control adenoviral vector, i.e. Ad-hrGFP was generated in the same way as Ad-KChIP2-hrGFP. 2.4. Real-time RT-PCR To evaluate developmental change of Kv4.2, Kv4.3 and KChIP2 mRNA, real-time RT-PCR (TaqMan) was done using an ABI PRISM 7700 (Applied Biosystems, Foster-City, CA, USA) [24,25]. Total RNA (40 ng) of E12, E18, P1 and P10 rat cardiac ventricle was RT into cDNA and amplified using a TaqMan® One-Step RT-PCR Master Mix Reagents (PE Biosystems). An optimal PCR curve could be observed within 40 cycles with gene-specific primer (internal) pairs and TaqMan probe (see Table 1). The TaqMan probe was

1075

labeled with fluorescent FAM (reporter) at the 5'-end and fluorescent TAMRA (quencher) at the 3'-end. The absolute copy numbers of beta-actin, Kv4.2, Kv4.3 and KChIP2 transcripts in samples were calculated by using a fluorescence curve obtained from amplicons of serially diluted standard RNA, which had been synthesized in in vitro transcription. For in vitro transcription, beta-actin, Kv4.2 and Kv4.3 cDNA were amplified by PCR with gene-specific primer (standard) pairs (see Table 1), using the same methods as for KChIP2 cloning, and then subcloned into pGEM-T vector (Promega). After the sequences were verified, the plasmids of beta-actin, Kv4.2, Kv4.3 and KChIP2 were linearized and transcribed by T7 or SP6 RNA polymerase as described previously [25]. 2.5. Cell culture and transfection of adenoviral vectors In order to transfect recombinant adenovirus, the isolated E12 cardiac ventricular myocytes were cultured. Cells were counted and were subsequently maintained on collagencoated coverslip fragments in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum, 30 µg/ml streptomycin and 30 unit/ml penicillin. After 12 h, recombinant adenoviral infection was performed (5–20 PFU/cell), and 2 h later, medium with adenovirus was removed. The myocytes were cultured for 48 h and then coverslip fragments with attached cells were perfused with the external solutions and patch-clamp experiment was performed, using the same methods as for freshly isolated E12 and P10 myocytes. Transfection was identified by the expression of hrGFP and the beating condition signified the existence of myocytes. 2.6. Indirect immunofluorescence and confocal microscopy Labeling was performed on hrGFP- and KChIP2-infected myocytes with anti-Kv4.2 primary antibodies (Alomone Laboratories, Jerusalem, Israel; 0.3 mg/ml) [26]. Freshly isolated E12 myocytes were adhered to laminin-coated coverslips, and infected with Ad-KChIP2-hrGFP and Ad-hrGFP the next day. After 48 h of this adenoviral infection followed by fixation and permeabilization, cells were incubated overnight at 4 °C with anti-Kv4.2 primary antibodies. Incubation with secondary antibodies was performed at room temperature for 2 h using Tex Red-conjugated anti-rabbit IgG (Molecular Probes; 2 mg/ml). The dilution of primary antibody was 1:80 and that of secondary antibody was 1:100. To determine non-specific binding, staining control experiments with secondary antibody in the absence of the primary antibody was also performed. Samples were examined by confocal laser scanning microscopy using a Zeiss LSM 510 equipped with an ArKr laser and an HeNe laser. We used the “Multi Track” scanning mode to avoid cross talk of fluorescence. 2.7. Statistics All averaged data are presented as mean ± S.E.M. Statistical significance was determined using t-test to compare two

1076

T. Kobayashi et al. / Journal of Molecular and Cellular Cardiology 35 (2003) 1073–1082

Fig. 1. (A, B) The waveforms and current–voltage relationships and (C, D) recovery from inactivation of Ito in freshly isolated E12 and P10 myocytes. (A) Representative outward currents recorded from freshly isolated rat ventricular myocytes in (a) E12 and (b) P10. Cells were clamped at –87 mV, and whole-cell currents were evoked by 400-ms step depolarization to values ranging from –57 to +53 mV in steps of 10 mV every 20 s. Arrows represent the zero current level. (B) Average current–voltage relationships of Ito recorded in freshly isolated E12 (open squares, n = 9, Cm = 10.42 ± 0.96 pF) and P10 (closed circles, n = 8, Cm = 13.18 ± 1.60 pF) myocytes. (C) Typical waveforms of recovery from inactivation at –87 mV recorded in freshly isolated (a) E12 and (b) P10 myocytes. The rates of recovery of Ito from steady-state inactivation were determined by a double-pulse protocol. After inactivating the currents during 400-ms prepulse (P1) to +53 mV, cells were hyperpolarized to –87 mV for times ranging from 1 to 2000 ms before a second (P2) depolarization to +53 mV. Arrows represent the zero current level. (D) Plots showing mean time courses of recovery from inactivation at –87 mV. The magnitude of Ito elicited by P2 pulse was expressed as a fraction (relative current) of the Ito density at P1 pulse and plotted against the interpulse duration for freshly isolated E12 (open squares, n = 6) and freshly isolated P10 (closed circles, n = 6) myocytes. Data were fitted using either a single- or a two-exponential function as appropriate. For single-exponential fits: I/I0 = 1 – exp(–t/tau), where I and I0 are Ito at P2 pulse and at P1 pulse, respectively, t is the time spent at the recovery potential (–87 mV), tau is the time constant. For two-exponential fits: I/I0 = 1 – Afastexp(–t/taufast) – Aslowexp(–t/tauslow), Afast + Aslow = 1, where Afast and Aslow are the contributions of the fast and slow components for recovery, respectively; taufast and tauslow are the time constants for recovery of the fast and slow components, respectively.

groups, and ANOVA and Fisher’s multiple comparison tests to compare multiple groups. A value of P < 0.05 indicates statistical significance. In the experiment of real-time RTPCR, data of each mRNA amount were log transformed and then, because data of log10 (beta-actin mRNA copy number) followed a normal distribution, statistical analysis was carried out. 3. Results

day 10 (P10) myocytes. Fig. 1B shows the current–voltage relationships of Ito in freshly isolated E12 and P10 myocytes. Ito densities at +53 mV in the E12 myocytes were 3.83 ± 0.70 pA/pF, a value significantly (P < 0.01) lower than those in the P10 myocytes (11.85 ± 1.71 pA/pF). Thus, Ito densities increased about 3 times from E12 to P10. The results demonstrate that Ito exists in E12 myocytes and that its current density increases with development, as reported previously [8–11].

3.1. Electrophysiological recordings of Ito in freshly isolated E12 and P10 myocytes

3.2. Recovery from inactivation of Ito in freshly isolated E12 and P10 myocytes

Previous study reported that there was no Ito in cardiomyocytes of day 13 rat embryos [7]. To confirm this finding, we measured Ito in embryonic and postnatal rat ventricular myocytes. Surprisingly, Ito could be observed in E12, though its amplitude was very small (Fig. 1A-a). The amplitude of Ito increased during development, and then a relatively large amplitude of Ito was recorded in freshly isolated postnatal

Ito is composed of two components, Ito,f and Ito,s, and the rate of recovery from inactivation of Ito,s is significantly slower than Ito,f [3–5]. In order to study the recovery process from inactivation of Ito, two identical voltage pulses were applied with varying interpulse intervals. Fig. 1C shows typical voltage-clamp recordings in the recovery process from inactivation of Ito in freshly isolated E12 and P10

T. Kobayashi et al. / Journal of Molecular and Cellular Cardiology 35 (2003) 1073–1082

1077

Table 2 Gating parameters of steady-state inactivation and recovery from inactivation for each myocytes

E12 fresh P10 fresh Non-infected Ad-hrGFP Ad-KChIP2-hrGFP

Recovery from inactivation 2-exp fit Afast taufast (ms)

Aslow

tauslow (ms)

0.75 ± 0.11 (n = 6) 12.3 ± 3.5

0.25 ± 0.11

1014 ± 216

0.96 ± 0.01 (n = 8) 7.2 ± 0.9

0.04 ± 0.01

1356 ± 277

Steady-state inactivation k 1-exp fit V0.5 (mV) tau (ms) 1238 ± 279 (n = 6) –51.0 ± 4.5 (n = 9) 9.59 ± 0.56 9.07 ± 1.03 –41.5 ± 2.8 (n = 5)† 1025 ± 157 (n = 5) –56.8 ± 4.2 (n = 6) 8.03 ± 1.72 1392 ± 147 (n = 8) –53.2 ± 2.6 (n = 7) 9.59 ± 1.02 –37.8 ± 1.1 (n = 8) * ,† 5.38 ± 0.74 *

* Significantly different (P < 0.05) vs. E12 fresh. † Significantly different (P < 0.05) vs. non-infected.

myocytes. Recovery from inactivation of Ito in E12 myocytes was very slow, while the recovery was very rapid in P10 myocytes. The time course of recovery from inactivation in E12 myocytes was best fitted by a single-exponential function (Fig. 1D and Table 2). In P10 myocytes, however, recovery from inactivation of Ito was best described by sum of two exponentials. The fast recovering component of Ito accounted for the greater part of Ito in freshly isolated P10 (0.75 ± 0.11), and was reminiscent of Ito,f [3,12]. The slow time constant in P10 myocytes (1014 ± 216 ms) was not significantly different compared with the time constants in E12 myocytes (1238 ± 279 ms). This slow recovering component observed in the present study was reminiscent of Ito,s [3,12]. The current density of Ito,s in P10 can be estimated about 2.96 pA/pF at 53 mV from the contribution of Ito,s in the whole Ito (0.25) (Table 2). This value is similar to that of E12 (3.83 pA/pF) which had no Ito,f, suggesting that the increased component from E12 to P10 was only Ito,f. These results indicate that Ito in E12 is solely composed of Ito,s and the developmental increase of Ito through E12 to P10 is due to the dramatic appearance and the increase of Ito,f amplitudes. 3.3. Expression of Kv4.2, Kv4.3 and KChIP2 In order to determine the molecular mechanism that underlies the developmental increase in Ito,f, we examined expression of molecular correlates of Ito,f (Kv4.2 and Kv4.3) and KChIP2 during E12 and P10 by real-time RT-PCR (Fig. 2). As the expression of beta-actin did not change significantly through E12 to P10, the data of Kv4.2, Kv4.3 and KChIP2 were not standardized by that of beta-actin. Amount of Kv4.2 mRNA in E12 and P10 is 2.25 × 104 ± 0.25 × 104 and 1.73 × 104 ± 0.14 × 104 copies, respectively (mean ± S.E.M., per 40 ng total RNA). In the same periods, amount of Kv4.3 mRNA is 8.78 × 103 ± 2.32 × 103 and 2.92 × 104 ± 0.71 × 104 copies, respectively. Absolute mRNA copy numbers of Kv4.2 and Kv4.3 were changed by 0.76and 3.32-fold from E12 to P10, respectively. On the other hand, amount of KChIP2 mRNA in the same periods is 1.12 × 103 ± 0.25 × 103 and 8.15 × 105 ± 1.58 × 105, respectively. The amount of KChIP2 mRNA increased dramatically by 731-fold during the same period. These findings indicate the possibility that the huge increased expression of KChIP2 accounts for the dramatic developmental increase in Ito,f in the heart.

3.4. Electrophysiological recordings of Ito in KChIP2-infected E12 ventricular myocytes The real-time PCR revealed that the expressed amount of Kv4.2/4.3 mRNA in E12 myocytes was comparable with that in P10 myocytes. If the developmental increase of KChIP2 expression is responsible for the appearance of Ito,f, only an over-expression of KChIP2 should be enough to produce great Ito,f in E12 myocytes. To confirm our assumption, E12 myocytes were infected with recombinant adenoviral vectors containing either hrGFP only (Ad-hrGFP) or KChIP2 and hrGFP (Ad-KChIP2-hrGFP). Fig. 3A illustrates representative records of Ito in non-infected (uninfected control), AdhrGFP-infected and Ad-KChIP2-hrGFP-infected myocytes. Only a small amplitude of Ito was observed in non-infected and Ad-hrGFP-infected myocytes. On the other hand, an extremely large amplitude of Ito was observed in AdKChIP2-hrGFP-infected myocytes. Fig. 3B shows the current–voltage relationships of Ito in each myocyte. The averaged current densities at +53 mV in non-infected and AdhrGFP-infected myocytes were 7.3 ± 1.3 and 6.8 ± 1.3 pA/pF, respectively. Ad-KChIP2-hrGFP-infected myocytes had a greater amount of Ito (83.6 ± 10.2 pA/pF; P < 0.01 vs. non-infected and Ad-hrGFP-infected myocytes). These data

Fig. 2. Relationships between developmental periods (E12, E18, P1 and P10) and the abundance of mRNA expression in beta-actin (closed circles, n = 6), Kv4.2 (closed triangles, n = 6), Kv4.3 (closed diamonds, n = 6) and KChIP2 (open squares, n = 6). Each mRNA copy number (mRNA expression amount) in 40 ng total RNA of ventricular myocytes was measured with real-time RT-PCR. mRNA expression amount was log transformed; log10 (mRNA copy number).

1078

T. Kobayashi et al. / Journal of Molecular and Cellular Cardiology 35 (2003) 1073–1082

Fig. 3. The waveforms and current–voltage relationships of Ito in noninfected, Ad-hrGFP-infected and Ad-KChIP2-hrGFP-infected myocytes. (A) Representative outward currents recorded from (a) non-infected, (b) Ad-hrGFP-infected and (c) Ad-KChIP2-hrGFP-infected myocytes. Cells were clamped at –87 mV, and whole-cell currents were evoked by 400-ms step depolarization to values ranging from –57 to +53 mV in steps of 10 mV every 20 s. Arrows represent the zero current level. (B) Average current– voltage relationships of Ito recorded from non-infected (open circles, n = 9, Cm = 30.62 ± 5.80 pF), Ad-hrGFP-infected (open triangles, n = 6, Cm = 30.58 ± 6.02 pF) and Ad-KChIP2-hrGFP-infected (closed squares, n = 16, Cm = 29.23 ± 4.30 pF) myocytes.

suggest that the over-expression of KChIP2 is enough to get large increased Ito in E12 myocytes. 3.5. Recovery from inactivation and steady-state inactivation of expressed Ito In order to characterize the large increased Ito in AdKChIP2-hrGFP-infected myocytes, two identical voltage pulses were applied with varying interpulse intervals. Fig. 4A shows typical voltage-clamp recordings in the recovery process from inactivation of Ito in non-infected, Ad-KChIP2hrGFP-infected and freshly isolated P10 myocytes. Recovery from inactivation of Ito in non-infected and Ad-hrGFPinfected myocytes were very slow, while the recovery was very rapid in Ad-KChIP2-hrGFP-infected myocytes. The time courses of recovery from inactivation in non-infected and Ad-hrGFP infected were best fitted by a singleexponential function as freshly isolated E12 myocytes

(Fig. 1C,D), and the recovery time constants were not significantly different (Table 2). On the other hand, in Ad-KChIP2hrGFP-infected myocytes, recovery from inactivation of Ito was best described by the sum of two exponentials as the case of freshly isolated P10 myocytes. The fast and slow time constants derived from Ad-KChIP2-hrGFP-infected and freshly isolated P10 myocytes were similar (7.2 ± 0.9 vs. 12.3 ± 3.5 ms and 1356 ± 277 vs. 1014 ± 216 ms). The slow time constants in Ad-KChIP2-hrGFP-infected and freshly isolated P10 myocytes (1356 ± 277 and 1014 ± 216 ms, respectively) were not significantly different from the time constants in non-infected, Ad-hrGFP-infected and freshly isolated E12 myocytes (1025 ± 157, 1392 ± 147 and 1238 ± 279 ms, respectively). The current density of Ito,s in AdKChIP2-hrGFP-infected myocytes can be estimated about 3.3 pA/pF from the contribution of Ito,s in the whole Ito (0.04 ± 0.01) (Table 2). This value is rather smaller than that of Ad-hrGFP-infected myocytes (6.8 pA/pF) which had no Ito,f. We do not know the reason why Ito,s component decreased. In any case, however, it can be mentioned that the overexpression of KChIP2 does not increase Ito,s component. These results indicate that the large increased Ito in AdKChIP2-hrGFP-infected myocytes mainly comes from the great increase of Ito,f, suggesting the over-expression of KChIP2 in E12 myocytes could induce great Ito,f and mimic Ito in P10 myocytes. The steady-state inactivation curve of Ito was determined by the two-step pulse protocol (see legend of Fig. 4C). Mean normalized currents are plotted as a function of conditioning potentials in Fig. 4C; the continuous and broken curves represent the best fits to the averaged data by the Boltzmann distribution. The voltages for half-inactivation (V0.5) in freshly isolated E12, non-infected and Ad-hrGFP-infected myocytes ranged between –51 and –57 mV (Table 2). In Ad-KChIP2-hrGFP-infected and freshly isolated P10 myocytes, V0.5 was shifted to a more depolarized potential compared with the other groups of myocytes (Fig. 4C and Table 2). The V0.5 in freshly isolated P10 myocytes was between that in Ad-KChIP2-hrGFP-infected myocytes and that in the other myocytes. This medium shift in the freshly isolated P10 myocytes may be explained by a relative contribution of Ito,f and Ito,s to the cells (see Section 4). 3.6. Immunocytochemistry In heterologous systems, the co-expression of KChIP2 with Kv4.2/4.3 channels was reported to modify the trafficking of Kv4.2/4.3 channels and to lead to an enhanced expression of Kv4.2/4.3 channel in surface membrane [15,16]. To determine whether a similar phenomenon was evoked in Ad-KChIP2-hrGFP-infected cardiomyocytes, immunofluorescent labeling was performed using specific antibodies against Kv4.2. Ad-hrGFP-and Ad-KChIP2-hrGFP-infected cells could be distinguished from non-infected cells by expression of GFP. In Ad-hrGFP-infected myocytes (in the absence of KChIP2 mRNA), Kv4.2 proteins were concentrated within the perinuclear endoplasmic reticulum

T. Kobayashi et al. / Journal of Molecular and Cellular Cardiology 35 (2003) 1073–1082

1079

Fig. 4. Recovery from inactivation of Ito in cultured myocytes and steady-state inactivation. (A) Typical waveforms of recovery from inactivation at –87 mV recorded in (a) non-infected, (b) Ad-KChIP2-hrGFP-infected and (c) freshly isolated P10 myocytes. The rates of recovery of Ito from steady-state inactivation were determined by a double-pulse protocol. After inactivating the currents during 400 ms prepulse (P1) to +53 mV, cells were hyperpolarized to –87 mV for times ranging from 1 to 2000 ms before a second (P2) depolarization to +53 mV. The insets show the currents recorded in the initial phase (from 1 to 50 ms) of recovery in Ad-KChIP2-hrGFP-infected and freshly isolated P10 myocytes. Arrows represent the zero current level. (B) Plots showing mean time courses of recovery from inactivation at –87 mV. The magnitude of Ito elicited by P2 pulse was expressed as a fraction (relative current) of the Ito density at P1 pulse and plotted against the interpulse duration for freshly isolated E12 (open squares, n = 6), freshly isolated P10 (closed circles, n = 6), non-infected (open circles, n = 5), Ad-hrGFP-infected (open triangles, n = 7) and Ad-KChIP2-hrGFP-infected (closed squares, n = 8) myocytes. Data were fitted using either a single- or a two-exponential function as appropriate. The inset shows the initial phase of recovery of the currents on an expanded timescale. (C) Steady-state inactivation of Ito in freshly isolated E12 (open squares, n = 9), freshly isolated P10 (closed circles, n = 5), non-infected (open circles, n = 6), Ad-hrGFP-infected (open triangles, n = 7) and Ad-KChIP2-hrGFP-infected (closed squares, n = 8) myocytes were determined by two-step pulse protocol; a conditioning pulse of 2000-ms duration from –87 to +13 mV in steps of 10 mV was followed by a step to +53 mV for 400 ms. The holding potential was –87 mV, and the interval between test pulses was 20 s. The magnitude of Ito detected at +53 mV after the conditioning pulse was normalized to Ito recorded at a conditioning potential of –87 mV in each individual experiment, and was given as a fraction of the conditioning pulse potential. Data were fitted assuming Boltzmann kinetics of steady-state inactivation. The data points were fitted with the Boltzmann-distribution equation: It/Imax = 1/{1 + exp[(V – V0.5)/k]}, where It is the peak current at a conditioning potential of V, Imax is the maximal peak current (at –87 mV), V0.5 is the half-inactivation potential and k is a slope factor.

(Fig. 5B). In contrast, in Ad-hrGFP-KChIP2-infected myocytes, Kv4.2 proteins were observed at the outer margins of the cells (Fig. 5A). This result indicates that the mechanism, by which KChIP2 increased Ito,f, is KChIP2 protein-induced promotion of trafficking of Kv4.2/4.3 channels to the cell membrane.

4. Discussion Two kinetically different components of Ito were observed in rat neonatal ventricular myocytes in the present study. Previous studies [3,5,12] also demonstrated two kinds of Ito which had the different kinetics of recovery from inactivation. It is difficult to directly compare rates of recovery from inactivation between previous studies and ours, because the temperature under which the experiments were performed was different (22–24 °C in previous studies [3,5,12] and

Fig. 5. (A) Immunocytochemistry of Kv4.2 in Ad-KChIP2-hrGFP-infected or (B) Ad-hrGFP-infected myocytes. Green fluorescence of GFP indicates adenoviral vector infection. Red indicates Kv4.2 protein with anti-Kv4.2 antibody (scale bar = 20 µm).

1080

T. Kobayashi et al. / Journal of Molecular and Cellular Cardiology 35 (2003) 1073–1082

36–37 °C in our study). However, both studies similarly showed that the difference of fast and slow rates of recovery from inactivation was in the order of 100. Therefore, we may observe the same components as previous studies did, i.e. Ito,f and Ito,s. Previous studies in rat [5,9] and mouse myocytes [4], discriminated Ito,f and Ito,s by using inactivation time course of the currents. It was thought that the decaying current should be fit with two-exponential function if there are two component of Ito. In heterologous expression experiments of Kv1.4 (which is thought to be molecular correlate of Ito,s) and Kv4.2 (which is thought to be molecular correlate of Ito,f), however, inactivating time constant in Kv1.4 was as fast as that in Kv4.2 [3,27,28]. These studies led us to speculate that it is difficult to discriminate Ito,f and Ito,s by using inactivation time course of the currents. This is one reason we have not examined the inactivating kinetics. In addition, we observed a slowly activating outward current (probably IK) in a part of examined cardiac myocytes. Activating time constant of this current was similar to the time constant of slow inactivating component of Ito in adult rat myocytes. It was difficult for us to discriminate three currents (Ito,f, Ito,s and probably IK) with using fit-exponential function. Therefore, the appropriate analysis was discrimination of Ito,f and Ito,s by recovery process from inactivation, as reported previously [3,5,12]. To our surprise, very low but detectable quantity of Ito was observed in freshly isolated E12 myocytes. We firstly expected that there was no Ito in E12 myocytes because Abrahamsson et al. [7] reported that Ito in E13 rat myocytes was never observed. The discrepancy may be ascribed to the different experimental conditions. In our experiment, the holding potential was –87 mV and step pulses were applied every 20 s. The present study revealed that the V0.5 of steadystate inactivation of Ito in freshly isolated E12 myocytes was –51 mV and the recovery time constant from inactivation in E12 was very slow (1.2 s). In Abrahamsson’s study, the holding potential was more positive (–50 mV) and the interpulse interval was shorter (5 s). A more hyperpolarized holding potential and a longer interpulse interval may be necessary to detect Ito in embryonic myocytes because Ito in embryonic myocytes was solely composed of Ito,s, not Ito,f. The present study showed the trend of depolarizing shift of V0.5 of steady-state inactivation in Ad-KChIP2-hrGFPinfected and freshly isolated P10 myocytes (Ito,f dominant), compared with the other groups (Ito,s only). Previous study [12] reported that neonatal and adult cardiomyocytes have almost same V0.5, although there was the difference of the rate of recovery from inactivation in these groups. Heterologous expression experiment showed that V0.5 in Kv4.2-only expressed oocytes was shifted to more hyperpolarized potential (the leftward shift) compared with that in Kv1.4-only expressed oocytes [27]. On the other hand, V0.5 in mammalian cell line co-expressing Kv4.2 and KChIP2 was shifted to more depolarized potential (the rightward shift) compared with that in cells expressing Kv4.2 alone [16]. Unfortunately, none of heterologous expression experiment reported the

difference of V0.5 in Kv1.4-only expressed and Kv4.2/4.3 and KChIP2 co-expressed cells/oocytes under identical experimental condition. Therefore, we speculate that (1) V0.5 of Ito,f is shifted to slightly depolarized potential compared with that of Ito,s and (2) the difference of V0.5 between Ito,f and Ito,s is small. These may be the reasons that there is the slight difference (but not statistically significant) of V0.5 between freshly isolated E12 and P10 myocytes. The appearance of Ito,f during early postnatal period could be explained by the change of the expression of Kv4.2/4.3, the molecular correlate of Ito,f. Previous studies, however, reported that Kv4.2 mRNA did not increase during early postnatal period. Roberds and Tamkun [29] reported that there was no change in Kv4.2 (RK5) mRNA through E14 to P10. They also reported that the abundance of Kv4.2 mRNA suddenly increased at P20. Xu et al. [11] also showed that there was no change of Kv4.2 mRNA through P0 to P10 although Ito current density increased. Consistent with these previous studies, the present study also showed the mismatch between the changes in amplitude of Ito,f and expression of Kv4.2/4.3 mRNA during E12 and P10. The increase of Ito,f component is tremendous between two stages because E12 myocytes have no Ito,f as shown in our analysis. On the other hand, absolute mRNA copy numbers of Kv4.2 and Kv4.3 between two stages were changed by 0.76- and 3.32-fold, respectively. These results led us to speculate that another factor in addition to Kv4.2/4.3 is necessary to elucidate the appearance of Ito,f with development. Recent studies reported that co-expression of KChIP2 with Kv4.2/4.3 increased Ito,f in heterologous systems [15,16]. Kuo et al. [18] reported that Ito was largely reduced in cardiomyocytes from KChIP2deficient mice. They, in addition, reported that the expression of KChIP2 in heart of wild type was not detectable at E15, but became detectable at E17. The increase in KChIP2 mRNA level is consistent with our findings. In the present study, the abundance of KChIP2 mRNA increased from E12 to P10 myocytes by 731-fold. We think the dramatic increase of Ito,f corresponds with the huge increase in KChIP2 mRNA from E12 to P10, rather than Kv4.2/4.3. It is still possible that evaluation of protein level of KChIP2 may be better even though there is good correlation between expression of KChIP2 mRNA and amplitude of Ito,f. Generally, mRNA level does not always go with protein level. Unfortunately, it is difficult for us to perform western blot analysis because E12 rat embryos have little amount of heart tissue. This is one of the reasons we have taken realtime RT-PCR method instead of northern blot analysis in this study. To confirm that KChIP2 is a major player to increase Ito,f during development, we used gene transfer technique with adenoviral vectors. E12 myocytes expressed almost equivalent Kv4.2/4.3 mRNA level as P10 myocytes. If the developmental change of KChIP2 is really responsible for the increase in Ito,f density, the forced expression of KChIP2 in E12 myocytes must induce Ito,f. Indeed, a much greater amplitude of Ito,f was observed when E12 myocytes were transfected

T. Kobayashi et al. / Journal of Molecular and Cellular Cardiology 35 (2003) 1073–1082

with Ad-KChIP2-hrGFP. Moreover, immunocytochemical experiments demonstrated that KChIP2 enhanced the trafficking of Kv4.2 channels to cell surface. These results suggest that enough amount of Kv4.2/4.3 proteins have already existed in E12 myocytes and that only KChIP2 expression is necessary to express Ito,f channels to the cell membrane. A functional role of KChIP2 is still controversial [17,30]. Rosati et al. [17] reported that the amount of KChIP2 mRNA level could explain the transmural gradients of Ito expression in mammalian ventricular wall. On the other hand, Deschenes et al. [30] reported there was no gradient in KChIP2 protein expression across canine and human ventricular walls, casting doubt on the physiological significance of KChIP2. In the present study, real-time PCR showed a large increase in KChIP2 mRNA expression during development. Over-expressing KChIP2 via adenoviral gene transfer to E12 myocytes, which had no Ito,f but almost the same amount of Kv4.2 mRNA as P10 myocytes, resulted in the extreme increase in Ito,f density. Therefore, our data argue for a functional role of KChIP2 at least in the increase of Ito,f during development. We cannot correctly control the expression level of KChIP2 via adenoviral vector. The level of the overexpressed KChIP2 may be higher than the changes of KChIP2 during development because the amount of Ito,f in Ad-KChIP2-hrGFP-infected myocytes was greater than in freshly isolated P10 myocytes. Another limitation is unknown factor caused by culturing. Ad-KChIP2-infected myocytes were cultured during 2–3 d before KChIP2 protein had expressed. There is a possibility that unknown factor might affect characteristics of Ito,f during culturing.

Acknowledgements This study was in part supported by a Grant-in-Aid from the Ministry of Education, Science and Culture of Japan to N.T. (No. 14370015).

References [1] [2]

[3]

[4]

[5]

[6]

Sanguinetti MC. When the KChIPs are down. Nat Med 2002;8:18–9. Greenstein JL, Wu R, Po S, Tomaselli GF, Winslow RL. Role of the calcium-independent transient outward current I(to1) in shaping action potential morphology and duration. Circ Res 2000;87:1026– 33. Wickenden AD, Jegla TJ, Kaprielian R, Backx PH. Regional contributions of Kv1.4, Kv4.2, and Kv4.3 to transient outward K+ current in rat ventricle. Am J Physiol 1999;276:H1599–607. Xu H, Guo W, Nerbonne JM. Four kinetically distinct depolarizationactivated K+ currents in adult mouse ventricular myocytes. J Gen Physiol 1999;113:661–78. Kassiri Z, Hajjar R, Backx PH. Molecular components of transient outward potassium current in cultured neonatal rat ventricular myocytes. J Mol Med 2002;80:351–8. Nerbonne JM. Molecular basis of functional voltage-gated K+ channel diversity in the mammalian myocardium. J Physiol 2000;525(Pt 2):285–98.

1081

[7]

Abrahamsson C, Palmer M, Ljung B, Duker G, Baarnhielm C, Carlsson L, et al. Induction of rhythm abnormalities in the fetal rat heart. A tentative mechanism for the embryotoxic effect of the class III antiarrhythmic agent almokalant. Cardiovasc Res 1994;28:337–44.

[8]

Kilborn MJ, Fedida D. A study of the developmental changes in outward currents of rat ventricular myocytes. J Physiol 1990;430:37– 60.

[9]

Guo W, Kamiya K, Toyama J. Modulated expression of transient outward current in cultured neonatal rat ventricular myocytes: comparison with development in situ. Cardiovasc Res 1996;32:524–33.

[10] Wahler GM, Dollinger SJ, Smith JM, Flemal KL. Time course of postnatal changes in rat heart action potential and in transient outward current is different. Am J Physiol 1994;267:H1157–66. [11] Xu H, Dixon JE, Barry DM, Trimmer JS, Merlie JP, McKinnon D, et al. Developmental analysis reveals mismatches in the expression of K+ channel alpha subunits and voltage-gated K+ channel currents in rat ventricular myocytes. J Gen Physiol 1996;108:405–19. [12] Wickenden AD, Kaprielian R, Parker TG, Jones OT, Backx PH. Effects of development and thyroid hormone on K+ currents and K+ channel gene expression in rat ventricle. J Physiol 1997;504(Pt 2):271–86. [13] Kassiri Z, Zobel C, Nguyen TT, Molkentin JD, Backx PH. Reduction of I(to) causes hypertrophy in neonatal rat ventricular myocytes. Circ Res 2002;90:578–85. [14] Fiset C, Clark RB, Shimoni Y, Giles WR. Shal-type channels contribute to the Ca2+-independent transient outward K+ current in rat ventricle. J Physiol 1997;500(Pt 1):51–64. [15] An WF, Bowlby MR, Betty M, Cao J, Ling HP, Mendoza G, et al. Modulation of A-type potassium channels by a family of calcium sensors. Nature 2000;403:553–6. [16] Bahring R, Dannenberg J, Peters HC, Leicher T, Pongs O, Isbrandt D. Conserved Kv4 N-terminal domain critical for effects of Kv channelinteracting protein 2.2 on channel expression and gating. J Biol Chem 2001;276:23888–94. [17] Rosati B, Pan Z, Lypen S, Wang HS, Cohen I, Dixon JE, et al. Regulation of KChIP2 potassium channel beta subunit gene expression underlies the gradient of transient outward current in canine and human ventricle. J Physiol 2001;533:119–25. [18] Kuo HC, Cheng CF, Clark RB, Lin JJ, Lin JL, Hoshijima M, et al. A defect in the Kv channel-interacting protein 2 (KChIP2) gene leads to a complete loss of I(to) and confers susceptibility to ventricular tachycardia. Cell 2001;107:801–13. [19] Nagashima M, Tohse N, Kimura K, Yamada Y, Fujii N, Yabu H. Alternation of inwardly rectifying background K+ channel during development of rat fetal cardiomyocytes. J Mol Cell Cardiol 2001;33: 533–43. [20] Tohse N, Nakaya H, Hattori Y, Endou M, Kanno M. Inhibitory effect mediated by alpha 1-adrenoceptors on transient outward current in isolated rat ventricular cells. Pflügers Arch 1990;415:575–81. [21] Yamada Y, Nagashima M, Tsutsuura M, Kobayashi T, Seki S, Makita N, et al. Cloning of a functional splice variant of L-type calcium channel beta 2 subunit from rat heart. J Biol Chem 2001;276: 47163–70. [22] Shinoura N, Muramatsu Y, Nishimura M, Yoshida Y, Saito A, Yokoyama T, et al. Adenovirus-mediated transfer of p33ING1 with p53 drastically augments apoptosis in gliomas. Cancer Res 1999;59: 5521–8. [23] Miyake S, Makimura M, Kanegae Y, Harada S, Sato Y, Takamori K, et al. Efficient generation of recombinant adenoviruses using adenovirus DNA-terminal protein complex and a cosmid bearing the full-length virus genome. Proc Natl Acad Sci USA 1996;93:1320–4.

1082

T. Kobayashi et al. / Journal of Molecular and Cellular Cardiology 35 (2003) 1073–1082

[24] Eads CA, Danenberg KD, Kawakami K, Saltz LB, Danenberg PV, Laird PW. CpG island hypermethylation in human colorectal tumors is not associated with DNA methyltransferase overexpression. Cancer Res 1999;59:2302–6. [25] Fujimaki S, Funato T, Harigae H, Imaizumi M, Suzuki H, Kaneko Y, et al. A quantitative reverse transcriptase polymerase chain reaction method for the detection of leukaemic cells with t(8;21) in peripheral blood. Eur J Haematol 2000;64:252–8. [26] Sedarat F, Xu L, Moore ED, Tibbits GF. Colocalization of dihydropyridine and ryanodine receptors in neonate rabbit heart using confocal microscopy. Am J Physiol Heart Circ Physiol 2000;279:H202–9.

[27] Wickenden AD, Tsushima RG, Losito VA, Kaprielian R, Backx PH. Effect of Cd2+ on Kv4.2 and Kv1.4 expressed in Xenopus oocytes and on the transient outward currents in rat and rabbit ventricular myocytes. Cell Physiol Biochem 1999;9:11–28. [28] Johns DC, Nuss HB, Marban E. Suppression of neuronal and cardiac transient outward currents by viral gene transfer of dominant-negative Kv4.2 constructs. J Biol Chem 1997;272:31598–603. [29] Roberds SL, Tamkun MM. Developmental expression of cloned cardiac potassium channels. FEBS Lett 1991;284:152–4. [30] Deschenes I, DiSilvestre D, Juang GJ, Wu RC, An WF, Tomaselli GF. Regulation of Kv4.3 current by KChIP2 splice variants: a component of native cardiac I(to)? Circulation 2002;106:423–9.