Introduction
The pore-forming subunit of the neuronal voltage-gated sodium channel (NaV1.6) isoform of the voltage-gated sodium channel was first discovered in and is now a well-established component of the peripheral and central nervous systems (Caldwell et al., Reference Caldwell, Schaller, Lasher, Peles and Levinson2000; Wang et al., Reference Wang, Ou and Wang2017). Hence, it is the common moniker neuronal sodium channel. Recently, NaV1.6 has been identified within cardiac myocytes, localized near calcium (Ca2+) handling machinery in transverse tubules (t-tubules) (Maier et al., Reference Maier, Westenbroek, McCormick, Curtis, Scheuer and Catterall2004; Zimmer et al., Reference Zimmer, Haufe and Blechschmidt2014; Radwański et al., Reference Radwański, Brunello, Veeraraghavan, Ho, Lou, Makara, Belevych, Anghelescu, Priori, Volpe, Hund, Janssen, Mohler, Bridge, Poelzing and Gyorke2015, Reference Radwański, Ho, Veeraraghavan, Brunello, Liu, Belevych, Unudurthi, Makara, Priori, Volpe, Armoundas, Dillmann, Knollmann, Mohler, Hund and Györke2016). These neuronal channels contribute a small portion of the total sodium current compared to cardiac sodium channels [pore-forming subunit of the cardiac voltage-gated sodium channel (NaV1.5)] (Maier, Reference Maier2009). However, recent studies indicate that Na+ influx via these channels may disproportionately impact Ca2+ dynamics in both health and disease, via electrogenic Na+–Ca2+ exchange mediated by the sodium–calcium exchanger (NCX) (Moreno & Clancy, Reference Moreno and Clancy2012; Radwański et al., Reference Radwański, Greer-Short and Poelzing2013, Reference Radwański, Brunello, Veeraraghavan, Ho, Lou, Makara, Belevych, Anghelescu, Priori, Volpe, Hund, Janssen, Mohler, Bridge, Poelzing and Gyorke2015, Reference Radwański, Ho, Veeraraghavan, Brunello, Liu, Belevych, Unudurthi, Makara, Priori, Volpe, Armoundas, Dillmann, Knollmann, Mohler, Hund and Györke2016; Helms et al., Reference Helms, Alvarado, Yob, Tang, Pagani, Russell, Valdivia and Day2016; Sato et al., Reference Sato, Clancy and Bers2017). Further, these studies suggest that such a role for NaV1.6 may be predicated upon its physical proximity to Ca2+ cycling proteins within t-tubules (Veeraraghavan et al., Reference Veeraraghavan, Gyorke and Radwański2017; Radwański et al., Reference Radwański, Johnson, Gyorke and Veeraraghavan2018). Thus, there is a significant need to understand the spatial organization of NaV1.6 within cardiac myocytes, particularly in relation to Ca2+ cycling proteins.
Super-resolution microscopy techniques, which are ideally suited to address this problem, require high-fidelity antibodies against target proteins. Therefore, we undertook the development of a novel antibody against NaV1.6 in order to facilitate the investigation of NaV1.6 localization in the heart and other tissues. Following an approach previously applied to the sodium channel NaV1.5 with significant success (Veeraraghavan et al., Reference Veeraraghavan, Hoeker, Alvarez-Laviada, Hoagland, Wan, King, Sanchez-Alonso, Chen, Jourdan, Isom, Deschenes, Smyth, Gorelik, Poelzing and Gourdie2018), we raised a rabbit polyclonal antibody against a C-terminal epitope on NaV1.6. Through the use of a variety of strategies, we demonstrate that this antibody recognizes NaV1.6 with high avidity and selectivity. Finally, we use this novel tool in super-resolution microscopy experiments to demonstrate for the first time that over half of the NaV1.6 channels in cardiac myocytes are located within 100 nm of ryanodine receptor Ca2+ release channels.
Methods
All animal procedures were approved by The Ohio State University Institutional Animal Care and Use Committee and conformed to the Guide for the Care and Use of Laboratory Animals published by the U.S. National Institutes of Health (NIH Publication No. 85-23, revised 2011).
Custom NaV1.6 Antibody Development
The development of a custom rabbit polyclonal antibody was undertaken as previously described (Veeraraghavan et al., Reference Veeraraghavan, Hoeker, Alvarez-Laviada, Hoagland, Wan, King, Sanchez-Alonso, Chen, Jourdan, Isom, Deschenes, Smyth, Gorelik, Poelzing and Gourdie2018). Our novel antibody was raised against a C-terminal epitope on NaV1.6: “ENGGTHREKKESTP”, which correspond to amino acids 1926–1939 on human NaV1.6 (Fig. 1). A C-terminal epitope was selected to enable easy access for antibody binding. Further, this specific region was chosen based on its uniqueness to NaV1.6 (compared to other NaV1.x proteins) and a high degree of conservation across mammalian species. A BLAST search revealed a highly significant (E = 3 × 10−7) correspondence between this epitope and the NaV1.6 protein from various species but no significant similarities (E > 3) to other known protein sequences.
Fig. 1. NaV1.6 C-terminal epitope. (a) Schematic showing the location of epitope on the NaV1.6 C-terminus. (b) Comparison of NaV isoforms. (c) Comparison with other species.
Immunization and care of rabbits, collection of sera, and affinity purification of the antibody were performed by Pierce Custom Antibody Services (ThermoFisher Inc). A New Zealand white rabbit was immunized with a peptide corresponding to the epitope with subsequent immunizations at 14, 42, 56, 104, 159, and 222 days after the initial immunization. Serum was collected prior to immunization (day 0) and at days 28, 56, 70, 72, 118, 120, 173, 236, and 238 following the initial immunization. Each serum sample was individually evaluated for immunoreactivity in confocal immuno-fluorescence studies using murine cardiac sections with the day 0 serum evaluated as a nonspecific background control. The serum from day 173 showed the strongest immunoreactivity (Fig. 2) and was hence selected for affinity purification. Affinity purification was performed using a purpose-built immobilized antigen column and the purified antibody evaluated by ELISA. The concentration of the purified antibody was measured at 0.1 mg/mL.
Fig. 2. Rabbit polyclonal anti-NaV1.6 antibody—raw serum: representative confocal images from (a) myocardial sections and (b) isolated cardiac myocytes immunolabeled using our rabbit polyclonal NaV1.6 antibody (raw serum; red) and a mouse monoclonal RyR2 antibody (green). Single optical sections (XY) are presented along with orthogonal projections (YZ, XZ) as well as images of individual fluorescence channels. (c,d) 3D views of the same samples. (e,f) Intensity profiles demonstrate the striated pattern of NaV1.6 immunosignals closely associating with RyR2 immunosignal.
Mouse Models
Wild-type (WT) mice (C57BL6 background; Cat. #000664) and global knockout of NaV1.6 (NaV1.6 KO) mice (C3Fe.Cg background; Cat. #003798) were purchased from Jackson Laboratories. Cardiac-specific NaV1.6 knockout mice (cNaV1.6KO) were obtained by crossing C57BL6 mice with loxP sites flanking Exon 1 of the Scn8a gene (custom generated by the University of Utah Transgenic and Gene Targeting Core and the University of Utah Mutation Generation and Detection Core) with mice expressing Cre driven by cardiac-specific alpha myosin-heavy chain (Myh6) promoter (Jackson Laboratories; Cat. #011038).
Tissue Collection and Myocyte Isolation
Male C57/BL6 mice (30 g, 6–18 weeks) were anesthetized with 5% isoflurane mixed with 100% oxygen (1 L/min). After the loss of consciousness, anesthesia was maintained with 3–5% isoflurane mixed with 100% oxygen (1 L/min). Once the animal was in a surgical plane of anesthesia, the heart was excised and frozen for cryosectioning or perfused (at 40–55 mm Hg) using Langendorff preparations with oxygenated Tyrode's solution (containing in mM: NaCl 140, KCl 5.4, MgCl2 0.5, dextrose 5.6, HEPES 10; pH adjusted to 7.4) at 37°C as previously described (Veeraraghavan & Poelzing, Reference Veeraraghavan and Poelzing2008; Radwański et al., Reference Radwański, Veeraraghavan and Poelzing2010, Reference Radwański, Brunello, Veeraraghavan, Ho, Lou, Makara, Belevych, Anghelescu, Priori, Volpe, Hund, Janssen, Mohler, Bridge, Poelzing and Gyorke2015; Veeraraghavan et al., Reference Veeraraghavan, Lin, Hoeker, Keener, Gourdie and Poelzing2015, Reference Radwański, Johnson, Gyorke and Veeraraghavan2018). Cardiac myocytes were isolated via enzymatic digestion, plated on laminin-coated glass coverslips, fixed with 2% paraformaldehyde (PFA) for 5 min, washed in phosphate-buffered saline (PBS) (3 × 10 min) and stored in PBS at 4°C for immunolabeling.
Cell Culture
Native Chinese hamster ovarian cells (CHO) modified to stably express either NaV1.5 or NaV1.6 sodium channels were plated on laminin-coated glass coverslips and maintained in culture in F-12 HAM media at 37°C under a 5% O2, 95% CO2 atmosphere. Once at 50% confluence, the cells were fixed with 2% PFA for 5 min, washed in PBS (3 × 10 min), and stored in PBS at 4°C for immunolabeling.
Fluorescent Immunolabeling
Immuno-fluorescent labeling of PFA-fixed tissue sections (5 µm) and cells was performed, as previously described (Veeraraghavan et al., Reference Veeraraghavan, Lin, Hoeker, Keener, Gourdie and Poelzing2015, Reference Veeraraghavan, Hoeker, Alvarez-Laviada, Hoagland, Wan, King, Sanchez-Alonso, Chen, Jourdan, Isom, Deschenes, Smyth, Gorelik, Poelzing and Gourdie2018; Radwański et al., Reference Radwański, Ho, Veeraraghavan, Brunello, Liu, Belevych, Unudurthi, Makara, Priori, Volpe, Armoundas, Dillmann, Knollmann, Mohler, Hund and Györke2016; Koleske et al., Reference Koleske, Bonilla, Thomas, Zaman, Baine, Knollmann, Veeraraghavan, Gyorke and Radwański2018). Briefly, samples were permeabilized with Triton X-100 (0.2% in PBS for 15 min at room temperature) and treated with a blocking agent (1% bovine serum albumin, 0.1% triton in PBS for 2 h at room temperature) prior to labeling with primary antibodies (overnight at 4°C). Ryanodine receptors were labeled with a mouse monoclonal antibody (Cat. #: MA3-916) from Invitrogen (Rockford, IL, USA), and β-tubulin III was labeled using a mouse monoclonal antibody (Cat. #: MAB5564) from Millipore Sigma (St. Louis, MO, USA). Samples were then washed in PBS (3 × 5 min at room temperature) prior to labeling with secondary antibodies. For confocal microscopy, samples were then labeled with goat anti-rabbit AlexaFluor 568 (1:4000; ThermoFisher Scientific, Grand Island, NY, USA) and goat anti-mouse AlexaFluor 488 (1:4000; ThermoFisher Scientific, Grand Island, NY, USA) secondary antibodies. Samples were then washed in PBS (3 × 5 min at room temperature) and mounted in ProLong Gold (Invitrogen, Rockford, IL, USA). For proximity ligation assay (PLA), samples were labeled using appropriate Duolink secondary antibodies (Sigma, St. Louis, MO, USA) per manufacturer's instructions (Radwański et al., Reference Radwański, Ho, Veeraraghavan, Brunello, Liu, Belevych, Unudurthi, Makara, Priori, Volpe, Armoundas, Dillmann, Knollmann, Mohler, Hund and Györke2016). Briefly, this assay uses complementary oligonucleotide-labeled secondary antibodies, which undergo ligation when co-labeled proteins are located <40 nm apart, to detect protein–protein interactions with high sensitivity. For super-resolution STochastic Optical Reconstruction Microscopy (STORM), samples were labeled with goat anti-mouse Alexa 647 (1:4000) and goat anti-rabbit Biotium CF 568 (1:4000) secondary antibodies (ThermoFisher Scientific, Grand Island, NY, USA). Samples were then washed in PBS (3 × 5 min at room temperature) and stored in the Scale U2 buffer for 48 h at 4°C (Veeraraghavan & Gourdie, Reference Veeraraghavan and Gourdie2016; Veeraraghavan et al., Reference Veeraraghavan, Lin, Keener, Gourdie and Poelzing2016, Reference Veeraraghavan, Hoeker, Alvarez-Laviada, Hoagland, Wan, King, Sanchez-Alonso, Chen, Jourdan, Isom, Deschenes, Smyth, Gorelik, Poelzing and Gourdie2018).
Confocal Microscopy
Confocal imaging was performed using an A1R-HD laser-scanning confocal microscope equipped with four solid-state lasers (405, 488, 560, and 640 nm, 30 mW each), a 63×/1.4 numerical aperture oil immersion objective, two GaAsP detectors, and two high sensitivity photomultiplier tube detectors (Nikon, Melville, NY, USA). Individual fluorophores were imaged sequentially with the excitation wavelength switching at the end of each frame.
Western Immunoblotting
Whole cell lysates were prepared from frozen WT mouse hearts as previously described (Koleske et al., Reference Koleske, Bonilla, Thomas, Zaman, Baine, Knollmann, Veeraraghavan, Gyorke and Radwański2018; Veeraraghavan et al., Reference Veeraraghavan, Hoeker, Alvarez-Laviada, Hoagland, Wan, King, Sanchez-Alonso, Chen, Jourdan, Isom, Deschenes, Smyth, Gorelik, Poelzing and Gourdie2018). These were electrophoresed on 4–15% TGX Stain-free gels (BioRad, Hercules, CA, USA) before being transferred onto a nitrocellulose membrane. The membranes were probed with our novel rabbit polyclonal antibody against NaV1.6 as well as mouse monoclonal antibody against glyceraldehyde 3-phosphate dehydrogenase (GAPDH) (loading control; Fitzgerald Industries, Acton, MA, USA), followed by goat anti-rabbit and goat anti-mouse horse radish peroxidase-conjugated secondary antibodies (Promega, Madison, WI, USA). Signals were detected by chemiluminescence using SuperSignal West Femto Extended Duration Substrate (ThermoFisher Scientific, Grand Island, NY, USA) and imaged using a Chemidoc MP imager (BioRad, Hercules, CA, USA).
STORM Super-Resolution Imaging
STORM imaging was performed as previously described (Veeraraghavan & Gourdie, Reference Veeraraghavan and Gourdie2016; Veeraraghavan et al., Reference Veeraraghavan, Hoeker, Alvarez-Laviada, Hoagland, Wan, King, Sanchez-Alonso, Chen, Jourdan, Isom, Deschenes, Smyth, Gorelik, Poelzing and Gourdie2018; Bonilla et al., Reference Bonilla, Belevych, Baine, Stepanov, Mezache, Bodnar, Liu, Volpe, Priori, Weisleder, Sakuta, Carnes, Radwański, Veeraraghavan and Gyorke2019). Briefly, imaging was performed using a Vutara 352 microscope (Bruker Nano Surfaces, Middleton, WI, USA) equipped with biplane three-dimensional (3D) detection, and fast scientific Complementary metal–oxide–semiconductor imaging achieving 20 nm lateral and 50 nm axial resolution. Individual fluorophore molecules were localized with a precision of 10 nm. Registration of the two color channels was accomplished using localized positions of several TetraSpeck Fluorescent Microspheres (ThermoFisher Scientific, Carlsbad, CA, USA) scattered throughout the field of view. The images were quantitatively analyzed using STORM-based relative localization analysis (STORM-RLA) as previously described (Veeraraghavan & Gourdie, Reference Veeraraghavan and Gourdie2016).
Results
NaV1.6 Antibody Development—Epitope Selection
In order to facilitate the investigation of the NaV1.6 neuronal sodium channel isoform in the heart and other tissues, we developed a novel high-affinity rabbit polyclonal antibody against a C-terminal epitope on NaV1.6 (Fig. 1). The region from amino acid 1926 to 1939 on the NaV1.6 C-terminus was chosen based on its uniqueness compared to other NaV isoforms (Fig. 1b). Additionally, this region is conserved across NaV1.6 from several mammalian species (Fig. 1c). Further, a BLAST search for this sequence yielded no significant results, affirming its specificity for NaV1.6.
Validation of NaV1.6 Antibody
Raw sera, collected from rabbits 173 days following immunization, showed the strong immuno-fluorescent signal (red) in a striated pattern with moderately low background in laser-scanning confocal microscopy (LSCM) images of transmural sections from murine ventricular myocardium (Fig. 2a) and in isolated murine cardiomyocytes (Fig. 2b). In both cases, NaV1.6 immuno-fluorescent signals demonstrated close association (indicated by yellow pixels in the overlay images) with ryanodine receptor 2 (RyR2; green), which is enriched at t-tubules (Radwański et al., Reference Radwański, Ho, Veeraraghavan, Brunello, Liu, Belevych, Unudurthi, Makara, Priori, Volpe, Armoundas, Dillmann, Knollmann, Mohler, Hund and Györke2016). The periodic, striated expression pattern of NaV1.6 and RyR2, as well as their close association, is illustrated by intensity profiles generated from a single optical section in Figures 2e and 2f. Orthogonal projections and 3D renderings demonstrate that the close association between NaV1.6 and RyR2 was consistent along the X, Y, and Z dimensions (Figs. 2a–2d) in both tissue and isolated myocytes.
The selectivity of the antibody for the NaV1.6 isoform was tested using Chinese hamster ovary (CHO) cells stably expressing either NaV1.6 (CHO-NaV1.6) or NaV1.5 (CHO-NaV1.5) isoform of voltage-gated sodium channels. The NaV1.5 isoform was chosen for this comparison given that it is the most abundant sodium channel protein in cardiac muscle. CHO-NaV1.6 cells expressed the strong immuno-fluorescent signal (red) at the cell borders and within perinuclear regions (nuclei in blue) (Figs. 3a, 3b), while CHO-NaV1.5 cells revealed the minimal immuno-fluorescent signal (Figs. 3c, 3d).
Fig. 3. Antibody specificity in the heterologous expression system: single optical sections and 3D views of confocal images from CHO cells stably expressing NaV1.6 (a,b) or NaV1.5 (c,d) labeled with our NaV1.6 antibody (raw serum; red) as well as a DAPI (4′,6-diamidino-2-phenylindole) nuclear stain (blue).
Next, we sought to improve the signal-to-background characteristics of our antibody through affinity purification. In LSCM images of transmural sections from murine ventricular myocardium (Fig. 4a) and in isolated murine cardiomyocytes (Fig. 4b), the purified antibody maintained the strong immuno-fluorescent signal (red) in a striated pattern as seen with the raw sera. However, the level of background fluorescence was markedly reduced. The close association between NaV1.6 and RyR2 immuno-fluorescent signals was maintained and became more apparent with the reduced background signal of the purified sera. The periodic, striated expression pattern of NaV1.6 and RyR2, as well as their close association, is illustrated by intensity profiles generated from a single optical section in Figures 4e and 4f. Once again, orthogonal projections and 3D renderings demonstrate that the close association between NaV1.6 and RyR2 was consistent in all three dimensions (Figs. 4a–4d) in both tissue and isolated myocytes. In a further experiment, murine ventricular sections were labeled with the purified antibody in the presence of a peptide corresponding to its epitope (5× the immunoglobulin concentration of the purified antibody). The peptide abrogated the NaV1.6 immuno-fluorescent signal (red) indicating a high specificity of the purified antibody to the targeted epitope (Figs. 5a, 5b).
Fig. 4. Rabbit polyclonal anti-NaV1.6 antibody—affinity purified: Representative confocal images from (a) myocardial sections and (b) isolated cardiac myocytes immunolabeled using our rabbit NaV1.6 antibody (raw serum; red) and a mouse monoclonal RyR2 antibody (green). Single optical sections (XY) are presented along with orthogonal projections (YZ, XZ) as well as images of individual fluorescence channels. (c,d) 3D views of the same samples. (e,f) Intensity profiles demonstrate the striated pattern of NaV1.6 immunosignals closely associating with RyR2 immunosignals.
Fig. 5. Antibody specificity—peptide inhibition. Confocal images of myocardial sections immunolabeled for RyR2 (green) and NaV1.6 (red) in the (a) absence and (b) presence of a peptide corresponding to the NaV1.6 C-terminal epitope. Western immunoblots of whole cell lysates of three WT murine hearts (1 per lane) in the (c) absence and (d) presence of a peptide corresponding to the NaV1.6 C-terminal epitope. In both cases, GAPDH was detected as a loading control.
Additionally, we validated our antibody in Western blot studies of whole cell lysates of adult murine myocardium. The antibody detected a distinct band at 250 kD, consistent with the predicted molecular weight of the NaV1.6 protein (Fig. 5c; larger representation in Supplementary Fig. 1). Furthermore, this band was abrogated in the presence of the peptide corresponding to the epitope (Fig. 5d). These results are congruent with those from the immunolabeling studies and further validate the specificity of our antibody.
Further validation of NaV1.6 isoform selectivity of our antibody was performed using mice with a global loss of NaV1.6 (NaV1.6 KO; Figs. 6c, 6f) as well as mice with cardiac-specific loss of NaV1.6 (cNaV1.6 KO; Figs. 6b, 6e). Confocal microscopy images of transmural cardiac sections from both knockout mice revealed near total attenuation of the NaV1.6 immuno-fluorescent signal (Fig. 6b). Further validation was undertaken through immuno-fluorescent staining of brain sections from these mice. NaV1.6 immuno-fluorescent signals were apparent in brain sections from the WT and cNaV1.6 KO mice, where they demonstrated a close association with β-tubulin (Figs. 6d, 6e). However, NaV1.6 immuno-fluorescent signals could not be observed in the global NaV1.6 KO brain tissue (Fig. 6f). These data are consistent with our antibody demonstrating specific reactivity to NaV1.6 in both the heart and brain tissue.
Fig. 6. Antibody specificity—KO models: (a–c) Confocal images of cardiac sections from the WT, cardiac-specific NaV1.6 KO (NaV1.6 cKO), and global NaV1.6 KO mice, labeled for NaV1.6 (red) and RyR2 (green). A clear NaV1.6 immunosignal was observed in WT but not NaV1.6 cKO or NaV1.6 KO. (d–f) Confocal images of brain sections labeled for NaV1.6 (red) and β-tubulin (green) show a clear NaV1.6 immunosignal in WT and NaV1.6 cKO but not NaV1.6 KO samples.
NaV1.6 Localization in Cardiac Myocytes
In order to evaluate NaV1.6 localization relative to Ca2+ cycling machinery, we performed a PLA to detect close approximation of NaV1.6 with RyR2 (Fig. 7). This assay generates a punctate signal at sites where co-labeled proteins are located within 40 nm of each other (Weibrecht et al., Reference Weibrecht, Leuchowius, Clausson, Conze, Jarvius, Howell, Kamali-Moghaddam and Soderberg2010). An overlay of a confocal image of NaV1.6-RyR2 PLA signal (green puncta) and DAPI nuclear stain (blue) with differential interference contrast showing myocyte topology (grayscale) is provided in Figure 7. This demonstrates the high frequency of close association between NaV1.6 and RyR2 with a striated distribution consistent with t-tubular localization. This concurs with previous studies suggesting colocalization of neuronal sodium channels, including NaV1.6, with RyR2 (Radwański et al., Reference Radwański, Ho, Veeraraghavan, Brunello, Liu, Belevych, Unudurthi, Makara, Priori, Volpe, Armoundas, Dillmann, Knollmann, Mohler, Hund and Györke2016).
Fig. 7. Proximity ligation assay. Confocal images of NaV1.6-RyR2 PLA signals (green) and nuclei (blue) overlaid on a differential interference contrast image of an isolated cardiac myocyte.
Lastly, we performed STORM on murine ventricular sections to quantitatively assess relative localization of NaV1.6 and RyR2 with sub-diffraction resolution. STORM affords 20 nm lateral and <50 nm axial resolution; however, it requires high-fidelity antibodies to achieve sufficient signal-to-noise ratios. A representative 3D STORM image in Figure 8a demonstrates both NaV1.6 (green) and RyR2 (red) molecules organizing into dense clusters, arrayed in a striated pattern. This is consistent with closely associated clusters of NaV1.6 and RyR2 localized along t-tubules. Results from the quantitative analysis of these data using STORM-RLA, a machine learning-based approach (Veeraraghavan & Gourdie, Reference Veeraraghavan and Gourdie2016), are presented in the form of a bivariate histogram in Figure 8b. Low-density NaV1.6 clusters were preferentially localized within 100 nm of RyR2 clusters but could also be identified at larger distances from RyR2. In contrast, high-density NaV1.6 clusters were exclusively localized within 50 nm of RyR2 clusters. Overall, 40 ± 4% of NaV1.6 clusters were located within 50 nm of RyR2 clusters with an additional 25 ± 2% located between 50 and 100 nm from RyR2 clusters. These data indicate a very close spatial approximation of NaV1.6 with RyR2 in cardiac muscle.
Fig. 8. Super-resolution STORM imaging. (a) Representative STORM image of RyR2-NaV1.6 from murine myocardium. The lateral resolution of 20 nm and axial resolution of <50 nm were achieved. Each localized molecule is represented as a 50 nm sphere for easy visualization. (b) Results from STORM-RLA (Veeraraghavan & Gourdie, Reference Veeraraghavan and Gourdie2016) quantification: bivariate histogram of NaV1.6 cluster density versus distance from the closest RyR2 cluster.
Discussion
Voltage-gated sodium channel isoforms, such as NaV1.6, which are expressed ubiquitously in the nervous system, are also present within the myocardium. Thus, they have been of interest in a wide range of physiological and pathophysiological contexts. The absence of quantitative, high-resolution data on their localization relative to Ca2+ handling machinery in cardiac myocytes, stemming from a lack of high-fidelity antibodies, has remained a significant barrier to understanding their functional roles. Here, we present results demonstrating the high avidity and specificity of a novel rabbit polyclonal antibody against the C-terminal epitope on NaV1.6. Further, we utilized this antibody to demonstrate enrichment of NaV1.6 in close proximity (<100 nm) of RyR2 Ca2+ release channels.
Our novel antibody was raised against a 12 amino acid sequence from the C-terminal domain of NaV1.6, which displays a high degree of conservation across multiple mammalian species while being unique relative to other sodium channel isoforms. Such a C-terminal epitope enables easy access for the antibody given its cytoplasmic location and lack of a highly ordered protein structure. Additionally, this sequence possesses favorable chemical properties when synthesized as a peptide, such as high-water solubility and lack of known binding motifs. Therefore, this region of NaV1.6 was selected as an ideal target for antibody development.
The antibody was validated based on four different criteria: signal-to-noise ratio, pattern of localization, specificity, and affinity. In confocal microscopy experiments, the raw antibody-containing serum demonstrated clear immunosignals with low background in both cellular and tissue preparations. These properties of the antibody were notably enhanced following affinity purification. These signals demonstrated a striated pattern of localization which coincided with ryanodine receptor calcium release channels (RyR2), consistent with previous studies of neuronal sodium channels in the heart (Maier et al., Reference Maier, Westenbroek, McCormick, Curtis, Scheuer and Catterall2004; Radwański & Poelzing, Reference Radwański and Poelzing2011; Radwański et al., Reference Radwański, Ho, Veeraraghavan, Brunello, Liu, Belevych, Unudurthi, Makara, Priori, Volpe, Armoundas, Dillmann, Knollmann, Mohler, Hund and Györke2016). These studies reported neuronal channels localizing to the transverse tubules of cardiac myocytes along with Ca2+ handling proteins such as RyR2. Thus, RyR2 was identified as a suitable protein to co-label in our antibody validation studies. Next, we validated the specificity and isoform selectivity of our antibody using multiple orthogonal approaches. The immunoreactivity of our antibody in both immuno-fluorescence and immunoblot studies was abrogated upon incubation with a peptide corresponding to its epitope target. In immunoblot studies, multiple bands were observed at and near the predicted weight of NaV1.6. This is consistent with previous observations for ion channel proteins, particularly sodium channels (van Bemmelen et al., Reference van Bemmelen, Rougier, Gavillet, Apotheloz, Daidie, Tateyama, Rivolta, Thomas, Kass, Staub and Abriel2004; Clatot et al., Reference Clatot, Hoshi, Wan, Liu, Jain, Shinlapawittayatorn, Marionneau, Ficker, Ha and Deschenes2017), where multiple bands were observed on Western blots, likely reflecting post-translational modifications. Additional experiments using heterologous expression systems and genetically modified mice with global or cardiac-specific knockout of NaV1.6 provided further evidence of specificity. Notably, the antibody displayed avidity and specificity in both the brain and heart tissue in these experiments, enhancing confidence in our results and indicating the broad utility of the antibody.
Although previous studies identified NaV1.6 at the t-tubules of cardiac myocytes, they lacked the resolution to determine whether NaV1.6 closely associated with Ca2+ handling proteins at the nanoscale. Armed with a high-affinity and -sensitivity antibody, we were able to assess the unique organization of NaV1.6 in cardiomyocytes at the nanoscale using PLA and STORM super-resolution microscopy. Punctate signals generated by PLA, corresponding to sites of the close association between NaV1.6 and RyR2, displayed noticeable alignment with t-tubules (identified from DIC). This was consistent with the striated pattern of NaV1.6 and RyR2 localization observed using STORM. The low background noise observed in the STORM images, enabled by the high fidelity of our antibody, allowed NaV1.6-RyR2 relative localization to be quantitatively assessed at the nanoscale. Machine learning-based cluster detection and STORM-RLA (Veeraraghavan & Gourdie, Reference Veeraraghavan and Gourdie2016) quantification revealed over 60% of NaV1.6 clusters are located within 100 nm of RyR2. To our knowledge, this is the first such quantitative assessment of NaV1.6 localization in cardiac muscle at sub-diffraction resolution. Importantly, these results identify NaV1.6 as a component of the Ca2+ cycling nanodomain, capable of influencing Ca2+ cycling via NCX. This supports a role for NaV1.6 in modulating normal Ca2+ cycling in health as well as in contributing to abnormal, arrhythmogenic Ca2+ release in disease.
In summary, we present a novel high-affinity and -specificity antibody, which may be used to dissect the role of NaV1.6 within cardiac nanodomains. We provide multiple lines of evidence demonstrating its selectivity and specificity in both the brain and the heart. Lastly, we utilized this high-fidelity antibody to quantitatively demonstrate for the first time that over half of the NaV1.6 channels in cardiac myocytes reside within 100 nm of ryanodine receptor Ca2+ release channels.
Supplementary material
To view supplementary material for this article, please visit https://doi.org/10.1017/S1431927619015289.
Acknowledgments
The authors thank the University of Utah Transgenic and Gene Targeting Core and the University of Utah Mutation Generation and Detection Core for the generation of cardiac-specific NaV1.6 knockout mice.
Financial support
The study was supported by the National Institutes of Health grants R01-HL074045, R01-HL063043, and R01-HL138579 awarded to S.G.; R00-HL127299 and an American Heart Association Transformational Project Award 19TPA34910191 to P.B.R.; and an American Heart Association Scientist Development Grant 16SDG29870007 awarded to R.V.
Conflict of interest
The authors declare that they have no competing interests.
