C. Marionneau and B. Couette contributed equally to this work.
Specific pattern of ionic channel gene expression associated with pacemaker activity in the mouse heart
Article first published online: 22 DEC 2004
The Journal of Physiology
Volume 562, Issue 1, pages 223–234, 1 January 2005
How to Cite
Marionneau, C., Couette, B., Liu, J., Li, H., Mangoni, M. E., Nargeot, J., Lei, M., Escande, D. and Demolombe, S. (2005), Specific pattern of ionic channel gene expression associated with pacemaker activity in the mouse heart. The Journal of Physiology, 562: 223–234. doi: 10.1113/jphysiol.2004.074047
- Issue published online: 22 DEC 2004
- Article first published online: 22 DEC 2004
- (Received 17 August 2004; accepted after revision 21 October 2004; first published online 21 October 2004)
Even though sequencing of the mammalian genome has led to the discovery of a large number of ionic channel genes, identification of the molecular determinants of cellular electrical properties in different regions of the heart has been rarely obtained. We developed a high-throughput approach capable of simultaneously assessing the expression pattern of ionic channel repertoires from different regions of the mouse heart. By using large-scale real-time RT-PCR, we have profiled 71 channels and related genes in the sinoatrial node (SAN), atrioventricular node (AVN), the atria (A) and ventricles (V). Hearts from 30 adult male C57BL/6 mice were microdissected and RNA was isolated from six pools of five mice each. TaqMan data were analysed using the threshold cycle (Ct) relative quantification method. Cross-contamination of each region was checked with expression of the atrial and ventricular myosin light chains. Two-way hierarchical clustering analysis of the 71 genes successfully classified the six pools from the four distinct regions. In comparison with the A, the SAN and AVN were characterized by higher expression of Navβ1, Navβ3, Cav1.3, Cav3.1 and Cavα2δ2, and lower expression of Kv4.2, Cx40, Cx43 and Kir3.1. In addition, the SAN was characterized by higher expression of HCN1 and HCN4, and lower expression of RYR2, Kir6.2, Cavβ2 and Cavγ4. The AVN was characterized by higher expression of Nav1.1, Nav1.7, Kv1.6, Kvβ1, MinK and Cavγ7. Other gene expression profiles discriminate between the ventricular and the atrial myocardium. The present study provides the first genome-scale regional ionic channel expression profile in the mouse heart.
In the adult mammalian heart, the heartbeat is initiated by primary pacemaker cells located in the sinoatrial node (SAN). From the SAN, the pacemaker impulse is spread to the atrium and the atrioventricular node (AVN), and then through the conduction system to the whole heart to cause coordinated contractions. Furthermore, the AVN also acts as the major subsidiary pacemaker, securing automaticity and taking the lead in case of failure of the SAN. Coordination between cardiac pacemaking and contraction requires regionally specialized electrical functions, which are reflected by the electrical heterogeneity between the heart chambers. It has been recognized that such heterogeneity is achieved by the differential expression of ionic channel genes and connexins (Cx) (for recent review see Kleber & Rudy, 2004). Particularly, the development of the patch-clamp methods in the 1980s and molecular biology in the 1990s has provided an enormous amount of information about how specialization is achieved by differential expression of cardiac ion channels (Boyett et al. 1996). Most of these studies, however, have focused on individual ionic channels, currents, and/or on the expression of channel subunits generating these currents (Schram et al. 2002). Meanwhile, in recent years, dramatic progress in our knowledge of mammalian genomes (e.g. mouse and man) has led to the identification of the complete ionic channel gene repertoire (International Human Genome Sequencing Consortium, Lander et al. 2001; FANTOM Consortium, Okazaki et al. 2002). How this repertoire is expressed in specific regions of the heart and how it functions in relation to the tissue's physiological (e.g. automaticity) and pathophysiological states (e.g. sick sinus syndrome) are important issues to be addressed.
High-throughput methods now allow the expression of thousands of genes to be measured simultaneously (Carulli et al. 1998; Pollack et al. 1999). We have previously established a specialized cDNA chip comprising probes for the collection of mouse and man ion channel and calcium regulator genes (IONchips; Le Bouter et al. 2003, 2004). However, the cDNA chip technology is currently limited by the amount of RNA it requires, thereby preventing the profiling of small-size structures. In the present study, we made use of high-density real-time RT-PCR array to differentiate the expression pattern of transcripts of ionic channels and proteins involved in Ca2+ homeostasis (Ca2+ regulators) in the mouse SAN and AVN, and compared this pattern with that of the atrium and ventricle. This provides the first genome-scale profile of the regional expression of ionic channels of the mouse heart. We found that the SAN and the AVN (nodes) closely cluster in the same branch, whereas the working atrial and ventricular myocardium cluster separately. We also found that the nodes distinguished themselves from the working myocardium through increased expression of a large panel of ion channel genes rather than through down-expression. Part of our data has previously been published in abstract form (Demolombe et al. 2004).
Animal experiments were performed in accordance with institutional guidelines for animal use in research. Hearts were excised from 10-week-old C57BL/6 male mice (Charles River) which had been killed by cervical dislocation. The SAN, AVN, atria (A) and ventricles (V) were carefully dissected as previously described (Mangoni & Nargeot, 2001; Lei et al. 2004) and flash-frozen in liquid nitrogen for further RNA isolation. In brief, whole hearts were washed by Langendorff perfusion for 5 min using oxygenated Tyrode solution. After removal of the ventricles, the right atrium and ventricular septum were placed in a Tyrode-perfused dissection chamber. Under the microscope, the right atrium was opened to expose the crista terminalis, the intercaval area and the interatrial septum. The preparation was pinned with the endocardial side exposed up. Figure 1 shows the final preparation. SAN and AVN regions were recognized by their anatomic landmarks. A thin strip of SAN tissue (approximately 1 × 0.8 mm), limited by crista terminalis, atrial septum and orifices of the venae cavae was cut from the right atrium, and a triangle-shaped piece of AVN tissue was cut from Koch's triangle. Total RNA from six pools of five mice each was isolated and DNase treated using the RNeasy Fibrous Tissue Mini or Micro Kit (Qiagen) by following the manufacturer's instructions. The quality of total RNA was assessed by microelectrophoresis on acrylamide gel (Agilent 2100 Bioanalyser, Agilent, Palo Alto, CA, USA) and by RT-PCR using interexonic primers for the cardiac α-actin gene (see Fig. 2). Lack of genomic DNA contamination was verified by PCR.
TaqMan real-time RT-PCR
TaqMan low-density arrays (Micro Fluidic Cards, Applied Biosystems, Foster City, CA, USA) were used in a two-step RT-PCR process. First strand cDNA was synthesized from 2 µg of total RNA using the High-Capacity cDNA Archive Kit (Applied Biosystems). PCR reactions were then carried out in Micro Fluidic Cards using the ABI PRISM 7900HT Sequence Detection System (Applied Biosystems). The 384 wells of each card were preloaded with 96 × 4 predesigned fluorogenic TaqMan probes and primers. The probes were labelled with the fluorescent reporter dye 6-carboxyfluorescein (FAM, Applera Corp., Norwalk, CT, USA) on the 5′ end, and with nonfluorescent quencher on the 3′ end. The genes selected for their cardiac expression encode 71 α- and β-ion channel subunits, ten proteins involved in calcium homeostasis, one ion channel expression regulator, six specific markers of cardiac regions, one marker for vascular vessels, one for neuronal tissue, one for fibroblasts, two inflammatory markers, and three reference genes used for normalization. The genes are listed in the online supplementary material. Two nanograms of cDNA combined with 1× TaqMan Universal Master Mix (Applied Biosystems) were loaded into each well. The Micro Fluidic Cards were thermal cycled at 50°C for 2 min and 94.5°C for 10 min, followed by 40 cycles at 97°C for 30 s, and 59.7°C for 1 min. Data were collected with instrument spectral compensations by the Applied Biosystems SDS 2.1 software, and analysed using the threshold cycle (Ct) relative quantification method (Livak & Schmittgen, 2001). The hypoxanthine guanine phosphoribosyl transferase (HPRT) reference gene was used for normalizing the data. Genes with Ct > 32 were eliminated for lack of reproducibility. The outliers were excluded using a robust statistical modified z-score method based on the median of absolute deviation (Iglewicz & Hoaglin, 1993; Barnett & Lewis, 1984). The nonexcluded values ( n= 4–6 for SAN, AVN, atria and ventricles) were averaged and then used for the 2−ΔCt× 100 calculation; 2−ΔCt corresponds to the ratio of each gene expression versus HPRT. It should be kept in mind that expression quantification for each transcript is relative since high-throughput TaqMan does not provide absolute quantification. However, it has been established that for a PCR efficiency of 100%, a Ct value of 30 corresponds approximately to 1000 copies of the transcript, and that this relation is linear between Ct values from 10 to 36 (Michael De Graaf, Applied Biosystems; personal communication).
SYBER Green RT-PCR
Five distinct pools of tissue were used. Reverse transcription was performed with total RNA (70 ng) using random primers and Superscript II RNase H reverse transcriptase (Invitrogen). The expression of 22 α- and β-ion channel subunits, including the HCN channel family (TaqMan assay for HCN4 was not available), was quantified by real-time PCR (ABI Prism 7000, Applied Biosystems). Experiments were performed using 1× SYBR Green PCR Master Mix (Applied Biosystems) with 300 nm of each primer. Primers were designed with the Primer Express software (ABI). The cycling conditions included a hot start at 95°C for 10 min, followed by 40 cycles at 95°C for 15 s and 60°C for 1 min. All primer pairs were tested using mouse genomic DNA as the template. Those giving 90–100% efficacy were chosen. In all cases, a single amplicon of the appropriate melting temperature and size was detected using the dissociation curve and gel electrophoresis, respectively. Results were normalized to HPRT and expressed according to the 2−ΔCt method. Similar results were obtained with the beta-glucuronidase (Gus), which was used as a reference gene (data not shown). As negative controls, RNA samples incubated without reverse transcriptase during cDNA synthesis showed no amplification.
Two-way hierarchical agglomerative clustering was applied to the gene expression matrix consisting of the 24 pools and a selection of analysed genes. The input consisted of the ΔCt for each gene and pool. We applied average linkage clustering with uncentred correlation using the Cluster program (Eisen et al. 1998). Clusters were visualized using the Treeview program. For every gene, compartments were compared and statistical differences were identified using one-way analysis of variance completed by a Tukey t test. A value of P < 0.05 was considered significant.
Study design and evaluation of node tissue contamination
The study included six pools of SAN, six pools of AVN, six pools of A and six pools of V. Each pool represented five adult inbred C57BL/6 male mice. Genes were selected for their involvement in the mouse cardiac electrical signalling. Several genes that played a key role in neuronal electrical signalling were further included, together with well-characterized genes involved in calcium homeostasis. A total of 96 transcripts were quantified using the Ct relative quantification method (Livak & Schmittgen, 2001). The HPRT reference gene was used for data normalization. Transcripts were quantified using fluorescent probes with 100% PCR efficacy. This standardized method permits comparison of relative transcript expression level between cardiac compartments, and also between genes. Expression levels are expressed as a ratio versus HPRT (×100).
By making use of specific markers of atrial, ventricular and nodal tissues, we first aimed to control possible contamination of small-size nodal samples by atrial and ventricular tissue (Fig. 2). Indeed, both the SAN and AVN regions are heterogeneous structures containing both pacemaker cells and cells displaying atrial phenotype. Quantitative estimation of the relative presence of atrial cells in nodal samples is thus needed with respect to the limited number of true pacemaker cells in the mouse SAN and AVN. We used the atrial myosin light chain gene (MLC1A) as a marker to calculate the percentage of contamination by atrial myocytes. MLC1A expression was not detected in the ventricle. The calculation was based on the assumption that no MLC1A was expressed in nodal cells. An estimate of 48% as maximal contamination for the SAN and 43% for the AVN was thus obtained. The expression of the brain natriuretic peptide (BNP) gene confirmed the percentage of SAN contamination by atrial cells. Expression of the ventricular myosin light chain gene (MLC2V) was high in ventricular tissue and null in the atria. Using this marker, we estimated that <23% of total RNA from the AVN samples were contaminated by ventricular myocytes. To evaluate contamination further, probes for cardiac troponin I (TnIc) and slow skeletal troponin I (TnIs) genes were used as positive controls. Gorza et al. (1993) and Brahmajothi et al. (1996) showed that TnIc is present in every region of the mature heart both in working myocytes (A and V) and in the nodes. In contrast, expression of TnIs is restricted to the conduction system. Our nodal samples were enriched approximately seven times in TnIs in comparison with the working myocardium.
Hierarchical clustering of cardiac regions and ion channel genes
An overview of global gene expression data is provided by a two-way hierarchical clustering analysis of genes and cardiac regions (Fig. 3). This computer analysis revealed a clear separation of each cardiac region (SAN, AVN, A and V), indicating that the expression profile of transcripts involved in electrical signalling discriminated each individual sample pool (SAN, AVN, A and V). Hierarchical clustering also demonstrated subclassification within the cardiac regions: the SAN and AVN pools were parts of the same tree branch whereas atrial and ventricular pools clustered on a distinct branch. The genes that were most relevant for the classification are indicated in Fig. 3. Groups indicated as A and C in contain genes with a lower and higher expression in the nodal tissues, respectively. It should be noted that the number of genes up-regulated in the nodes (shown in group C) was greater than the number of down-regulated genes (shown in group A). In other words, the nodes were characterized by over-expression rather than through down-expression of ionic channel genes. Groups B and D contain genes with a higher expression in V and a higher expression in A, respectively. Thus, clustering demonstrates that the ionic channel expression pattern accurately discriminates specialized regions from the mouse heart, and distinguishes between automatic tissues from the working myocardium.
Specific regional repertoire of ionic channel transcripts
Among the 71 ionic channel genes, 14 exhibited very low or undetectable expression in the mouse heart. These included genes encoding Cav1.1, Cavγ6, CFTR, Kir1.1, Kir3.2, Kv3.1, Mirp1, TRAAK and TWIK2. Figures 4, 5 and 6 illustrate the relative levels of expression determined by TaqMan for the 67 ionic channel and Ca2+ regulator genes with measurable expression in at least one compartment. The nodes exhibited a specific expression pattern, which involved genes from every channel family.
Among the voltage-gated sodium channels (Fig. 4), the specific expression pattern of the SAN and AVN was mainly based on β-subunits such as Navβ1 (SCN1B) and Navβ3 (SCN3B). For example, expression of Navβ1 transcripts in the AVN was nearly twofold greater than that in the A. Note that Navβ1 transcripts were about ten times more expressed than Navβ3 transcripts.
As expected, the gap junction protein Cx40 (GJA5) largely predominated in the A as compared to the nodes, and remained low in the V, whereas expression of Cx43 (GJA1) predominated in the working myocardium. The expression of Cx37 (GJA4) and Cx45 (GJA7) was not associated with a specific region.
In the Ca2+ channel area, the voltage-dependent T-type Ca2+ channel alpha-1G subunit Cav3.1 (CACNA1G) was predominantly expressed in the nodes. The L-type Ca2+ channel alpha-1D subunit Cav1.3 (CACNA1D) was also preferentially expressed in the nodes, although in general its expression was lower than the alpha-1C subunit Cav1.2 (CACNA1C) (Fig. 4). The voltage-dependent Ca2+ channel alpha 2/delta auxiliary subunit 2, Cavα2δ2 (CACNA2D2), largely predominated in the nodes, whereas the expression of Cavα2δ1 (CACNA2D1) was more widely distributed. A profile comparable to that for Cav1.3 was obtained for the auxiliary subunit Cavβ3 (CACNB3), albeit at a very low level.
Among the voltage-gated K+ channel family, the brain-type α-subunits Kv1.1 (KCNA1) and Kv1.6 (KCNA6) expressed at low levels, predominated in the nodes (Fig. 5). Kv1.4 (KCNA4) expression was lower whereas Kv1.5 (KCNA5) expression was higher in the V in comparison with other regions. The voltage-gated K+ channel, Shal-related family member Kv4.2 (KCND2), which plays a key role in the mouse ventricular repolarization, also discriminated the nodes by its lower expression (Fig. 5). As expected, there were less transcripts of Kv4.3 (KCND3) than Kv4.2. The KvLQT1 regulator β-subunit MinK (KCNE1) was expressed at low levels in the AVN, and remained undetectable in the SAN and the A. KChiP2 (KCNIP2) and SUR2 (ABCC9) largely predominated in the ventricle, whereas Kvβ1 (KCNAB1) predominated in the AVN. Among the inward rectifiers, Kir2.2 (KCNJ12) predominated in the atria. The G-protein-activated Kir3.1 (KCNJ3) also predominated in the A compared to the nodes or V.
The expression of genes involved in Ca2+ homeostasis did not univocally discriminate tissue with intrinsic automaticity, with the exception of the ryanodine receptor RYR2, which was expressed at a lower level in the SAN than in the other cardiac regions (Fig. 6). In general, genes involved in Ca2+ homeostasis were 10–150 times more expressed than the α- and β-subunit ionic channels. Ankyrin B, an anchoring protein involved in the targeting of ion channels and transporters, showed reduced expression in the SAN (not illustrated).
Data obtained with quantitative TaqMan technique were further evaluated by the SYBR Green quantitative RT-PCR technique. The expression of 22 genes, selected for their differential expression between cardiac regions, was further determined (not illustrated). On purpose, the amplified cDNA sequence was chosen distinct from the sequence amplified by the TaqMan probe. In addition, we used distinct pools of tissue. The relative quantification obtained with TaqMan was similar to that determined with SYBR Green. On average, the difference between TaqMan and SYBR Green measurements did not exceed 2%. The expression of the hyperpolarization-activated, cyclic nucleotide-gated cation channel gene family HCN was investigated with the SYBR Green method (Fig. 7). Relative to the A and V, SAN tissues showed preferential expression of HCN1, and very robust expression of HCN4. However, HCN channel transcripts were not exclusively present in pacemaker cells, since ventricular myocytes expressed consistent levels of HCN2 and HCN4. HCN3 expression remained very low in all tested compartments.
Statistical analyses were further conducted to compare the expression profile of each ion channel gene between one compartment and the other compartments (the degree of significance is indicated as a, b or c in Figs 4–7). The AVN was targeted for higher expression of Nav1.1 (SCN1A), Nav1.7 (SCN9A), Cav3.1, the gamma subunit of voltage-dependent Ca2+ channel Cavγ7 (CACNG7), Kv1.6 and Kvβ1. The tetrodotoxin (TTX)-sensitive brain-type Na+ channel α-subunits Nav1.1 and Nav1.7 were expressed predominantly in the AVN.
The only transcripts that were predominantly expressed in the SAN were HCN1 and HCN4. Inversely, the SAN compartment was characterized by reduced expression of Cavβ2 (CACNB2), ATP-dependent K+ channel Kir6.2 (KCNJ11) and K+ intermediate/small-conductance Ca2+-activated channel SK1 (KCNN1). The SAN was also characterized by a reduced expression of the cardiac RYR2.
A large number of channel α- and β-subunits were predominantly expressed in the ventricle. These included Nav1.5 (SCN5A), Cav1.2, Cavα2δ1, Cavβ2, Kv1.5, Kir6.1 (KCNJ8), the voltage-gated K+ channel interacting protein KChIP2, the sulphonylurea receptor SUR2 and phospholamban (PLN). Ventricles were also distinguished by a lower expression of Kir3.1, Kir3.4 (KCNJ5), Cx40, and of the voltage-dependent Ca2+ channel auxiliary subunits Cavα2δ2 and Cavα2δ3 (CACNA2D3).
The atrial tissue compartment exhibited higher expression of Kir2.2, Kir3.1, Cx40 and Na+,K+-ATPase α1 (ATP1A1) associated with a lower expression of the K+ channel β-subunits MinK and Kvβ1. All these differences reached statistical significance.
The identification of the molecular determinants of cellular electrical activity in pacemaker tissue and the working myocardium constitutes an important issue towards understanding the regulation of the heartbeat in normal and physiopathological conditions. Here, we describe a large-scale expression profile of ionic channels in different regions of the heart, including the SAN, AVN, A and V. We demonstrated that the expression pattern of the genes involved in electrical signalling discriminated each cardiac region, and we highlighted that the nodal tissues closely clustered together with a profile distinct from that of the working myocardium.
The remarkable portrait of the nodal tissues
The L-type Cav1.2 and T-type Cav3.1 Ca2+ channels are highly expressed in the nodes, albeit not specifically (Fig. 4). The Cav1.3 transcript is more specific for the nodes. The Cav1.3 channel has been reported to play an important role in cardiac pacemaker activity, since its inactivation leads to sinus node dysfunction at rest and reduced cellular automaticity (Platzer et al. 2000; Zhang et al. 2002; Mangoni et al. 2003). Cavα2δ2 and Cav3.1 are the only transcripts which exhibit high expression in the nodal tissues and almost no expression in working myocytes. Based on these data, it is tempting to speculate that both Cavα2δ2 and Cav3.1 constitute specific markers of the nodal function. Consistently with the expression data, the low-voltage-activated T-type Ca2+ current is robustly expressed in both the mouse and rabbit SAN (Mangoni & Nargeot, 2001; Protas et al. 2001). In addition, the cardiac phenotype of Cav3.1 knockout mice (Kim et al. 2001) showed both bradycardia and slowing of the atrioventricular conduction (Mangoni et al. unpublished observations). Finally, it has been shown that Cavα2δ2 increases the T-type current amplitude related to Cav3.1 by 176% (Gao et al. 2000).
All four HCN isoforms have been found in mammalian heart (Stieber et al. 2004). Expression of these isoforms varies somewhat among species, cardiac tissue, and the developmental stage analysed. The dominant HCN isoform in the adult SAN of all species investigated so far (rabbit, mouse, dog) is HCN4 (Stieber et al. 2004), accounting for ∼80% of the total HCN message. Our expression profile of the HCN gene family (Fig. 7) is consistent with previous findings (Moosmang et al. 2001) indicating significant expression of the HCN1. The functional role of the HCN1 gene in cardiac pacemaking has not been elucidated yet. The functional importance of the HCN4 gene in cardiac automaticity has recently been confirmed by a knockout mouse model, which shows that this gene is essential for embryonic cardiac pacemaking (Stieber et al. 2003). We also detected expression of the HCN2 gene in the SAN and AVN. This result is consistent with previous findings showing a contribution of HCN2 channels in the proper setting of the heart rate (Ludwig et al. 2003). In conclusion, as for genes encoding Ca2+ channels, the expression profile obtained for the HCN gene family is consistent with the current literature on the functional role of these channels in cardiac pacemaking.
Among the voltage-gated Na+ channel α- and β-subunits, pacemaker tissues distinguish themselves by predominant expression of Navβ1 and Navβ3 (Fig. 4). Expression of these β-subunits in the mouse SAN was previously demonstrated at the protein level (Maier et al. 2003). Na+ channel β-subunits are multifunctional. Their effects include modulation of channel gating, facilitation of channel trafficking and regulation of cell adhesion (Isom, 2001). Functional expression of these subunits in vitro has shown that both Navβ1 (Isom et al. 1992) and Navβ3 (Morgan et al. 2000) accelerate Na+ current (INa) inactivation kinetics. Since brain Na+ channels, Nav1.1 and Nav1.7, exhibit slow inactivation kinetics, coexpression with Navβ1 and Navβ3 could increase the rate of inactivation of these channels, adapting their function to the fast heart rate of the mouse. The overall expression of Nav1.1 in our SAN samples was low, especially when compared to that of Nav1.5, which showed substantial expression both in the nodes and in the working myocardium. The expression of Nav1.1 mRNA in the SAN was first shown in tissue sections from newborn rabbits (Baruscotti et al. 1997). Expression of Nav1.5 in the mouse SAN is consistent with previous evidence indicating that Nav1.5 is expressed in large SAN cells, as well as in cells located in the intercaval region (Maier et al. 2003). By employing electrophysiology on single cells and immunolabelling of tissue sections, we have recently confirmed these findings and found that the expression of Nav1.5 is complex (Lei et al. 2004). Indeed, we have shown that although Nav1.5 was not expressed in small SAN cells, it is expressed in large SAN cells (likely to be from the periphery of the SAN) and in the region of the SAN located on the endocardial face of the crista terminalis (the periphery). Interestingly, the expression of the Nav1.5 protein is consistent with the observation that the density of the TTX-resistant INa component in SAN is positively correlated with the cell size. In contrast, the density of TTX-sensitive INa did not vary with cell size (Lei et al. 2004).
The neuronal Kv1.1 and Kv1.6 targeted nodal cells even if their expression level remained low in comparison with other K+ channels (Fig. 5). The mouse Kv1.1 clone generates a slowly inactivating current (Klumpp et al. 1991) poorly adapted to the physiology of pacemaker cells. However, coexpression of Kvβ1 subunits (expressed also in nodal cells and especially in AVN cells) alters Kv1.1 channel function from noninactivating delayed rectifiers to rapidly inactivating channels (Jing et al. 1999). Inversely, recombinant Kvβ3, but not Kvβ1, confers rapid inactivation to Kv1.6 subunits (Rhodes et al. 1997; Bahring et al. 2004). Electrophysiological investigation of mouse SAN and AVN cells is needed to elucidate the functional role of these K+ channels in pacemaking and conduction.
The significant expression of Kir2.1 in the mouse SAN is somewhat surprising (Fig. 5). Indeed, it is a classical concept that pacemaker cells express low inward rectifier K+ current (IKI) (Irisawa et al. 1993), which is coded by Kir2.1 and Kir2.2 subunits. Also, it has been proposed that the lack of IKI may constitute a fundamental requirement for pacemaking (Miake et al. 2002). Still, Kir2.1 expression in the mouse SAN was not as low as one would anticipate. However, our expression findings are fairly consistent with Noma group's data showing the presence of IKI in the mouse SAN (Cho et al. 2003), although we cannot exclude that part of Kir subunit expression in the SAN was due to contamination with atrial tissue and/or with fibroblasts.
The remarkable portrait of the working myocardium
Both atrial and ventricular cells exhibit a predominant expression of K+ channel Kv4.2 and Cx43 transcripts as compared to the nodal cells (Figs 4 and 5). It is well established that the Kv4.2 channel gene rates a current very similar to fast transient outward K+ current (Ito,f), a major repolarizing current in the mouse heart (Guo et al. 1999). In contrast to previous findings that demonstrated higher expression of Kv4.2 mRNA in ventricular than in atrial cells (Xu et al. 1999), we found a similar level of expression in both compartments. In adult mouse myocytes, one can assume that the lower density of Ito,f in atrial than in ventricular cells (Xu et al. 1999) is supported by preferential expression of the KChIP2 protein in the ventricular cells. The essential role of KChIP2 was demonstrated by genetic invalidation in mice, which completely eliminated the cardiac voltage-gated transient outward K+ current (Ito) current (Kuo et al. 2001), and also by the dramatic improvement of Kv4.2 cell surface expression in the presence of KChIP2 (Shibata et al. 2003).
Cx43 transcript expression predominates in the working myocardium (Fig. 4). The heterozygous Cx43+ mouse shows reduced ventricular conduction velocity (Thomas et al. 1998), but normal atrial conduction and SAN or AVN node function. This observation suggests that expression of Cx40 prevented the development of an abnormal phenotype in Cx43+ mice. However, the presence of Cx40 in the mouse nodal structure is controversial. In contrast to our data, Verheijck et al. (2001) and Miquerol et al. (2004) did not observe Cx40 expression in the SAN. Such a discrepancy can be due to the presence of atrial myocytes in our SAN samples. Consistent with our results, the expression of Cx40 has been previously reported at the protein level in the mouse AVN (Miquerol et al. 2004).
Among the 61 channel α- and β-subunits expressed in the mouse heart (the present study), eight presented higher expression in the V than in the A. In the mouse, a key role of Nav1.5 in ventricular depolarization and conduction of the impulse has been demonstrated (Papadatos et al. 2002). The expression profile for Kv1.5, which predominates in the mouse ventricle, is in contrast with previous findings in human (Wang et al. 1993), dog (Fedida et al. 2003), rat (Dixon & McKinnon, 1994) and pig (Knobloch et al. 2002), which show preferential expression in the atria. Mice in which Kv1.5 has been inactivated (London et al. 1998) exhibit QT prolongation, highlighting the key role of Kv1.5 in ventricular repolarization. The ATP-dependent K+ channel Kir6.1 is two to three times more expressed in the V than in other cardiac regions (Fig. 5). In the mouse heart, this channel is expressed in mitochondria (Lacza et al. 2003) and also in the coronary arteries (Miki et al. 2002). It has been suggested from Western blot analysis that Kir6.1 subunits are present in the plasma membrane of murine cardiomyocytes (Rosner et al. 2002). Kir6.1-null mice (Miki et al. 2002) exhibit a normal pinacidil-induced glibenclamide-sensitive outward current in their cardiomyocytes, suggesting that Kir6.1 is not a component of the sarcolemmal KATP channel. Our Kir6.1 expression pattern may be caused by a possible enrichment of ventricular tissue with mitochondria and vessels. Among the channel β-subunits that we screened, the sulphonylurea receptor SUR2 was the most abundant transcript and predominated in the ventricle. SUR2A is associated with Kir6.2 for constituting the cardiac KATP channel (Inagaki et al. 1996), and SUR2B is associated with Kir6.2 for constituting the nonvascular smooth muscle KATP channels (Yamada et al. 1997). SUR2B is associated with Kir6.1 for constituting the vascular smooth muscle KATP channels (Yamada et al. 1997). Neither SUR2A nor SUR2B are present in the mitochondria (Lacza et al. 2003).
Limitations of the study
The heart is a nonhomogeneous tissue made not only with myocytes, but also with smooth muscle and other noncardiac cells such as fibroblasts and intracardiac neurones. In addition, heterogeneity also applies to various cardiac cell types within the same structure. An example of this heterogeneity is the SAN. Indeed, it has been shown that the SAN region is composed of a central core enriched of dominant primary pacemaker cells, together with a periphery which includes both transitional cells as well as cells showing atrial phenotype (Boyett et al. 2000; Verheijck et al. 2001; Mangoni & Nargeot, 2001). This heterogeneous cellular organization has been proposed to play an important role in the regulation of pacemaker activity of the SAN in toto (for review see Boyett et al. 2000). By essence, our technique cannot take into account the fine structure of the nodes and of the other cardiac chambers. Another limitation of our approach is found in the anatomical definition of the SAN and the AVN. Cellular heterogeneity in the different cardiac regions has some consequences on the interpretation of our data. For example, the presence of abundant nerve terminals in the AVN may partly account for the source of neuronal Na+ channels such as Nav1.1 and Nav1.7. Also a consequence, the level of expression of genes that was down-regulated in the nodes may have been overestimated by about 40%. This applies, for example, to Cx40, Cx43, Kv4.2, Kir3.1 and Kir2.1. Similarly, the level of expression of genes that are up-regulated in the nodes may have been underestimated by about 40%. These include genes encoding Cav3.1, Nav1.1, Nav1.7, Kv1.1, Kv1.6, MinK and Kvβ1. It is interesting to note that all these transcripts exhibit a clear predominance in the AVN, suggesting that contamination of the SAN with atrial myocytes may mask some differences between the SAN and the working myocardium. Moreover, message levels quantified in the present study are not necessarily correlated with protein expression, as with post-transcriptional or/and translational adaptation. Besides all the limitations, it is important to stress that the expression pattern of several ionic channels belonging to different gene families that we reported here is consistent with the current literature on their functional role. This observation demonstrates the validity of our high-throughput approach for the characterization of the global expression pattern of ionic channels in the different cardiac tissues that we have considered.
- 2004). Differential modulation of Kv1 channel-mediated currents by co-expression of Kvbeta3 subunit in a mammalian cell-line. Mol Membr Biol 21, 19–25.DOI: 10.1080/09687680310001597749 , & (
- 1984). Outliers in Statistical Data. John Wiley & Sons, New York . & (
- 1997). The newborn rabbit sino-atrial node expresses a neuronal type I-like Na+ channel. J Physiol 498, 641–648. , , , & (
- 1996). A list of vertebrate cardiac ionic currents nomenclature, properties, function and cloned equivalents. Cardiovasc Res 32, 455–481. , , , , & (
- 2000). The sinoatrial node, a heterogeneous pacemaker structure. Cardiovasc Res 47, 658–687. , & (
- 1996). In situ hybridization reveals extensive diversity of K+ channel mRNA in isolated ferret cardiac myocytes. Circ Res 78, 1083–1089. , , , , & (
- 1998). High throughput analysis of differential gene expression. J Cell Biochem Suppl 30– 31 , 286–296. , , , , , et al. (
- 2003). The electrophysiological properties of spontaneously beating pacemaker cells isolated from mouse sinoatrial node. J Physiol 550, 169–180. , & (
- 2004). Expression of ion channel genes in mouse heart. J Physiol 557, C8. , , , & (
- 1994). Quantitative analysis of potassium channel mRNA expression in atrial and ventricular muscle of rats. Circ Res 75, 252–260. & (
- 1998). Cluster analysis and display of genome-wide expression patterns. Proc Natl Acad Sci U S A 95, 14863–14868.DOI: 10.1073/pnas.95.25.14863 , , et al. (
- 2003). Kv1.5 is an important component of repolarizing K+ current in canine atrial myocytes. Circ Res 93, 744–751.DOI: 10.1161/01.RES.0000096362.60730.AE , , , , , et al. (
- 2000). Functional properties of a new voltage-dependent calcium channel alpha(2) delta auxiliary subunit gene (CACNA2D2). J Biol Chem 275, 12237–12242.DOI: 10.1074/jbc.275.16.12237 , , , , , et al. (
- 1993). Regional differences in troponin I isoform switching during rat heart development. Dev Biol 156, 253–264.DOI: 10.1006/dbio.1993.1074 , , , & (
- 1999). Molecular basis of transient outward K+ current diversity in mouse ventricular myocytes. J Physiol 521, 587–599.DOI: 10.1111/j.1469-7793.1999.00587.x , , & (
- 1993). How to Detect and Handle Outliers. American Society for Quality Control, Milwaukee , WI , USA . & (
- 1996). A family of sulfonylurea receptors determines the pharmacological properties of ATP-sensitive K+ channels. Neuron 16, 1011–1017.DOI: 10.1016/S0896-6273(00)80124-5 , , , , , et al. (
- 1993). Cardiac pacemaking in the sinoatrial node. Physiol Rev 73, 197–227. , & (
- 2001). Sodium channel beta subunits: anything but auxiliary. Neuroscientist 7, 42–54. (
- 1992). Primary structure and functional expression of the beta 1 subunit of the rat brain sodium channel. Science 256, 839–842. , , , , , et al. (
- 1999). Fast inactivation of a brain K+ channel composed of Kv1.1 and Kvbeta1.1 subunits modulated by G protein beta gamma subunits. EMBO J 18, 1245–1256.DOI: 10.1093/emboj/18.5.1245 , , , , & (
- 2001). Lack of the burst firing of thalamocortical relay neurons and resistance to absence seizures in mice lacking alpha (1G) T-type Ca2+ channels. Neuron 31, 35–45.DOI: 10.1016/S0896-6273(01)00343-9 , , , , , et al. (
- 2004). Basic mechanisms of cardiac impulse propagation and associated arrhythmias. Physiol Rev 84, 431–488.DOI: 10.1152/physrev.00025.2003 & (
- 1991). The potassium channel MBK1 (Kv1.1) is expressed in the mouse retina. Cell Mol Neurobiol 11, 611–622.DOI: 10.1007/BF00741449 , , , & (
- 2002). Electrophysiological and antiarrhythmic effects of the novel I (Kur) channel blockers, S9947 and S20951, on left vs. right pig atrium in vivo in comparison with the IKr blockers dofetilide, azimilide, d,l-sotalol and ibutilide. Naunyn Schmiedebergs Arch Pharmacol 366, 482–487.DOI: 10.1007/s00210-002-0599-x , , , , & (
- 2001). 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 107, 801–813. , , , , , et al. (
- 2003). Heart mitochondria contain functional ATP-dependent K+ channels. J Mol Cell Cardiol 35, 1339–1347.DOI: 10.1016/S0022-2828(03)00249-9 , , , , & (
- 2001). International Human Genome Sequencing Consortium. Initial sequencing and analysis of the human genome. Nature 409, 860–921.DOI: 10.1038/35057062 , , , , , et al. (
- 2003). Microarray analysis reveals complex remodeling of cardiac ion channel expression with altered thyroid status: relation to cellular and integrated electrophysiology. Circ Res 92, 234–242. , , , , , et al. (
- 2004). Chronic amiodarone remodels expression of ion channel transcripts in the mouse heart. Circulation 110, 3028–3035. , , , , , et al. (
- 2004). Requirement of neuronal- and cardiac-type sodium channels for murine sinoatrial node pacemaking. J Physiol 559, 835–848. , , , , , et al. (
- 2001). Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 25, 402–408.DOI: 10.1006/meth.2001.1262 & (
- 1998). Long QT and ventricular arrhythmias in transgenic mice expressing the N terminus and first transmembrane segment of a voltage-gated potassium channel. Proc Natl Acad Sci U S A 95, 2926–2931.DOI: 10.1073/pnas.95.6.2926 , , , , , et al. (
- 2003). Absence epilepsy and sinus dysrhythmia in mice lacking the pacemaker channel HCN2. EMBO J 22, 216–224.DOI: 10.1093/emboj/cdg032 , , , , , et al. (
- 2003). An unexpected requirement for brain-type sodium channels for control of heart rate in the mouse sinoatrial node. Proc Natl Acad Sci U S A 100, 3507–3512.DOI: 10.1073/pnas.2627986100 , , , , , et al. (
- 2003). Functional role of L-type Cav1.3 Ca2+ channels in cardiac pacemaker activity. Proc Natl Acad Sci U S A 100, 5543–5548.DOI: 10.1073/pnas.0935295100 , , , , , et al. (
- 2001). Properties of the hyperpolarization-activated current (If) in isolated mouse sino-atrial cells. Cardiovasc Res 52, 51–64.DOI: 10.1016/S0008-6363(01)00370-4 & (
- 2002). Biological pacemaker created by gene transfer. Nature 419, 132–133.DOI: 10.1038/419132b , & (
- 2002). Mouse model of Prinzmetal angina by disruption of the inward rectifier Kir6.1. Nat Med 8, 466–472.DOI: 10.1038/nm0502-466 , , , , , et al. (
- 2004). Architectural and functional asymmetry of the His-Purkinje system of the murine heart. Cardiovasc Res 63, 77–86.DOI: 10.1016/j.cardiores.2004.03.007 , , , , , , et al. (
- 2001). Cellular expression and functional characterization of four hyperpolarization-activated pacemaker channels in cardiac and neuronal tissues. Eur J Biochem 268, 1646–1652.DOI: 10.1046/j.1432-1033.2001.02036.x , , , , & (
- 2000). Beta 3: an additional auxiliary subunit of the voltage-sensitive sodium channel that modulates channel gating with distinct kinetics. Proc Natl Acad Sci U S A 97, 2308–2313.DOI: 10.1073/pnas.030362197 , , , , , et al. (
- 2002). FANTOM Consortium RIKEN Genome Exploration Research Group Phase I & II Team. Analysis of the mouse transcriptome based on functional annotation of 60,770 full-length cDNAs. Nature 420, 563–573.DOI: 10.1038/nature01266 , , , , , et al. (
- 2002). Slowed conduction and ventricular tachycardia after targeted disruption of the cardiac sodium channel gene Scn5a. Proc Natl Acad Sci U S A 99, 6210–6215.DOI: 10.1073/pnas.082121299 , , , , , et al. (
- 2000). Deafness and sinoatrial node dysfunction in mice lacking class D,L-type Ca2+ channels. Cell 102, 89–97. , , , , , et al. (
- 1999). Genome-wide analysis of DNA copy-number changes using cDNA microarrays. Nat Genet 23, 41–46. , , , , , et al. (
- 2001). L-type but not T-type calcium current changes during postnatal development in rabbit sinoatrial node. Am J Physiol Heart Circ Physiol 281, H1252–1259. , & (
- 1997). Association and colocalization of the Kvbeta1 and Kvbeta2 beta-subunits with Kv1 alpha-subunits in mammalian brain K+ channel complexes. J Neurosci 17, 8246–8258. , , , , & (
- 2002). Kir6.1 K+ channel subunits are expressed in mouse ventricular myocytes and associate with the actin cytoskeleton. Biophys J 82, 589a (abstract). , , , , & (
- 2002). Differential distribution of cardiac ion channel expression as a basis for regional specialization in electrical function. Circ Res 90, 939–950.DOI: 10.1161/01.RES.0000018627.89528.6F , , & (
- 2003). A fundamental role for KChIPs in determining the molecular properties and trafficking of Kv4.2 potassium channels. J Biol Chem 278, 36445–36454.DOI: 10.1074/jbc.M306142200 , , , , , et al. (
- 2003). The hyperpolarization-activated channel HCN4 is required for the generation of pacemaker action potentials in the embryonic heart. Proc Natl Acad Sci U S A 100, 15235–15240.DOI: 10.1073/pnas.2434235100 , , , , , et al. (
- 2004). Pacemaker channels and sinus node arrhythmia. Trends Cardiovascular Med 14, 23–28.DOI: 10.1016/j.tcm.2003.09.006 , & (
- 1998). Disparate effects of deficient expression of connexin43 on atrial and ventricular conduction: evidence for chamber-specific molecular determinants of conduction. Circulation 97, 686–691. , , , , , et al. (
- 2001). Electrophysiological features of the mouse sinoatrial node in relation to connexin distribution. Cardiovasc Res 52, 40–50.DOI: 10.1016/S0008-6363(01)00364-9 , , , , & (
- 1993). Sustained depolarization-induced outward current in human atrial myocytes. Evidence for a novel delayed rectifier K+ current similar to Kv1.5 cloned channel currents. Circ Res 73, 1061–1076. , & (
- 1999). Elimination of the transient outward current and action potential prolongation in mouse atrial myocytes expressing a dominant negative Kv4 alpha subunit. J Physiol 519, 11–21.DOI: 10.1111/j.1469-7793.1999.0011o.x , & (
- 1997). Sulphonylurea receptor 2B and Kir6.1 form a sulphonylurea-sensitive but ATP-insensitive K+ channel. J Physiol 499, 715–720. , , , , , et al. (
- 2002). Functional roles of Cav1.3 (α1D) calcium channel in sinoatrial nodes: insight gained using gene-targeted null mutant mice. Circ Res 90, 981–987.DOI: 10.1161/01.RES.0000018003.14304.E2 , , , , , et al. (
Supported by special grants from Ouest genopole, INSERM, CNRS, le Ministère français de la Recherche and the Wellcome Trust (M.L.). We thank Karine Haurogné and Jean-Marie Heslan for expert technical assistance.
The online version of this paper can be accessed at: DOI: 10.1113/jphysiol.2004.074047 http://jp.physoc.org/cgi/content/full/jphysiol.2004.074047/DC1 and contains supplementary material consisting of two tables.
This material can also be found at: http://www.blackwellpublishing.com/products/journals/suppmat/tjp/tjp609/tjp609sm.htm
Appendix A1. Containing two tables.
|TJP_609_sm_appendixa1.pdf||64K||Supporting info item|
Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.