Background Of the five sodium channel subtypes expressed in cardiac tissues, the rat (rH1) and human (hH1) isoforms are thought to be the predominant subtypes on the basis of heterologous expression studies. In this study, subtype-specific antibodies and immunocytochemistry were used to confirm protein expression and to localize rH1 protein in cardiac tissues.
Methods and Results Subtype-specific antibodies immunolabeled adult rat heart tissue in a manner identical to that obtained with subtype-nonselective antibodies. All antibodies specifically bound to the surface and t-tubular systems of atrial and ventricular muscle cells. Cytoplasmic labeling, reflecting nascent sodium channels or cytoplasmic stores of sodium channel protein, was apparent. Most notably, all antibodies also specifically labeled the subset of intercalated disks located at the ends but not the sides of adjacent ventricular muscle cells.
Conclusions rH1 is the predominant subtype expressed on rat atrial and ventricular muscle cells. rH1 protein localization in surface and t-tubular membranes is consistent with its proposed role in coordinating membrane depolarization along the length and deep within cardiac muscle cells. rH1 protein localization in terminal intercalated disks suggests that sodium channels may also act as a localized voltage-dependent current amplifier, raising the safety margin for conduction; they also may contribute to anisotropic or saltatory conduction in cardiac tissues. These electrophysiological properties would be particularly important under conditions of altered channel function resulting from ion channel gene defects (eg, long QT syndrome), antiarrhythmic drug therapy, ischemia, or other heart diseases by influencing the electrophysiological substrate for ventricular tachyarrhythmias. (Circulation. 1996;94:3083-3086.)
Molecular studies have identified five distinct sodium channel transcripts in cardiac tissues.1 Identification of rat (rH1) and human (hH1) isoforms as the predominant sodium channel subtypes expressed in rat and human cardiac tissues, respectively, is supported by heterologous expression of cRNA encoding both isoforms. Sodium currents resistant to both TTX and μ-conotoxin but possessing use- and frequency-dependent behavior with class 1 antiarrhythmic drugs were observed, characteristics expected of the predominant sodium channel in the heart.2 3 4 To confirm protein expression and to determine protein localization in atrial and ventricular muscle cells, we used three anti-peptide rH1-specific antibodies to examine rH1 subtype expression in the rat heart. A fourth polyclonal antibody, denoted R12 and developed against a portion of the conserved interdomain 3-4 region of Na(v)1 sodium channels, was used to investigate whether rH1 is the predominant subtype expressed in atrial and ventricular muscle cells. While our findings support this identification, our observation that rH1 protein is localized in terminal intercalated disks suggests additional but previously unsuspected mechanisms by which sodium channels can contribute to cardiac impulse conduction in ventricular muscle.
Preparation of Site-Directed Antisera
rH1-specific polyclonal antibodies were prepared against synthetic oligopeptides corresponding to unique portions of the rH1 cardiac sodium channel sequence.5 Antibody R12 is directed against a sequence that is 100% conserved in all Na(v)1 sodium channels (see Fig 1⇓ and Table⇓). The remaining polyclonal antibodies, which served as negative controls, were previously used in studies of the adult rat skeletal muscle sodium channel (rSkM1). Antibody specificity was previously documented by the ability to identify sodium channel protein on immunoblots and to specifically immunoprecipitate sodium channel protein from crude membrane preparations.5 6 7 8
Adult male Wistar rat hearts freshly frozen in isopentane (Sigma Chemical Co) were stored briefly in liquid nitrogen until use. Cryostat sections (6 μm) were melted onto coverslips and blocked for 20 minutes with 4% BSA/PBS (Sigma). Sections were then incubated for 2 hours with affinity-purified primary antibody diluted 1:25-1:50 (≈1 μg/mL) in 2% BSA/PBS before being washed three times, 20 minutes each, with PBS/0.2% Tween 20. Secondary antibody (rhodamine-conjugated goat anti-rabbit antibody, Cappel) was diluted 1:1000 in 2% BSA/PBS and incubated for 1 hour before the sections were washed with PBS/0.2% Tween 20. All incubations and washes were carried out at room temperature. Sections were blotted dry, mounted with 50% glycerol in PBS, and viewed by use of a Zeiss epifluorescence microscope.
No significant differences in immunolabeling were observed in specimens obtained from different regions or different portions of each region of the heart other than histological differences expected between atrial and ventricular muscle cells. Full-thickness and tangential sections demonstrated no differences in immunolabeling caused by location (epicardial, middle, or endocardial) in the ventricular wall.
Immunolabeling of atrial tissue demonstrated nearly uniform labeling of atrial surface membranes (Fig 2A⇓). One or more eccentrically located focal accumulations of label were present in each cell, most likely representing perinuclear staining of nascent channel protein. The cytoplasm had a faintly stippled appearance, suggesting specific immunolabeling of the rudimentary t-tubular system of atrial cells and possibly cytoplasmic stores of protein available for transport to the surface membrane. No enhanced labeling of gap junction–type structures was observed in atrial tissues.
Immunolabeling of cross sections of adult rat ventricular muscle produced similar findings: nearly uniform surface membrane labeling, eccentrically located focal accumulation of label, but a more intensely and coarsely stippled cytoplasm (Fig 2B⇑). Longitudinal sections of ventricular muscle revealed more specific detail (Fig 2C⇑): Parallel lines corresponding to the Z lines of the fully developed t-tubular system of ventricular muscle were readily apparent (see the inset in Fig 2⇑C). In addition, each of the rH1-specific antibodies specifically labeled the intercalated disk region at the ends of adjacent ventricular muscle cells more intensely than either the t-tubular or surface membranes (Fig 2C⇑).
Extensive negative controls included preadsorption of IgG with immunizing peptide, omission of primary antibody, and use of IgG specific for a sodium channel subtype not specifically expressed in cardiac tissues (antibodies I-31, I-467, and I-1771 directed against similar regions of the adult rat skeletal muscle sodium channel).7 All controls produced background levels of immunolabeling (Fig 2D⇑). Each rH1-specific IgG also produced only background labeling with sections derived from adult rat skeletal muscle, brain, or kidney (data not shown).
To investigate whether significant pools of sodium channel protein other than the rH1 subtype are expressed in atrial and ventricular muscle cells, antibody R12, developed against a portion of the interdomain 3-4 region that is 100% conserved in most sodium channels, was used for immunolabeling. Although staining of neural and conduction system elements was observed with this antisera,9 both the intensity and the pattern of labeling of atrial and ventricular muscle cells were identical to those observed with the rH1-specific antisera. Even though immunocytochemistry is not a quantitative technique, these results suggest that other sodium channel subtypes, if present in atrial and ventricular muscle cells, are neither expressed in significant quantities nor located in additional subcellular compartments. These findings support the notion that the rH1 subtype comprises a substantial portion if not the majority of sodium channel protein expressed in atrial and ventricular muscle cells.
Labeling of both surface and t-tubular membrane systems of atrial and ventricular muscle cells is consistent with the expected presence of rH1 sodium channel protein in these membranes. In these locations, rH1 sodium channel protein would serve to ensure uniform conduction of electric depolarization both along the length and deep within cardiac muscle cells. Labeling of terminal intercalated disks in ventricular muscle suggests that this subset of nexi contains a targeting mechanism that gives rise to rH1 protein localization. Labeling of terminal intercalated disks appears to be more intense than labeling of either surface or t-tubular membranes. This could be due to clustering of sodium channel protein, giving rise to an increased density of rH1 sodium channels per unit membrane, similar to clustering of sodium channel protein at nodes of Ranvier or axon hillocks in nervous tissues10 11 12 13 and at neuromuscular junctions in skeletal muscle.14 15 Alternatively, the intense labeling may represent a constant density of sodium channel per unit membrane (similar to that in surface or t-tubular membranes) that, because of membrane redundancy, gives the appearance of an increased concentration of channel protein.
Each rH1-specific antibody used in this study specifically identifies the rH1 α-subunit on Western blots and is able to specifically immunoprecipitate rH1 protein from partially purified rat heart membrane proteins.5 The extensive use of controls, the ability of the R12 antisera to duplicate the results of the rH1-specific antisera, and the selective labeling of only a subset of intercalated disks suggest that nonspecific labeling of membrane regions containing high protein concentrations is an unlikely explanation for the patterns of immunolabeling observed in this study. Additional studies are needed, however, to determine the functional status of rH1 protein in each membrane environment.
At least one other ion channel has been immunolocalized to ventricular intercalated disks. By use of two anti-peptide polyclonal antibodies, the Kv1.5 potassium channel was immunolocalized to ventricular but not atrial intercalated disks in unfixed sections of explanted human cardiac tissue.16 Unlike the present study, however, both terminal and lateral intercalated disks were labeled, and little or no visible immunolabeling of surface or t-tubular membranes was observed.
Given our present understanding of cardiac electrophysiology and the space constant of ventricular muscle cells, it is unclear why sodium or other ion channels should be located in regions of intercellular communication. A previous report suggested that intercalated disks might represent a membrane insertion site for newly synthesized channel proteins or that this region may be more amenable for anchoring membrane proteins.16 However, the involvement of type 1 antiarrhythmic drugs in arrhythmia prevention, termination, and induction (through proarrhythmic side effects) and the involvement of gene defects of sodium and potassium channels in the genesis of the long QT syndrome17 18 suggest other possibilities.
The first is that sodium channels in terminal intercalated disks might serve as voltage-dependent current amplifiers. Thus, rather than relying solely on passive electric conduction through connexin connections, rH1 protein in intercalated disks would increase the safety factor for electric conduction by enhancing electrotonic conduction. A second and related possibility is that sodium channels in terminal intercalated disks may allow electric impulses to be conducted in a saltatory fashion, having direct effects on conduction velocity and tissue refractoriness. Thus, regulation of the density of channels in terminal intercalated disks would provide a means of modulating both conduction velocity and tissue refractoriness by altering action potential upstroke velocity. A third possibility is that sodium channels concentrated only in terminal intercalated disks may contribute to the anisotropic behavior of cardiac muscle by enhancing longitudinal rather than transverse conduction. Thus, by having direct effects on the safety factor for conduction, conduction velocity, refractoriness, and the anisotropic behavior of cardiac tissues, sodium channels in intercalated disks could play an important role in both arrhythmia initiation and propagation; this could explain their success (and failure) as targets for antiarrhythmic drug therapy. Abnormal or inhomogeneous coupling of myocardial cells induced by defective ion channels (as in the LQT3 mutation in the long QT syndrome18 ), antiarrhythmic drugs, cardiac ischemia, or other heart disease could thus promote arrhythmias by differential effects on sodium channels in terminal intercalated disks in different parts of the ventricle, thus giving rise to a functional arrhythmic circuit. Each of these hypotheses needs to be tested experimentally.
This work was supported in part by a Grant-in-Aid from the American Heart Association (93013120) and by a Merit Review Award from the Department of Veterans Affairs.
- Received August 20, 1996.
- Revision received October 3, 1996.
- Accepted October 9, 1996.
- Copyright © 1996 by American Heart Association
Cohen SA, Barchi RL. Cardiac sodium channel structure and function. Trend Cardiovasc Med.. 1992;2:133-140.
Gellens ME, George AL, Chen LQ, Chahine M, Horn R, Barchi RL, Kallen RG. Primary structure and functional expression of the human cardiac tetrodotoxin-insensitive voltage-dependent sodium channel. Proc Natl Acad Sci U S A.. 1992;89:554-558.
White MM, Chen LQ, Kleinfield R, Kallen RG, Barchi RL. SkM2, a Na+ channel cDNA clone from denervated skeletal muscle, encodes a tetrodotoxin-insensitive Na+ channel. Mol Pharmacol.. 1991;39:604-608.
Cohen SA, Levitt LK. Partial characterization of the rH1 sodium channel protein from rat heart using subtype-specific antibodies. Circ Res.. 1993;73:735-742.
Haimovich B, Bonilla E, Casadei J, Barchi R. Immunocytochemical localization of the mammalian voltage-dependent sodium channel using polyclonal antibodies against the purified protein. J Neurosci.. 1984;4:2259-2268.
Zwerling SJ, Cohen SA, Barchi RL. Analysis of protease-sensitive regions in the skeletal muscle sodium channel in vitro and implications for channel tertiary structure. J Biol Chem.. 1991;266:4574-4580.
Sun W, Barchi RL, Cohen SA. Probing sodium channel cytoplasmic domain structure: evidence for the interaction of the rSkM1 amino- and carboxy-termini. J Biol Chem.. 1995;270:22271-22276.
Petrecca K, Amellal F, Laird DW, Cohen SA, Shrier A. Sodium channel distribution within the rabbit AV node and surrounding myocardium as analyzed by confocal microscopy. PACE.. 1996;19:665. Abstract.
Devor M, Govrinlippmann R, Angelides K. Na+ channel immunolocalization in peripheral mammalian axons and changes following nerve injury and neuroma formation. J Neurosci.. 1993;13:1976-1992.
Wollner D, Catterall W. Localization of sodium channels in axon hillocks and initial segments of retinal ganglion cells. Proc Natl Acad Sci U S A.. 1986;83:8424-8428.
Angelides KJ, Elmer LW, Loftus D, Elson E. Distribution and lateral mobility of voltage-dependent sodium channels in neurons. J Cell Biol.. 1988;106:1911-1926.
Caldwell JH, Campbell DT, Beam KG. Sodium channel distribution in vertebrate skeletal muscle. J Gen Physiol.. 1986;87:907-932.
Lupa MT, Krzemien DM, Schaller KL, Caldwell JH. Aggregation of sodium channels during development and maturation of the neuromuscular junction. J Neurosci.. 1993;13:1326-1336.
Mays DJ, Foose JM, Philipson LH, Tamkun MM. Localization of the Kv1.5 K+ channel protein in explanted cardiac tissue. J Clin Invest.. 1995;96:282-292.