- Email: [email protected]
nuclear calcium handling
Phospholamban is concentrated in the nuclear envelope of cardiomyocytes and involved in perinuclear/nuclear calcium handling Adonis Z. Wu PhD, Dongzhu Xu MD, PhD, Na Yang MD, Shien-Fong Lin PhD, Peng-Sheng Chen MD, Steven E. Cala PhD, Zhenhui Chen PhD PII: DOI: Reference:
S0022-2828(16)30357-1 doi: 10.1016/j.yjmcc.2016.09.008 YJMCC 8456
To appear in:
Journal of Molecular and Cellular Cardiology
Received date: Revised date: Accepted date:
17 July 2016 26 August 2016 13 September 2016
Please cite this article as: Wu Adonis Z., Xu Dongzhu, Yang Na, Lin Shien-Fong, Chen Peng-Sheng, Cala Steven E., Chen Zhenhui, Phospholamban is concentrated in the nuclear envelope of cardiomyocytes and involved in perinuclear/nuclear calcium handling, Journal of Molecular and Cellular Cardiology (2016), doi: 10.1016/j.yjmcc.2016.09.008
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
SC
RI
PT
Phospholamban is concentrated in the nuclear envelope of cardiomyocytes and involved in perinuclear/nuclear calcium handling
NU
Adonis Z. Wu, PhD1,2, Dongzhu Xu, MD. PhD 1,3, Na Yang MD1,4, Shien-Fong Lin, PhD1,2, Peng-Sheng Chen, MD1, Steven E. Cala, PhD5, Zhenhui Chen, PhD1* 1
D
MA
Krannert Institute of Cardiology and Division of Cardiology, Department of Medicine, Indiana University School of Medicine, Indianapolis, IN, USA; 2National Chiao-Tung University, Hsinchu, Taiwan; 3Cardiovascular Division, Institute of Clinical Medicine, Faculty of Medicine, University of Tsukuba, Japan; 4Department of Gynecological and Obstetric ultrasound, First Affiliated Hospital of Harbin Medical University, Heilongjiang, P.R. China and 5Department of Physiology, Wayne State University, Detroit, MI
AC CE P
*Address correspondence to:
TE
Wu: PLB in nuclear envelope of cardiomyocytes
Zhenhui Chen, Ph.D.
Krannert Institute of Cardiology and Division of Cardiology Department of Medicine Indiana University School of Medicine 1800 N Capitol Ave, Indianapolis 46202, IN, USA. Tel: 1-317-274-0964 Fax: 1-317-962-0505 Email: [email protected]
1
ACCEPTED MANUSCRIPT Abstract
PT
Aims- Phospholamban (PLB) regulates the cardiac Ca2+-ATPase (SERCA2a) in sarcoplasmic reticulum (SR). However, the localization of PLB at subcellular sites outside the SR and possible contributions to Ca2+ cycling remain unknown. We examined the intracellular distribution of PLB and tested whether a pool of PLB exists in the nuclear envelope (NE) that might regulate perinuclear/nuclear Ca2+ (nCa2+) handling in cardiomyocytes (CMs).
MA
NU
SC
RI
Methods and Results- Using confocal immunofluorescence microscopy and immunoblot analyses of CMs and CM nuclei, we discovered that PLB was highly concentrated in NE. Moreover, the ratio of PLB levels to SERCA levels was greater in NE than in SR. The increased levels of PLB in NE was a consistent finding using a range of antibodies, tissue samples, and species. To address a possible role in affecting Ca2+ handling, we used Fluo-4 based confocal Ca2+ imaging, with scan-lines across cytosol and nuclei, and evaluated the effects of PLB on cytosolic and nCa2+ uptake and release in mouse CMs. In intact CMs, isoproterenol increased amplitude and decreased the decay time of Ca2+ transients not only in cytosol but also in nuclear regions. In saponin-permeabilized mouse CMs ([Ca2+]i =400nM), we measured spontaneous Ca2+ waves after specific reversal of PLB activity by addition of the Fab fragment of an antiPLB monoclonal antibody (100 μg/mL). This highly selective immunological reagent enhanced Ca2+ uptake (faster decay times) and Ca2+ release (greater intensity) in both cytosol and across the nuclear regions.
AC CE P
TE
D
Conclusions- Besides SR, PLB is concentrated in NE of CMs, and may be involved in modulation of nCa2+ dynamics.
Keywords: Phospholamban; calcium signaling; cardiomyocyte; nuclear membranes; sarcoplasmic reticulum Ca2+ ATPase ; perinuclear Ca2+ dynamics.
Abbreviations: CM, cardiomyocyte; Fab, the Fab fragment of the monoclonal anti-PLB antibody 2D12; IP3R,inositol 1,4,5-trisphosphate receptor; nCa2+, perinuclear/nuclear Ca2+; NE, nuclear envelope; NM, perinuclear/nuclear membranes; PLB, phospholamban; RyR2, ryanodine receptor; SCW, spontaneous Ca2+ wave;. SR, sarcoplasmic reticulum; SERCA2a, isoform of Ca2+-ATPase in cardiac SR.
2
ACCEPTED MANUSCRIPT 1. Introduction
RI
PT
Phospholamban (PLB) is a phosphoprotein regulator of the cardiac sarcoplasmic reticulum (SR) Ca 2+ATPase (SERCA2a isoform) and critically regulates the intracellular Ca2+ ([Ca2+]i) homeostasis in cardiomyocytes (CMs)[1-3]. While dephosphorylated PLB inhibits SERCA2a activity, phosphorylation of PLB by cAMP-dependent protein kinase or Ca2+/calmodulin-dependent protein kinase II reverses SERCA2a inhibition, thus increasing the Ca2+ uptake into luminal SR [4]. Due to its significant contribution to intracellular Ca2+ handling, PLB is as an important target for understanding CM function in physiological conditions and for the therapeutic approach to treat cardiac disease [3].
TE
D
MA
NU
SC
SR Ca2+ cycling plays an important role in excitation-contraction (E-C) coupling [5, 6]. Even in diastole, SR Ca2+ is released through the cardiac ryanodine receptor channel (RyR2) into cytosol as Ca2+ sparks or spontaneous Ca2+ waves (SCWs), which can trigger downstream physiological and pathophysiological events. Intracellular Ca2+ is recycled back into lumen of SR by SERCA2a, which is regulated by PLB. An early study in ventricular myocytes demonstrated that phosphorylation of PLB by isoproterenol addition or dissociation of PLB from SERCA2a by the PLB antibody, 2D12, not only stimulated the rate of Ca2+ uptake, but also enhanced Ca2+ release [7]. Studies using genetically engineered PLB mice further demonstrated that by controlling the rate of SERCA2a, PLB regulates SR Ca2+ content and consequent Ca2+ release through RyR2, thus modulating intracellular Ca2+ dynamics [8, 9]. We recently showed that the anti-PLB Fab fragment penetrates into saponin-permeabilized CMs and binds PLB in situ [10]. Addition of anti-PLB Fab significantly increased the intensity (F/F0), frequency, and velocity of SCWs, and decreased the decay time (DT50) of SCWs [10]. Therefore, PLB is a powerful modulator of both contractile function and rhythmic Ca2+ activity. Furthermore, anti-PLB Fab is a novel reagent to study PLB and other factors in regulation of intracellular Ca2+ dynamics. Ca2+ handling occurs not only in SR, but also in perinuclear/nuclear membranes (NM). Nuclear Ca2+ (nCa ) signaling plays an important role in regulation of cell growth, differentiation, survival, and death in various types of cells [11-13]. In CMs, nCa2+ stores inside NM share a common lumen with the SR through a continuous intracellular membrane system [14]. Cellular E-C coupling, a major function of SR, has little effect on nCa2+ stores, underlining the highly specialized nature of local nuclear Ca2+ control. Recently, Bers and co-workers suggested an excitation-transcription coupling mechanism, linking Ca2+ release from nCa2+ stores to gene regulation in CMs [15]. Such an important role for nCa2+ stores was further suggested by recent findings of Ljubojevic et al. [16], who found that heart failure in an animal model of pressure overload, or in failed human heart, led to altered levels of SERCA and IP3-dependent inositol 1,4,5-trisphosphate receptor (IP3R) in the NM, with structural and functional remodeling in nCa2+ stores. Yet, the nCa2+ release mechanism remains controversial. Perinuclear/nuclear Ca2+ release in the form of nuclear sparks/waves[17] occurs through IP3R [18-22], but this Ca2+ release might also involve RyR and CSQ [23, 24]. There are marked differences in the biophysical characteristics between SR and perinuclear Ca2+ transients, with the latter having lower amplitude and slower rise and decay times [17]. The detailed mechanisms for regulation of nCa2+ stores remains poorly understood, even as SERCAbased Ca2+ uptake is known to exist in the nuclear regions [16, 19, 20].
AC CE P
2+
Considering the fact that PLB plays a crucial role in the modulation of Ca2+ dynamics in SR, it seems reasonable to link PLB to regulation of the poorly understood SERCA-based nCa2+ stores. Besides free SR, PLB has been localized to a non-junctionally attached membrane compartment known as corbular SR [25], although the exact placement of this membrane subcompartment in the CM biosynthetic/secretory pathway remains uncertain [26]. A single study reported PLB present in NM in smooth muscle [27], but no data exists regarding whether PLB is present in the NM of CMs. Furthermore, to date, no study has examined PLB regulation of SERCA Ca2+ uptake into nCa2+ stores. Taking advantage of anti-PLB Fab as a specific tool to acutely reverse PLB inhibition in CMs, we investigated the expression of PLB in NM 3
ACCEPTED MANUSCRIPT
AC CE P
TE
D
MA
NU
SC
RI
PT
and its contribution to nCa2+ handling.
4
ACCEPTED MANUSCRIPT 2. Methods 2.1 Cardiomyocyte and preparation permeabilization
RI
PT
Use of animals in the study is approved by the IACUC of Indiana University School of Medicine and the Methodist Research Institute, Indianapolis, Indiana and conforms the NIH Guide for the care and use of laboratory animals. CM isolation from adult C57BL/6 mice followed protocols previously reported [10]. Rabbit hearts were homogenized and centrifuged through 2.15M sucrose to isolate SR and nuclei [18, 28].
SC
2.2 Antibodies
MA
NU
The 2D12 monoclonal antibody binds specifically to PLB residues 7-13, and polyclonal PLB9 (raised against 10-20 of PLB) recognizes all isoforms of PLB. Monoclonal antibodies 1F1 and 1G7 were raised against the first 10 residues of dog and mouse isoforms of PLB, respectively. Because of a one single amino acid difference between dog (Asp2) and mouse/rabbit (Glu2) isoform of PLB, 1F1 [29, 30] or 1G7 [10] is species-specific for detection of dog or mouse/rabbit isoform of PLB, respectively. The monoclonal antibody against SERCA2a, 2A7-A1 [29] was used in the study. Antibodies against sarcalumenin and lamin A/C were purchased from Life Technology (MA3-932) and Cell Signaling Technology (#2032), respectively.
TE
D
Confocal immunofluorescence microscopy on paraformaldehyde fixed heart sections, isolated CMs, or nuclei were performed as previously described [10]. In some experiments, protein-A affinity purified antibodies were covalently labeled with various Alexa-Fluor fluorescent dyes (Invitrogen).
AC CE P
Immunoblot analyses were performed as previously described [29, 30]. We used 5µg of SR microsomes and 15µg of nuclei for SDS-PAGE and transferring to nitrocellulose membranes. Antibodybound protein bands were visualized with 125I-protein A, followed by autoradiography and quantification (Personal Fx, BioRad). 2.3 Intracellular Ca2+ imaging and analysis Intracellular Ca2+ activities were imaged at room temperature with the Leica TCS SP8 LSCM inverted microscope fitted with a 40x 1.42 NA oil immersion objective. Ca2+ transients were generated with field stimulations in intact mouse CMs loaded with fluo-4-AM (Invitrogen). Normal Tyrode’s solution contained 1.8mM Ca2+. Spontaneous Ca2+ activity of saponin-permeabilized CMs was imaged using the Ca2+ indicator dye Fluo-4, as previous described [10]. Mock internal solution contains (in mmol/L): 100 potassium aspartate, 20 KCl, 5 KH2PO4, 5 MgATP, 10 phosphocreatine, 5 U/ml creatine phosphokinase, 10 HEPES, 0.5 EGTA, 1 MgCl2, 0.015 Fluo-4 (Invitrogen), and 8% w/v dextran (molecular weight 40,000), pH 7.2 (KOH). CaCl2 was added to make the free [Ca2+]i of 400nM (WebMaxC Extended (http://www.maxchelator.stanford.edu)). 2.4 Statistical analysis Results were expressed as mean ± SEM. The statistical significance was evaluated using either paired or unpaired t tests, and followed by post hoc analyses. A value of p<0.05 was considered a statistically significant difference.
5
ACCEPTED MANUSCRIPT 3. Results 3.1 Localization of PLB in the perinuclear/nuclear membranes of cardiomyocytes
NU
SC
RI
PT
Taking advantage of our well-characterized monoclonal antibodies [10, 29, 30], and using immunofluorescence confocal microscopy, we examined the intracellular distribution of PLB dog heart tissue sections. Fig. 1A shows dual-staining of dog heart tissue sections with protein A affinity-purified monoclonal antibody against PLB 2D12 (covalently linked to Alexa Fluor-488, 2D12-488) and the monoclonal antibody against SERCA2a 2A7-A1 (covalently linked to Alexa fluor-594, 2A7-A1-594). Both antibodies most heavily immunolabeled and co-stained SR along z-lines at a spacing of about 2μm [10]. Importantly, 2D12 generated a bright ring of fluorescence around myonuclei (Fig. 1A, right panels), suggesting a robust level of PLB localized to the NM. Nuclei from non CMs did not have any such staining pattern (Fig. 1A, overlap), confirming the specificity of the antibodies. Close examination of the nuclear region also showed immunoreactive staining of invaginations into the nucleus (Fig. 1A, magnified in the bottom panels), suggesting that PLB might extend from the nuclear envelope (NE) into bilayer invaginations [16, 31].
TE
D
MA
The fluorescence intensity profile of 2D12-488, as measured longitudinally across sarcomeres and one nucleus (Fig. 1A, arrow), was co-localized with SERCA (2A7-A1-594) in both SR and NM (Fig. 1B), as previously reported for SERCA [18]. Interestingly, whereas signal intensities for 2A7-A1 antibodies were similar for NM and SR, intensities of 2D12 signal were significantly stronger in NM than that in SR. Peak intensity ratios of NM to SR for PLB and SERCA were 1.8±0.5 (n=36 nuclei in 4 dogs) and 0.9±0.2, respectively (Fig. 1B). These indicate that there are more PLB in NM than that in SR.
AC CE P
We expanded our analysis by examining intracellular distributions of PLB in heart tissue sections from mice, rabbits, and humans. Dual-staining with 2D12 and 2A7-A1, labeled with various fluorochromes, showed the same pattern of PLB localization in SR and NM (Fig. 2A, left panels), and overlap with SERCA2a. In control experiments (Fig. 2A), incubation of anti-PLB 2D12 with PLB peptide (1-31) eliminated all 2D12 staining [10], suggesting that these staining patterns are specific for anti-PLB 2D12 binding to PLB. We repeated these experiments using multiple species-specific PLB antibodies. As shown in Fig. 2B, dog specie-specific monoclonal PLB antibody 1F1 detected PLB in dog cardiac SR and NM (Canine, 1F1), but not in mouse CMs. Similarly, when using a mouse specie-specific monoclonal PLB antibody, 1G7, we detected PLB in mouse SR and NM (1G7), but not in dog CMs. Moreover, the polyclonal antibody PLB9 (which reacts with all PLB sequences regardless of species) detected PLB expression in SR and NM in both dog and mouse (PLB9). Note again that ER or nuclei of non-CMs were not stained (Fig. 2B, canine), confirming highly specific binding of these antibodies to PLB in cardiac SR and NM. Negative control experiments, carried out in the absence of primary antibodies, protein A-488 alone, did not generate any immunostaining (Fig. 1B, Protein A-488). These results demonstrate that PLB is present in cardiac NM, but at levels higher than levels in SR. 3.2 Location of PLB in isolated cardiac nuclei To further confirm the data from immunofluorescence microscopy, we examined whether PLB could be detected in isolated cardiac nuclei. Isolated cardiac nuclei, prepared from fresh rabbit left ventricle tissue, produced a highly homogeneous preparation after pelleting through a sucrose barrier (Supplementary Fig. 1). The Amido Black-stained nitrocellulose blot showed distinct differences in levels of numerous protein bands in SR and nuclei (Fig. 3A, compare rSR vs rNu), confirming the separation of SR and nuclei. Importantly, both monoclonal PLB antibodies 2D12 and 1G7, as well as 6
ACCEPTED MANUSCRIPT
RI
PT
polyclonal antibody PLB9, specifically identified double bands with identical pattern to that of SR, which correspond to pentamers and monomers of PLB (Fig. 3B). Dog specie-specific PLB antibody 1F1 only recognized PLB in dog SR (dSR), but not in rabbit SR and isolated nuclei. Furthermore, using lamin A/C (LMNA) and sarcalumenin (SAR) as proteins markers for nuclei and SR, respectively, we confirmed the high purity of SR and nuclei in these samples (Fig. 3B). Finally, we compared the expression ratio of PLB to SERCA between SR and nuclei. Fig.3C showed that the PLB to SERCA expression ratio in nuclei was significantly higher (3.55±0.58, p<0.05) than that in SR (normalized, 1.00±0.08). The high expression ratio of PLB to SERCA in NE very strongly supports the quantitative and qualitative immunofluorescence data (Fig. 1B).
TE
D
MA
NU
SC
We also performed immunofluorescence studies on the isolated nuclei from rabbit hearts. In isolated nuclei, dual-staining with 2D12-594 and 2A7-A1-488 identified both PLB and SERCA2a in nuclear envelope (NE) of cardiac nuclei (Fig. 4A). In addition to staining of the NE, both PLB and SERCA appeared to invaginate into the nuclei (Fig. 4A, Magnified). In addition to DAPI signals, we co-stained nuclei with 2D12 and lamin A/C (LMNA). Fig. 4B shows co-localization of PLB and lamin A/C in nuclei of CMs. Nuclei from non-CMs (LMNA and DAPI positive) did not have any PLB or SERCA staining, again confirming the specificity of the antibodies. Furthermore, NE of these cardiac nuclei were stained strongly by the rabbit specie-specific 1G7 (Fig. 4C, 1G7), but not by dog specie-specific 1F1. Moreover, dual-staining shows only PLB9 staining, but not detectable expression of a specific SR protein marker, sarcalumenin (SAR), in cardiac nuclei. In control experiments, anti-sarcalumenin antibody did stained SR (data not shown), again confirming the purity of the nuclei. Finally, in control experiments, secondary antibody alone had no detectable staining. Combing these data from immunostaining and western blot analysis, we conclude that PLB is highly concentrated in NE of CMs. 3.3 Effect of isoproterenol on perinuclear/nuclear Ca transients.
AC CE P
The localization of PLB in NE led us to test the initial hypothesis that PLB regulates nuclear SERCA and thus nCa2+ handling. We performed fluorescence confocal Ca2+ imaging in isolated mouse CMs with scan-lines across the length of the cell in a medial plane that shows the full nuclear diameter (Fig. 5A, white line, top panels), which is identified by bright fluo-4 labeling. This approach has been successfully used in several studies to characterize biophysical properties of nCa2+ release [16, 23]. Ca2+ transients were generated by 1 Hz field stimulations. Compared with that seen in cytosol, Ca2+ transients across perinuclear/nuclear regions were characterized by higher basal fluorescence signals (F), smaller amplitude (F/F0), and slower rise (time to peak, TtP) and decay times (DT50) (Fig. 5B, compare Nu vs Cy traces), consistent with the biophysical properties of nCa2+ transients reported by multiple labs [16, 23]. Addition of isoproterenol (100nM), which ultimately leads to phosphorylation of PLB and increases SERCA Ca2+ uptake into SR, significantly increased F/F0 (Fig. 5C) and decreased the decay time (Fig. 5E, DT50) of the Ca2+ transients not only at cytoplasmic regions, but also at perinuclear/nuclear regions (Table 1). Interestingly, while causing no significant change in the TtP at cytoplasmic regions, addition of isoproterenol significantly decreased the TtP at perinuclear/nuclear regions (Fig. 5D, Table 1), suggesting critical differences between Ca2+ transients in cytoplasmic and perinuclear regions. The decrease in DT50 of the Ca2+ transients suggest that phosphorylation of PLB increased SERCA Ca2+ uptake into nuclear Ca2+ stores. 3.4 Effect of anti-PLB Fab on properties of perinuclear spontaneous Ca2+ waves. Isoproterenol treatment results in multiple downstream effects. Therefore, taking advantage of this recently reported novel reagent, the Fab fragment of 2D12 anti-PLB antibodies (anti-PLB-Fab) [10], we examined the specific reversal of PLB inhibition on perinuclear/nuclear Ca2+ uptake and release in CMs. 7
ACCEPTED MANUSCRIPT Anti-PLB-Fab, covalently linked to Alexa-594, most heavily immunolabeled SR [10] and NE of mouse CM (Fig. 6A). All staining was eliminated by co-incubation of a peptide of PLB residue 1-31 with antiPLB Fab-594 (Fig. 6B), suggesting that these staining patterns are specific for anti-PLB Fab binding to PLB, further supporting the specificity of anti-PLB-Fab binding to PLB in SR and NE.
AC CE P
TE
D
MA
NU
SC
RI
PT
We next performed fluorescence confocal Ca2+ imaging in saponin-permeabilized CMs with scanlines across the full nuclear diameter (Fig. 6B, white line, top panels), which is identified by bright fluo-4 labeling. At 400nM of [Ca2+]i, line-scan Ca2+ images revealed Ca2+ sparks and SCWs in the CM (Fig. 6B, Ctl). Compared to SCWs in the cytosol (SCWCy), SCWNu had relatively high basal fluorescence amplitudes, low fluorescence amplitudes (F/F0), and slow TtP and DT50 (Fig.6Bb,c compare Nu vs Cy trace, Table 1), consistent with that recorded in intact CMs and reports by others [16, 17, 23]. Addition of anti-PLB Fab (100 μg/mL), which acutely reversed PLB inhibition of SERCA, significantly increased the frequency of SCWCy, enhanced amplitude (ΔF/F0) and decreased DT50 but with no significant change in the TtP of SCWCy (Fig. 6B (d-f), Cy), confirming our recent work [10]. Importantly, after addition of antiPLB Fab, ΔF/F0 of SCWNu significantly increased, while both TtP and DT50 significantly decreased (Fig. 6B (d-f), Nu, p<.05, Table 1). In control experiments, the addition of boiled anti-PLB Fab, or the antiSERCA2a antibody 2A7-A1 did not affect SCWs (data not shown) [10]. These results suggest for the first time that acute reversal of PLB actions by anti-PLB Fab leads to increased SERCA-generated Ca2+ uptake from perinuclear/nuclear regions.
8
ACCEPTED MANUSCRIPT 4.
Discussion
PT
In this study, we showed that PLB is present at relatively high levels in nuclear membranes compared to SR, and exhibits a regulation of nuclear SERCA2a that reflects its possible role in regulation of nCa2+ uptake and release. 4.1 Localization of PLB to NE of CMs.
TE
D
MA
NU
SC
RI
Since PLB was originally localized to SR in adult CMs in the 1980s [25], little detailed characterization of subcellular localization of PLB have been reported. Ferguson et al. [27] described PLB present in nuclear membranes of smooth muscle cells, but PLB has never been localized to NE in heart tissue or CMs. Given the long history of PLB studies, we were cautious in this study, employing multiple approaches to ensure the specific detection of PLB expression in NE in CMs. We first analyzed tissues, cells, and isolated nuclei using multiple extensively characterized, highly-specific PLB monoclonal and polyclonal antibodies [10, 29, 30]. Secondly, our approach used protein-A affinity purified antibodies covalently labeled with fluorescent dyes, allowing us to perform dual staining of PLB and SERCA, each potentially present in NE. Third, we detected different PLB isoforms with the use of species-specific PLB antibodies 1F1 and 1G7, providing proper controls for each. Fourth, immunoblot analysis showed an identical pattern of PLB5/PLB1 in isolated nuclei and in SR. Considering the fact that these PLB antibodies recognize PLB residues 1-10 (1G7, 1F1), 7-13 (2D12), and 10-20 (PLB9), we believed that the signals in nuclei were form PLB, not a new PLB-homologous protein. Finally, because it is impractical to compare quantitatively the amount of PLB between pure SR membranes and whole nuclei, we showed that higher expression ratio of PLB to SERCA in nuclei than that in SR. Taken together, these approaches yielded a large consistent set of data, supporting the conclusion that PLB is expressed at elevated levels in NE of CMs.
AC CE P
Free SR proteins and junctional SR proteins (e.g., RyR2) have different intracellular distributions, suggesting that divergent trafficking pathways may exist for the proteins resident to the two known sets of SR subcompartments. Recently, Sleiman et al [26] proposed a basic model of junctional SR (jSR) biogenesis and protein trafficking from rough ER to jSR. They showed that jSR resident proteins, including calsequestrin, triadin, junctin, (parts of the RyR2 complex which carry out Ca2+ release), may sort at translocons in NM, from whence a posttranslational trafficking pathway commences along a transverse/radial, microtubule-dependent pathway. These authors were also able to distinguish a second longitudinal ER subcompartment (referred to as smooth SR/free SR). Whereas junctional transmembrane proteins junctin and triadin transition after NE translation to a transverse pathway towards junctional SR sites, they have hypothesized that resident free SR proteins might follow a different, and as yet unknown route to fill free SR. The data shown in this study suggest that PLB may also be synthesized in perinuclear rough ER, where it can undergo a relative NE retention and accumulation, leading to a steadystate perinuclear concentration. Although the mechanism for this retention will require further studies, the accumulation of the proteins used in Ca2+ sequestration into perinuclear SR subcompartments suggests that new cell biological mechanisms might underlie physiological control of homeostasis between contractile and transcriptional subdomains. 4.2. PLB and possible regulation of nCa2+ stores Molecular mechanisms by which the nCa2+ stores are regulated remain unknown. In the present study, we provide initial evidence that PLB can play a role in regulation of nCa2+ dynamics in CMs. We used our recently reported specific effect of anti-PLB Fab on properties of SCWCy as a positive control in the study, and showed significantly increased Ca2+ uptake (reduction in DT50) and release (increase in ΔF/F0 and frequency) in SCWNu. Although NM and SR form a contiguous lumenal Ca2+ reservoir [14], there are 9
ACCEPTED MANUSCRIPT
SC
RI
PT
differences between biophysical properties of SCWCy and SCWNu, (e. g., small F/F0, slow TtP and DT50), consistent with the idea that local Ca2+ signaling is highly specialized and differentially modulated [16, 17], Interestingly, reversal of PLB inhibition resulted in similar changes in SCW Cy and SCWNu in terms of F/F0 and t1/2, but significant differences in TtP. These differences between SCWCy and SCWNu further suggested that the Ca2+ uptake and release activities we measured were generated in perinuclear/nuclear regions and were regulated by PLB. We speculate that the significant decrease in TtP might reflect differences in density of T-tubules and in geometry between lumens of SR and NM, the latter is generally thought to have a much smaller lumen than SR. Whether higher levels of intralumenal Ca 2+ within NM would cause increased Ca2+ release rate has not been determined, to our knowledge. Based on these results, we propose that PLB is a unique regulator for nCa2+ signaling, offering a potential missing component to an understanding of nCa2+ store regulation. 4.3 PLB and heart disease.
AC CE P
TE
D
MA
NU
Several PLB mutations have been identified that can cause human heart failure [32-34]. Furthermore, several animal models of disease that result from PLB mutations cannot be explained solely by dysfunction of SR Ca2+ dynamics. For example, a naturally occurring PLB mutant that exhibits a lethal cardiomyopathy, R14Del-PLB, was recently found to exhibit “an abnormal cytoplasmic distribution and perinuclear accumulation [of PLB]” in a patient’s heart (Fig. 1b of ref.[35]). In an earlier study, R14DelPLB was shown to be mis-routed to interact with the Na+,K+-pump in the plasma membranes [36]. These results highlight the complexity and impact of trafficking and distribution of PLB/SERCA2a in perinuclear and NE, yet mechanisms remain very poorly defined. Furthermore, although similar to PLBKO in terms of global SR Ca2+ handling, over-expression of an R14Del-PLB mutation on top of a PLBKO background caused cardiac hypertrophy [36]. Moreover, several mouse models, e.g., over-expression of gain-of-function PLB mutations [37], can generate a severe cardiac hypertrophy by an uncertain molecular mechanism. Louch et al [38] showed that SERCA2b knock-in in SERCA2a-KO/PLB-KO mice normalized global Ca2+ dynamics, but also caused cardiac hypertrophy. They suggested that global SR Ca2+ handling is not the primary cause of hypertrophic signaling. It should be noted that because the regulating role of PLB nCa2+ stores is unknown, none of these studies examined nCa2+ signaling, which is known to contribute significantly to activation of cardiac hypertrophic signaling [39, 40]. Our findings that PLB regulates nCa2+ handling may provide an important mechanistic link between alteration of PLB function and cardiac disease. The localization and role of PLB in NE in disease conditions need to be studied in the future. 4.4 Limitations and future directions. In the present study, the resolution of PLB detection in NE is limited by the use of diffractionlimited light microscopy. Although these light microscopic studies shown here provide an interpretive template for such studies, electron microscopy and super-resolution imaging systems would provide better optical resolution and aid in establishing the precise intracellular sites of PLB concentration. Preliminary super-resolution data using STED imaging (T. Kohl, Z. Chen, and S. Lehnart, manuscript in preparation) indicates that PLB is localized to the NE as opposed to alternate perinuclear membranes. The concentration and calibration differences of fluo-4 based Ca2+ indicators may contribute to the difference in basal fluorescence intensity in SR and in the nuclei. Due to these issues, it is possible that SCWNu is contaminated with SCWCy. Due to the resolution of Ca2+ imaging, we cannot distinguish the Ca2+ activity in perinuclear membranes and inside the nucleus. Although we showed the highly specific the anti-PLB Fab effect on SCWNu, future testing of Ca2+ release directly in isolated cardiac nuclei might add clarity to our findings. Finally, while the current study explored the logical hypothesis that PLB in NE regulates nuclear SERCA, our finding of higher concentration of PLB in NE than in SR hints at the
10
ACCEPTED MANUSCRIPT possibility that PLB may functionally interact with other nuclear proteins. This possibility will be explored in further studies.
RI
Source of Funding
PT
In summary, we have provided compelling new evidence that PLB is present at high levels in the NE of CMs where it can contribute to the modulation of nCa2+ handling. These findings should lead to further studies into PLB biology, PLB effects on nCa2+ signaling in CMs and its possible contribution to nCa2+mediated transcriptional regulation.
NU
SC
This study was supported in part by NIH Grants P01 HL78931, R0171140, a Medtronic-Zipes Endowment (PSC), the Indiana University Health-Indiana University School of Medicine Strategic Research Initiative (PSC) and the Dr. Charles Fisch Cardiovascular Research Award endowed by Dr. Suzanne B. Knoebel of the Krannert Institute of Cardiology (ZC).
MA
Acknowledgments
Disclosures
AC CE P
None.
TE
D
We thank Glen A. Schmeisser, Jian Tan, Xiao Shi (Michelle) and Jin Guo for great technical supports. We also thank Dechun Yin for helpful suggestions.
11
ACCEPTED MANUSCRIPT References
AC CE P
TE
D
MA
NU
SC
RI
PT
[1] Simmerman HK, Jones LR. Phospholamban: protein structure, mechanism of action, and role in cardiac function. Physiol Rev. 1998;78:921-47. [2] MacLennan DH, Kranias EG. Phospholamban: a crucial regulator of cardiac contractility. Nat Rev Mol Cell Biol. 2003;4:566-77. [3] Kranias EG, Hajjar RJ. Modulation of cardiac contractility by the phospholamban/SERCA2a regulatome. Circ Res. 2012;110:1646-60. [4] Lindemann JP, Jones LR, Hathaway DR, Henry BG, Watanabe AM. beta-Adrenergic stimulation of phospholamban phosphorylation and Ca2+-ATPase activity in guinea pig ventricles. J Biol Chem. 1983;258:464-71. [5] Bers DM. Cardiac excitation-contraction coupling. Nature. 2002;415:198-205. [6] Eisner D, Bode E, Venetucci L, Trafford A. Calcium flux balance in the heart. J Mol Cell Cardiol. 2013;58:110-7. [7] Sham JS, Jones LR, Morad M. Phospholamban mediates the beta-adrenergic-enhanced Ca2+ uptake in mammalian ventricular myocytes. Am J Physiol. 1991;261:H1344-9. [8] Santana LF, Kranias EG, Lederer WJ. Calcium sparks and excitation-contraction coupling in phospholamban-deficient mouse ventricular myocytes. J Physiol. 1997;503 ( Pt 1):21-9. [9] Bai Y, Jones PP, Guo J, Zhong X, Clark RB, Zhou Q, et al. Phospholamban knockout breaks arrhythmogenic Ca2+ waves and suppresses catecholaminergic polymorphic ventricular tachycardia in mice. Circ Res. 2013;113:517-26. [10] Chan YH, Tsai WC, Song Z, Ko CY, Qu Z, Weiss JN, et al. Acute reversal of phospholamban inhibition facilitates the rhythmic whole-cell propagating calcium waves in isolated ventricular myocytes. J Mol Cell Cardiol. 2015;80C:126-35. [11] Stehno-Bittel L, Perez-Terzic C, Clapham DE. Diffusion across the nuclear envelope inhibited by depletion of the nuclear Ca2+ store. Science. 1995;270:1835-8. [12] Hardingham GE, Chawla S, Johnson CM, Bading H. Distinct functions of nuclear and cytoplasmic calcium in the control of gene expression. Nature. 1997;385:260-5. [13] Alonso MT, Garcia-Sancho J. Nuclear Ca2+ signalling. Cell calcium. 2011;49:280-9. [14] Wu X, Bers DM. Sarcoplasmic reticulum and nuclear envelope are one highly interconnected Ca2+ store throughout cardiac myocyte. Circ Res. 2006;99:283-91. [15] Wu X, Zhang T, Bossuyt J, Li X, McKinsey TA, Dedman JR, et al. Local InsP3-dependent perinuclear Ca2+ signaling in cardiac myocyte excitation-transcription coupling. J Clin Invest. 2006;116:675-82. [16] Ljubojevic S, Radulovic S, Leitinger G, Sedej S, Sacherer M, Holzer M, et al. Early remodeling of perinuclear Ca2+ stores and nucleoplasmic Ca2+ signaling during the development of hypertrophy and heart failure. Circulation. 2014;130:244-55. [17] Luo D, Yang D, Lan X, Li K, Li X, Chen J, et al. Nuclear Ca2+ sparks and waves mediated by inositol 1,4,5-trisphosphate receptors in neonatal rat cardiomyocytes. Cell calcium. 2008;43:165-74. [18] Bare DJ, Kettlun CS, Liang M, Bers DM, Mignery GA. Cardiac type 2 inositol 1,4,5-trisphosphate receptor: interaction and modulation by calcium/calmodulin-dependent protein kinase II. J Biol Chem. 2005;280:15912-20. [19] Gerasimenko OV, Gerasimenko JV, Tepikin AV, Petersen OH. ATP-dependent accumulation and inositol trisphosphate- or cyclic ADP-ribose-mediated release of Ca2+ from the nuclear envelope. Cell. 1995;80:439-44. [20] Humbert JP, Matter N, Artault JC, Koppler P, Malviya AN. Inositol 1,4,5-trisphosphate receptor is located to the inner nuclear membrane vindicating regulation of nuclear calcium signaling by inositol 1,4,5-trisphosphate. Discrete distribution of inositol phosphate receptors to inner and outer nuclear membranes. J Biol Chem. 1996;271:478-85. [21] Zima AV, Bare DJ, Mignery GA, Blatter LA. IP3-dependent nuclear Ca2+ signalling in the mammalian heart. J Physiol. 2007;584:601-11. 12
ACCEPTED MANUSCRIPT
AC CE P
TE
D
MA
NU
SC
RI
PT
[22] Zima AV, Blatter LA. Inositol-1,4,5-trisphosphate-dependent Ca2+ signalling in cat atrial excitationcontraction coupling and arrhythmias. J Physiol. 2004;555:607-15. [23] Guo A, Cala SE, Song LS. Calsequestrin accumulation in rough endoplasmic reticulum promotes perinuclear Ca2+ release. J Biol Chem. 2012;287:16670-80. [24] Escobar M, Cardenas C, Colavita K, Petrenko NB, Franzini-Armstrong C. Structural evidence for perinuclear calcium microdomains in cardiac myocytes. J Mol Cell Cardiol. 2011;50:451-9. [25] Jorgensen AO, Jones LR. Immunoelectron microscopical localization of phospholamban in adult canine ventricular muscle. J Cell Biol. 1987;104:1343-52. [26] Sleiman NH, McFarland TP, Jones LR, Cala SE. Transitions of protein traffic from cardiac ER to junctional SR. J Mol Cell Cardiol. 2015;81C:34-45. [27] Ferguson DG, Young EF, Raeymaekers L, Kranias EG. Localization of phospholamban in smooth muscle using immunogold electron microscopy. J Cell Biol. 1988;107:555-62. [28] Tadevosyan A, Allen BG, Nattel S. Isolation and study of cardiac nuclei from canine myocardium and adult ventricular myocytes. Methods in molecular biology. 2015;1234:69-80. [29] Chen Z, Akin BL, Jones LR. Mechanism of reversal of phospholamban inhibition of the cardiac Ca2+-ATPase by protein kinase A and by anti-phospholamban monoclonal antibody 2D12. J Biol Chem. 2007;282:20968-76. [30] Chen Z. Competitive displacement of wild-type phospholamban from the Ca-free cardiac calcium pump by phospholamban mutants with different binding affinities. J Mol Cell Cardiol. 2014;76c:130-7. [31] Fricker M, Hollinshead M, White N, Vaux D. Interphase nuclei of many mammalian cell types contain deep, dynamic, tubular membrane-bound invaginations of the nuclear envelope. The Journal of cell biology. 1997;136:531-44. [32] Haghighi K, Kolokathis F, Gramolini AO, Waggoner JR, Pater L, Lynch RA, et al. A mutation in the human phospholamban gene, deleting arginine 14, results in lethal, hereditary cardiomyopathy. Proc Natl Acad Sci USA. 2006;103:1388-93. [33] Haghighi K, Kolokathis F, Pater L, Lynch RA, Asahi M, Gramolini AO, et al. Human phospholamban null results in lethal dilated cardiomyopathy revealing a critical difference between mouse and human. J Clin Invest. 2003;111:869-76. [34] Schmitt JP, Kamisago M, Asahi M, Li GH, Ahmad F, Mende U, et al. Dilated cardiomyopathy and heart failure caused by a mutation in phospholamban. Science. 2003;299:1410-3. [35] Karakikes I, Stillitano F, Nonnenmacher M, Tzimas C, Sanoudou D, Termglinchan V, et al. Correction of human phospholamban R14del mutation associated with cardiomyopathy using targeted nucleases and combination therapy. Nat Commun. 2015;6:6955. [36] Haghighi K, Pritchard T, Bossuyt J, Waggoner JR, Yuan Q, Fan GC, et al. The human phospholamban Arg14-deletion mutant localizes to plasma membrane and interacts with the Na/KATPase. J Mol Cell Cardiol. 2012;52:773-82. [37] Haghighi K, Schmidt AG, Hoit BD, Brittsan AG, Yatani A, Lester JW, et al. Superinhibition of sarcoplasmic reticulum function by phospholamban induces cardiac contractile failure. J Biol Chem. 2001;276:24145-52. [38] Louch WE, Vangheluwe P, Bito V, Raeymaekers L, Wuytack F, Sipido KR. Phospholamban ablation in hearts expressing the high affinity SERCA2b isoform normalizes global Ca2+ homeostasis but not Ca2+-dependent hypertrophic signaling. Am J Physiol Heart Circ Physiol. 2012;302:H2574-82. [39] Nakayama H, Bodi I, Maillet M, DeSantiago J, Domeier TL, Mikoshiba K, et al. The IP3 receptor regulates cardiac hypertrophy in response to select stimuli. Circ Res. 2010;107:659-66. [40] Houser SR, Molkentin JD. Does contractile Ca2+ control calcineurin-NFAT signaling and pathological hypertrophy in cardiac myocytes? Science signaling. 2008;1:pe31.
13
ACCEPTED MANUSCRIPT
RI
PT
Table 1 Characteristics of spontaneous Ca2+ wave (SCW) initiation Ca2+ transients were measured in intact mouse CMs before (Ctl) and after addition of 100nM isoproterenol (ISO) as indicated in Figure 5. Spontaneous Ca2+ waves SCWs were measured in permeabilized mouse CMs before (Ctl) and after addition of anti-PLB Fab (Fab) at [Ca2+]i of 400nM as indicated in Figure 6. The characteristics of Ca2+ transients or SCWs at the cytoplasmic (Cy) and perinuclear (Nu) regions, including F/F0 , DT50, and time-to-peak (TtP) were measured. Data are means ± SEM. (n>9 in at least 4 mice; * indicates p<0.05 vs control by paired t test). Permeabilized Cytosol
ISO
ΔF/F0
3.0±0.2
4.2±0.4*
2.7±0.2
3.7±0.3*
DT50 (ms)
115±8
94±7*
162±9
120±7*
TtP
61.9±5.4
57.4±5.5
96.9±6.3
70.0±6.6*
perinuclear
Ctl
Fab
1.8±0.1
2.6±0.1*
1.0±0.1
1.4±0.1*
82±4
60±3*
151±8
110±6*
71±4
65±2
124±8
84±4*
Ctl
Fab
AC CE P
TE
D
MA
NU
Ctl
Perinuclear/ nuclear Ctl ISO
SC
Intact Cytosol
14
ACCEPTED MANUSCRIPT
SC
RI
PT
Figure 1. Intracellular distribution of PLB in canine cardiomyocytes. Confocal immunofluorescence microscopy images were obtained from canine heart tissue sections. A. Dual-staining images with the monoclonal anti-PLB antibody 2D12 (conjugated with Alexa fluor-488, left panels) and 2A7-A1 (conjugated with Alexa fluor-594, middle panels). Merged images are in the right panels. The regions indicated in the boxes were refocused to show magnified images in the bottom panels. Nuclei were stained with DAPI in blue in all images. B. Intensity profiles of sarcomeres and a nucleus. Region pointed by an arrow in A was analyzed (purple box). Graph shows intensity ratio between NM and SR for PLB and SERCA. * indicates p< 0.05. Bar represents 10 µm. Similar patterns of staining were obtained in at least 10 samples.
AC CE P
TE
D
MA
NU
Figure 2. Species-specific detection of PLB in cardiomyocytes from various species. Confocal immunofluorescence microscopy images were obtained from mouse CMs, rabbit and human heart tissue sections. A. Dual-staining with 2D12-488 and 2A7-A1-594 for mouse CMs, and dual-staining with 2D12594 and 2A7-A1-488 for rabbit and human heart tissue sections. Immunostainings with co-incubation of 2D12 and a PLB peptide 1-31 are shown in the right panels. B. Immunostaining of PLB using speciesspecific anti-PLB antibodies (Abs) in mouse CMs (upper panels) and dog heart tissue sections (lower panels). Abs used were: dog specie-specific PLB monoclonal Ab 1F1, mouse specie-specific PLB monoclonal Ab 1G7, and polyclonal antibody PLB9. Protein A-488 was used to visualize antibody binding signals. Immunostainings with protein A-488 alone but no primary antibody incubation are in right panels. Bar represents 10 µm. Similar patterns of staining were obtained in at least 10 samples in each condition.
Figure 3. Immunoblot analysis of PLB protein levels in isolated cardiac nuclei. 5µg rabbit cardiac SR (rSR) and 15µg nuclei (rNu) were subjected to SDS-PAGE. A. nitrocellulose stained with Amido Black, B. autoradiograms after incubation with antibodies followed by 125I-proteinA. Identical blots were stained with various antibodies (indicated at the bottom) against PLB (2D12, 1G7, PLB9, and 1F1 (canine SR was added as a positive control)), against sarcalumenin (SAR), against lamin A/C (LMNA), and against SERCA (2A7-A1). C. The immunoblot and plot showing the expression ratio of PLB to SERCA. PLB/SERCA in SR is normalized to 1 (n=5, * indicates p<0.05). PLB1 and PLB5, PLB monomer and pentamer, respectively.
Figure 4. Immunofluorescence analysis of PLB in isolated cardiac nuclei. Confocal immunofluorescence microscopy images were obtained from isolated rabbit cardiac nuclei. A. dualstaining with 2D12, conjugated with Alexa fluor-594, and 2A7-A1, conjugated with Alexa fluor-488. B. dual-staining with 2D12 and polyclonal lamin A/C antibodies (Ab), visualized by goat anti-mouse and goat anti-rabbit secondary Abs. Merged and magnified images (indicated by boxes) are shown in two right panels. C. Ab used (From left to right): rabbit and dog species-specific PLB monoclonal Ab 1G7 and 1F1, respectively, visualized with protein A-488; dual-staining with PLB9 and sarcalumenin Ab (SAR); no primary Ab with secondary Abs alone. Blue is DAPI in all images. Bar represents 10 µm. Similar patterns of staining were obtained in at least 10 samples in each condition.
15
ACCEPTED MANUSCRIPT
RI
PT
Figure 5. The effect of isoproterenol on cytoplasmic and perinuclear/nuclear Ca2+ transients in intact mouse CMs. Intact CMs were in Tyrode’s solution with 1.8mM Ca2+. A. Representative confocal line-scan Ca2+ images using Fluo-4-AM were obtained in the same CM before (Ctl) and after addition of 100 nM isoproterenol (ISO). Nucleus is between red lines. Scan-line (dashed white) is over cytosol and nucleus. B. Traces show intensities of fluorescent signals (F) of Ca2+ transients at cytoplasmic and perinuclear/nuclear regions. C-E. Bar graphs show ensemble average cell parameters for all measurements (C, intensity, ΔF/F0; D, time to peak, TtP; E, 50% decay time, DT50) of perinuclear/nuclear (Nu) and cytoplasmic (Cy) regions before and after addition of isoproterenol. See also Table 1 for data. * indicates p<0.05 vs control (average of at least 10 CMs in 6 mice).
AC CE P
TE
D
MA
NU
SC
Figure 6. Effects of anti-PLB Fab on spontaneous Ca2+ waves (SCWs) in cytoplasmic and perinuclear regions of CMs. A. Mouse CM stained by anti-PLB Fab, conjugated with Alexa 594, nuclei (DAPI) are in blue in all images. The right panel shows data from co-incubation of anti-PLB Fab-594 and a PLB peptide 1-31. B. representative confocal line-scan Ca2+ images using Fluo-4 Ca2+ indicator were obtained in the same permeabilized mouse CM (top panels) before (Ctl) or after (Fab) addition of 100 µg/ml anti-PLB Fab. The nucleus is between the red lines. Scan-line (dashed white) is over cytosol and nucleus. Ca2+ concentration was 400 nM. b. Traces show intensities of fluorescent signals (F) of SCWs at cytoplasmic and perinuclear/nuclear regions. c. ensemble averages of traces. d-f. Bar graphs show ensemble parameters (d, intensity, ΔF/F0; e, time to peak, TtP; f, 50% decay time, DT50) of perinuclear/nuclear (Nu) and cytoplasmic (Cy) before and after addition of anti-PLB Fab. See also Table 1 for data. * indicates p<0.05 vs control (average of at least 9 CMs in 4 mice).
16
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
AC CE P
TE
D
MA
Figure 1
17
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
AC CE P
TE
D
Figure 2
18
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
AC CE P
TE
D
MA
Figure 3
19
AC CE P
Figure 4
TE
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
20
Figure 5
AC CE P
TE
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
21
Figure 6
AC CE P
TE
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
22
PT
ACCEPTED MANUSCRIPT
AC CE P
TE
D
MA
NU
SC
RI
Graphical abstract
23
ACCEPTED MANUSCRIPT Highlights: We tested if phospholamban (PLB) is in nuclear envelope (NE) of cardiomyocytes (CMs).
PT
Multiple PLB antibodies stained CM perinuclear/nuclear membranes in several species.
RI
Immunoblot/fluorescence assays showed high levels of PLB in NE of isolated CM nuclei.
SC
The Fab fragment of PLB antibody increased perinuclear/nuclear Ca uptake and release.
AC CE P
TE
D
MA
NU
PLB was localized in NE in high concentration and involved in CM nuclear Ca handling.
24
S0022-2828(16)30357-1 doi: 10.1016/j.yjmcc.2016.09.008 YJMCC 8456
To appear in:
Journal of Molecular and Cellular Cardiology
Received date: Revised date: Accepted date:
17 July 2016 26 August 2016 13 September 2016
Please cite this article as: Wu Adonis Z., Xu Dongzhu, Yang Na, Lin Shien-Fong, Chen Peng-Sheng, Cala Steven E., Chen Zhenhui, Phospholamban is concentrated in the nuclear envelope of cardiomyocytes and involved in perinuclear/nuclear calcium handling, Journal of Molecular and Cellular Cardiology (2016), doi: 10.1016/j.yjmcc.2016.09.008
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT
SC
RI
PT
Phospholamban is concentrated in the nuclear envelope of cardiomyocytes and involved in perinuclear/nuclear calcium handling
NU
Adonis Z. Wu, PhD1,2, Dongzhu Xu, MD. PhD 1,3, Na Yang MD1,4, Shien-Fong Lin, PhD1,2, Peng-Sheng Chen, MD1, Steven E. Cala, PhD5, Zhenhui Chen, PhD1* 1
D
MA
Krannert Institute of Cardiology and Division of Cardiology, Department of Medicine, Indiana University School of Medicine, Indianapolis, IN, USA; 2National Chiao-Tung University, Hsinchu, Taiwan; 3Cardiovascular Division, Institute of Clinical Medicine, Faculty of Medicine, University of Tsukuba, Japan; 4Department of Gynecological and Obstetric ultrasound, First Affiliated Hospital of Harbin Medical University, Heilongjiang, P.R. China and 5Department of Physiology, Wayne State University, Detroit, MI
AC CE P
*Address correspondence to:
TE
Wu: PLB in nuclear envelope of cardiomyocytes
Zhenhui Chen, Ph.D.
Krannert Institute of Cardiology and Division of Cardiology Department of Medicine Indiana University School of Medicine 1800 N Capitol Ave, Indianapolis 46202, IN, USA. Tel: 1-317-274-0964 Fax: 1-317-962-0505 Email: [email protected]
1
ACCEPTED MANUSCRIPT Abstract
PT
Aims- Phospholamban (PLB) regulates the cardiac Ca2+-ATPase (SERCA2a) in sarcoplasmic reticulum (SR). However, the localization of PLB at subcellular sites outside the SR and possible contributions to Ca2+ cycling remain unknown. We examined the intracellular distribution of PLB and tested whether a pool of PLB exists in the nuclear envelope (NE) that might regulate perinuclear/nuclear Ca2+ (nCa2+) handling in cardiomyocytes (CMs).
MA
NU
SC
RI
Methods and Results- Using confocal immunofluorescence microscopy and immunoblot analyses of CMs and CM nuclei, we discovered that PLB was highly concentrated in NE. Moreover, the ratio of PLB levels to SERCA levels was greater in NE than in SR. The increased levels of PLB in NE was a consistent finding using a range of antibodies, tissue samples, and species. To address a possible role in affecting Ca2+ handling, we used Fluo-4 based confocal Ca2+ imaging, with scan-lines across cytosol and nuclei, and evaluated the effects of PLB on cytosolic and nCa2+ uptake and release in mouse CMs. In intact CMs, isoproterenol increased amplitude and decreased the decay time of Ca2+ transients not only in cytosol but also in nuclear regions. In saponin-permeabilized mouse CMs ([Ca2+]i =400nM), we measured spontaneous Ca2+ waves after specific reversal of PLB activity by addition of the Fab fragment of an antiPLB monoclonal antibody (100 μg/mL). This highly selective immunological reagent enhanced Ca2+ uptake (faster decay times) and Ca2+ release (greater intensity) in both cytosol and across the nuclear regions.
AC CE P
TE
D
Conclusions- Besides SR, PLB is concentrated in NE of CMs, and may be involved in modulation of nCa2+ dynamics.
Keywords: Phospholamban; calcium signaling; cardiomyocyte; nuclear membranes; sarcoplasmic reticulum Ca2+ ATPase ; perinuclear Ca2+ dynamics.
Abbreviations: CM, cardiomyocyte; Fab, the Fab fragment of the monoclonal anti-PLB antibody 2D12; IP3R,inositol 1,4,5-trisphosphate receptor; nCa2+, perinuclear/nuclear Ca2+; NE, nuclear envelope; NM, perinuclear/nuclear membranes; PLB, phospholamban; RyR2, ryanodine receptor; SCW, spontaneous Ca2+ wave;. SR, sarcoplasmic reticulum; SERCA2a, isoform of Ca2+-ATPase in cardiac SR.
2
ACCEPTED MANUSCRIPT 1. Introduction
RI
PT
Phospholamban (PLB) is a phosphoprotein regulator of the cardiac sarcoplasmic reticulum (SR) Ca 2+ATPase (SERCA2a isoform) and critically regulates the intracellular Ca2+ ([Ca2+]i) homeostasis in cardiomyocytes (CMs)[1-3]. While dephosphorylated PLB inhibits SERCA2a activity, phosphorylation of PLB by cAMP-dependent protein kinase or Ca2+/calmodulin-dependent protein kinase II reverses SERCA2a inhibition, thus increasing the Ca2+ uptake into luminal SR [4]. Due to its significant contribution to intracellular Ca2+ handling, PLB is as an important target for understanding CM function in physiological conditions and for the therapeutic approach to treat cardiac disease [3].
TE
D
MA
NU
SC
SR Ca2+ cycling plays an important role in excitation-contraction (E-C) coupling [5, 6]. Even in diastole, SR Ca2+ is released through the cardiac ryanodine receptor channel (RyR2) into cytosol as Ca2+ sparks or spontaneous Ca2+ waves (SCWs), which can trigger downstream physiological and pathophysiological events. Intracellular Ca2+ is recycled back into lumen of SR by SERCA2a, which is regulated by PLB. An early study in ventricular myocytes demonstrated that phosphorylation of PLB by isoproterenol addition or dissociation of PLB from SERCA2a by the PLB antibody, 2D12, not only stimulated the rate of Ca2+ uptake, but also enhanced Ca2+ release [7]. Studies using genetically engineered PLB mice further demonstrated that by controlling the rate of SERCA2a, PLB regulates SR Ca2+ content and consequent Ca2+ release through RyR2, thus modulating intracellular Ca2+ dynamics [8, 9]. We recently showed that the anti-PLB Fab fragment penetrates into saponin-permeabilized CMs and binds PLB in situ [10]. Addition of anti-PLB Fab significantly increased the intensity (F/F0), frequency, and velocity of SCWs, and decreased the decay time (DT50) of SCWs [10]. Therefore, PLB is a powerful modulator of both contractile function and rhythmic Ca2+ activity. Furthermore, anti-PLB Fab is a novel reagent to study PLB and other factors in regulation of intracellular Ca2+ dynamics. Ca2+ handling occurs not only in SR, but also in perinuclear/nuclear membranes (NM). Nuclear Ca2+ (nCa ) signaling plays an important role in regulation of cell growth, differentiation, survival, and death in various types of cells [11-13]. In CMs, nCa2+ stores inside NM share a common lumen with the SR through a continuous intracellular membrane system [14]. Cellular E-C coupling, a major function of SR, has little effect on nCa2+ stores, underlining the highly specialized nature of local nuclear Ca2+ control. Recently, Bers and co-workers suggested an excitation-transcription coupling mechanism, linking Ca2+ release from nCa2+ stores to gene regulation in CMs [15]. Such an important role for nCa2+ stores was further suggested by recent findings of Ljubojevic et al. [16], who found that heart failure in an animal model of pressure overload, or in failed human heart, led to altered levels of SERCA and IP3-dependent inositol 1,4,5-trisphosphate receptor (IP3R) in the NM, with structural and functional remodeling in nCa2+ stores. Yet, the nCa2+ release mechanism remains controversial. Perinuclear/nuclear Ca2+ release in the form of nuclear sparks/waves[17] occurs through IP3R [18-22], but this Ca2+ release might also involve RyR and CSQ [23, 24]. There are marked differences in the biophysical characteristics between SR and perinuclear Ca2+ transients, with the latter having lower amplitude and slower rise and decay times [17]. The detailed mechanisms for regulation of nCa2+ stores remains poorly understood, even as SERCAbased Ca2+ uptake is known to exist in the nuclear regions [16, 19, 20].
AC CE P
2+
Considering the fact that PLB plays a crucial role in the modulation of Ca2+ dynamics in SR, it seems reasonable to link PLB to regulation of the poorly understood SERCA-based nCa2+ stores. Besides free SR, PLB has been localized to a non-junctionally attached membrane compartment known as corbular SR [25], although the exact placement of this membrane subcompartment in the CM biosynthetic/secretory pathway remains uncertain [26]. A single study reported PLB present in NM in smooth muscle [27], but no data exists regarding whether PLB is present in the NM of CMs. Furthermore, to date, no study has examined PLB regulation of SERCA Ca2+ uptake into nCa2+ stores. Taking advantage of anti-PLB Fab as a specific tool to acutely reverse PLB inhibition in CMs, we investigated the expression of PLB in NM 3
ACCEPTED MANUSCRIPT
AC CE P
TE
D
MA
NU
SC
RI
PT
and its contribution to nCa2+ handling.
4
ACCEPTED MANUSCRIPT 2. Methods 2.1 Cardiomyocyte and preparation permeabilization
RI
PT
Use of animals in the study is approved by the IACUC of Indiana University School of Medicine and the Methodist Research Institute, Indianapolis, Indiana and conforms the NIH Guide for the care and use of laboratory animals. CM isolation from adult C57BL/6 mice followed protocols previously reported [10]. Rabbit hearts were homogenized and centrifuged through 2.15M sucrose to isolate SR and nuclei [18, 28].
SC
2.2 Antibodies
MA
NU
The 2D12 monoclonal antibody binds specifically to PLB residues 7-13, and polyclonal PLB9 (raised against 10-20 of PLB) recognizes all isoforms of PLB. Monoclonal antibodies 1F1 and 1G7 were raised against the first 10 residues of dog and mouse isoforms of PLB, respectively. Because of a one single amino acid difference between dog (Asp2) and mouse/rabbit (Glu2) isoform of PLB, 1F1 [29, 30] or 1G7 [10] is species-specific for detection of dog or mouse/rabbit isoform of PLB, respectively. The monoclonal antibody against SERCA2a, 2A7-A1 [29] was used in the study. Antibodies against sarcalumenin and lamin A/C were purchased from Life Technology (MA3-932) and Cell Signaling Technology (#2032), respectively.
TE
D
Confocal immunofluorescence microscopy on paraformaldehyde fixed heart sections, isolated CMs, or nuclei were performed as previously described [10]. In some experiments, protein-A affinity purified antibodies were covalently labeled with various Alexa-Fluor fluorescent dyes (Invitrogen).
AC CE P
Immunoblot analyses were performed as previously described [29, 30]. We used 5µg of SR microsomes and 15µg of nuclei for SDS-PAGE and transferring to nitrocellulose membranes. Antibodybound protein bands were visualized with 125I-protein A, followed by autoradiography and quantification (Personal Fx, BioRad). 2.3 Intracellular Ca2+ imaging and analysis Intracellular Ca2+ activities were imaged at room temperature with the Leica TCS SP8 LSCM inverted microscope fitted with a 40x 1.42 NA oil immersion objective. Ca2+ transients were generated with field stimulations in intact mouse CMs loaded with fluo-4-AM (Invitrogen). Normal Tyrode’s solution contained 1.8mM Ca2+. Spontaneous Ca2+ activity of saponin-permeabilized CMs was imaged using the Ca2+ indicator dye Fluo-4, as previous described [10]. Mock internal solution contains (in mmol/L): 100 potassium aspartate, 20 KCl, 5 KH2PO4, 5 MgATP, 10 phosphocreatine, 5 U/ml creatine phosphokinase, 10 HEPES, 0.5 EGTA, 1 MgCl2, 0.015 Fluo-4 (Invitrogen), and 8% w/v dextran (molecular weight 40,000), pH 7.2 (KOH). CaCl2 was added to make the free [Ca2+]i of 400nM (WebMaxC Extended (http://www.maxchelator.stanford.edu)). 2.4 Statistical analysis Results were expressed as mean ± SEM. The statistical significance was evaluated using either paired or unpaired t tests, and followed by post hoc analyses. A value of p<0.05 was considered a statistically significant difference.
5
ACCEPTED MANUSCRIPT 3. Results 3.1 Localization of PLB in the perinuclear/nuclear membranes of cardiomyocytes
NU
SC
RI
PT
Taking advantage of our well-characterized monoclonal antibodies [10, 29, 30], and using immunofluorescence confocal microscopy, we examined the intracellular distribution of PLB dog heart tissue sections. Fig. 1A shows dual-staining of dog heart tissue sections with protein A affinity-purified monoclonal antibody against PLB 2D12 (covalently linked to Alexa Fluor-488, 2D12-488) and the monoclonal antibody against SERCA2a 2A7-A1 (covalently linked to Alexa fluor-594, 2A7-A1-594). Both antibodies most heavily immunolabeled and co-stained SR along z-lines at a spacing of about 2μm [10]. Importantly, 2D12 generated a bright ring of fluorescence around myonuclei (Fig. 1A, right panels), suggesting a robust level of PLB localized to the NM. Nuclei from non CMs did not have any such staining pattern (Fig. 1A, overlap), confirming the specificity of the antibodies. Close examination of the nuclear region also showed immunoreactive staining of invaginations into the nucleus (Fig. 1A, magnified in the bottom panels), suggesting that PLB might extend from the nuclear envelope (NE) into bilayer invaginations [16, 31].
TE
D
MA
The fluorescence intensity profile of 2D12-488, as measured longitudinally across sarcomeres and one nucleus (Fig. 1A, arrow), was co-localized with SERCA (2A7-A1-594) in both SR and NM (Fig. 1B), as previously reported for SERCA [18]. Interestingly, whereas signal intensities for 2A7-A1 antibodies were similar for NM and SR, intensities of 2D12 signal were significantly stronger in NM than that in SR. Peak intensity ratios of NM to SR for PLB and SERCA were 1.8±0.5 (n=36 nuclei in 4 dogs) and 0.9±0.2, respectively (Fig. 1B). These indicate that there are more PLB in NM than that in SR.
AC CE P
We expanded our analysis by examining intracellular distributions of PLB in heart tissue sections from mice, rabbits, and humans. Dual-staining with 2D12 and 2A7-A1, labeled with various fluorochromes, showed the same pattern of PLB localization in SR and NM (Fig. 2A, left panels), and overlap with SERCA2a. In control experiments (Fig. 2A), incubation of anti-PLB 2D12 with PLB peptide (1-31) eliminated all 2D12 staining [10], suggesting that these staining patterns are specific for anti-PLB 2D12 binding to PLB. We repeated these experiments using multiple species-specific PLB antibodies. As shown in Fig. 2B, dog specie-specific monoclonal PLB antibody 1F1 detected PLB in dog cardiac SR and NM (Canine, 1F1), but not in mouse CMs. Similarly, when using a mouse specie-specific monoclonal PLB antibody, 1G7, we detected PLB in mouse SR and NM (1G7), but not in dog CMs. Moreover, the polyclonal antibody PLB9 (which reacts with all PLB sequences regardless of species) detected PLB expression in SR and NM in both dog and mouse (PLB9). Note again that ER or nuclei of non-CMs were not stained (Fig. 2B, canine), confirming highly specific binding of these antibodies to PLB in cardiac SR and NM. Negative control experiments, carried out in the absence of primary antibodies, protein A-488 alone, did not generate any immunostaining (Fig. 1B, Protein A-488). These results demonstrate that PLB is present in cardiac NM, but at levels higher than levels in SR. 3.2 Location of PLB in isolated cardiac nuclei To further confirm the data from immunofluorescence microscopy, we examined whether PLB could be detected in isolated cardiac nuclei. Isolated cardiac nuclei, prepared from fresh rabbit left ventricle tissue, produced a highly homogeneous preparation after pelleting through a sucrose barrier (Supplementary Fig. 1). The Amido Black-stained nitrocellulose blot showed distinct differences in levels of numerous protein bands in SR and nuclei (Fig. 3A, compare rSR vs rNu), confirming the separation of SR and nuclei. Importantly, both monoclonal PLB antibodies 2D12 and 1G7, as well as 6
ACCEPTED MANUSCRIPT
RI
PT
polyclonal antibody PLB9, specifically identified double bands with identical pattern to that of SR, which correspond to pentamers and monomers of PLB (Fig. 3B). Dog specie-specific PLB antibody 1F1 only recognized PLB in dog SR (dSR), but not in rabbit SR and isolated nuclei. Furthermore, using lamin A/C (LMNA) and sarcalumenin (SAR) as proteins markers for nuclei and SR, respectively, we confirmed the high purity of SR and nuclei in these samples (Fig. 3B). Finally, we compared the expression ratio of PLB to SERCA between SR and nuclei. Fig.3C showed that the PLB to SERCA expression ratio in nuclei was significantly higher (3.55±0.58, p<0.05) than that in SR (normalized, 1.00±0.08). The high expression ratio of PLB to SERCA in NE very strongly supports the quantitative and qualitative immunofluorescence data (Fig. 1B).
TE
D
MA
NU
SC
We also performed immunofluorescence studies on the isolated nuclei from rabbit hearts. In isolated nuclei, dual-staining with 2D12-594 and 2A7-A1-488 identified both PLB and SERCA2a in nuclear envelope (NE) of cardiac nuclei (Fig. 4A). In addition to staining of the NE, both PLB and SERCA appeared to invaginate into the nuclei (Fig. 4A, Magnified). In addition to DAPI signals, we co-stained nuclei with 2D12 and lamin A/C (LMNA). Fig. 4B shows co-localization of PLB and lamin A/C in nuclei of CMs. Nuclei from non-CMs (LMNA and DAPI positive) did not have any PLB or SERCA staining, again confirming the specificity of the antibodies. Furthermore, NE of these cardiac nuclei were stained strongly by the rabbit specie-specific 1G7 (Fig. 4C, 1G7), but not by dog specie-specific 1F1. Moreover, dual-staining shows only PLB9 staining, but not detectable expression of a specific SR protein marker, sarcalumenin (SAR), in cardiac nuclei. In control experiments, anti-sarcalumenin antibody did stained SR (data not shown), again confirming the purity of the nuclei. Finally, in control experiments, secondary antibody alone had no detectable staining. Combing these data from immunostaining and western blot analysis, we conclude that PLB is highly concentrated in NE of CMs. 3.3 Effect of isoproterenol on perinuclear/nuclear Ca transients.
AC CE P
The localization of PLB in NE led us to test the initial hypothesis that PLB regulates nuclear SERCA and thus nCa2+ handling. We performed fluorescence confocal Ca2+ imaging in isolated mouse CMs with scan-lines across the length of the cell in a medial plane that shows the full nuclear diameter (Fig. 5A, white line, top panels), which is identified by bright fluo-4 labeling. This approach has been successfully used in several studies to characterize biophysical properties of nCa2+ release [16, 23]. Ca2+ transients were generated by 1 Hz field stimulations. Compared with that seen in cytosol, Ca2+ transients across perinuclear/nuclear regions were characterized by higher basal fluorescence signals (F), smaller amplitude (F/F0), and slower rise (time to peak, TtP) and decay times (DT50) (Fig. 5B, compare Nu vs Cy traces), consistent with the biophysical properties of nCa2+ transients reported by multiple labs [16, 23]. Addition of isoproterenol (100nM), which ultimately leads to phosphorylation of PLB and increases SERCA Ca2+ uptake into SR, significantly increased F/F0 (Fig. 5C) and decreased the decay time (Fig. 5E, DT50) of the Ca2+ transients not only at cytoplasmic regions, but also at perinuclear/nuclear regions (Table 1). Interestingly, while causing no significant change in the TtP at cytoplasmic regions, addition of isoproterenol significantly decreased the TtP at perinuclear/nuclear regions (Fig. 5D, Table 1), suggesting critical differences between Ca2+ transients in cytoplasmic and perinuclear regions. The decrease in DT50 of the Ca2+ transients suggest that phosphorylation of PLB increased SERCA Ca2+ uptake into nuclear Ca2+ stores. 3.4 Effect of anti-PLB Fab on properties of perinuclear spontaneous Ca2+ waves. Isoproterenol treatment results in multiple downstream effects. Therefore, taking advantage of this recently reported novel reagent, the Fab fragment of 2D12 anti-PLB antibodies (anti-PLB-Fab) [10], we examined the specific reversal of PLB inhibition on perinuclear/nuclear Ca2+ uptake and release in CMs. 7
ACCEPTED MANUSCRIPT Anti-PLB-Fab, covalently linked to Alexa-594, most heavily immunolabeled SR [10] and NE of mouse CM (Fig. 6A). All staining was eliminated by co-incubation of a peptide of PLB residue 1-31 with antiPLB Fab-594 (Fig. 6B), suggesting that these staining patterns are specific for anti-PLB Fab binding to PLB, further supporting the specificity of anti-PLB-Fab binding to PLB in SR and NE.
AC CE P
TE
D
MA
NU
SC
RI
PT
We next performed fluorescence confocal Ca2+ imaging in saponin-permeabilized CMs with scanlines across the full nuclear diameter (Fig. 6B, white line, top panels), which is identified by bright fluo-4 labeling. At 400nM of [Ca2+]i, line-scan Ca2+ images revealed Ca2+ sparks and SCWs in the CM (Fig. 6B, Ctl). Compared to SCWs in the cytosol (SCWCy), SCWNu had relatively high basal fluorescence amplitudes, low fluorescence amplitudes (F/F0), and slow TtP and DT50 (Fig.6Bb,c compare Nu vs Cy trace, Table 1), consistent with that recorded in intact CMs and reports by others [16, 17, 23]. Addition of anti-PLB Fab (100 μg/mL), which acutely reversed PLB inhibition of SERCA, significantly increased the frequency of SCWCy, enhanced amplitude (ΔF/F0) and decreased DT50 but with no significant change in the TtP of SCWCy (Fig. 6B (d-f), Cy), confirming our recent work [10]. Importantly, after addition of antiPLB Fab, ΔF/F0 of SCWNu significantly increased, while both TtP and DT50 significantly decreased (Fig. 6B (d-f), Nu, p<.05, Table 1). In control experiments, the addition of boiled anti-PLB Fab, or the antiSERCA2a antibody 2A7-A1 did not affect SCWs (data not shown) [10]. These results suggest for the first time that acute reversal of PLB actions by anti-PLB Fab leads to increased SERCA-generated Ca2+ uptake from perinuclear/nuclear regions.
8
ACCEPTED MANUSCRIPT 4.
Discussion
PT
In this study, we showed that PLB is present at relatively high levels in nuclear membranes compared to SR, and exhibits a regulation of nuclear SERCA2a that reflects its possible role in regulation of nCa2+ uptake and release. 4.1 Localization of PLB to NE of CMs.
TE
D
MA
NU
SC
RI
Since PLB was originally localized to SR in adult CMs in the 1980s [25], little detailed characterization of subcellular localization of PLB have been reported. Ferguson et al. [27] described PLB present in nuclear membranes of smooth muscle cells, but PLB has never been localized to NE in heart tissue or CMs. Given the long history of PLB studies, we were cautious in this study, employing multiple approaches to ensure the specific detection of PLB expression in NE in CMs. We first analyzed tissues, cells, and isolated nuclei using multiple extensively characterized, highly-specific PLB monoclonal and polyclonal antibodies [10, 29, 30]. Secondly, our approach used protein-A affinity purified antibodies covalently labeled with fluorescent dyes, allowing us to perform dual staining of PLB and SERCA, each potentially present in NE. Third, we detected different PLB isoforms with the use of species-specific PLB antibodies 1F1 and 1G7, providing proper controls for each. Fourth, immunoblot analysis showed an identical pattern of PLB5/PLB1 in isolated nuclei and in SR. Considering the fact that these PLB antibodies recognize PLB residues 1-10 (1G7, 1F1), 7-13 (2D12), and 10-20 (PLB9), we believed that the signals in nuclei were form PLB, not a new PLB-homologous protein. Finally, because it is impractical to compare quantitatively the amount of PLB between pure SR membranes and whole nuclei, we showed that higher expression ratio of PLB to SERCA in nuclei than that in SR. Taken together, these approaches yielded a large consistent set of data, supporting the conclusion that PLB is expressed at elevated levels in NE of CMs.
AC CE P
Free SR proteins and junctional SR proteins (e.g., RyR2) have different intracellular distributions, suggesting that divergent trafficking pathways may exist for the proteins resident to the two known sets of SR subcompartments. Recently, Sleiman et al [26] proposed a basic model of junctional SR (jSR) biogenesis and protein trafficking from rough ER to jSR. They showed that jSR resident proteins, including calsequestrin, triadin, junctin, (parts of the RyR2 complex which carry out Ca2+ release), may sort at translocons in NM, from whence a posttranslational trafficking pathway commences along a transverse/radial, microtubule-dependent pathway. These authors were also able to distinguish a second longitudinal ER subcompartment (referred to as smooth SR/free SR). Whereas junctional transmembrane proteins junctin and triadin transition after NE translation to a transverse pathway towards junctional SR sites, they have hypothesized that resident free SR proteins might follow a different, and as yet unknown route to fill free SR. The data shown in this study suggest that PLB may also be synthesized in perinuclear rough ER, where it can undergo a relative NE retention and accumulation, leading to a steadystate perinuclear concentration. Although the mechanism for this retention will require further studies, the accumulation of the proteins used in Ca2+ sequestration into perinuclear SR subcompartments suggests that new cell biological mechanisms might underlie physiological control of homeostasis between contractile and transcriptional subdomains. 4.2. PLB and possible regulation of nCa2+ stores Molecular mechanisms by which the nCa2+ stores are regulated remain unknown. In the present study, we provide initial evidence that PLB can play a role in regulation of nCa2+ dynamics in CMs. We used our recently reported specific effect of anti-PLB Fab on properties of SCWCy as a positive control in the study, and showed significantly increased Ca2+ uptake (reduction in DT50) and release (increase in ΔF/F0 and frequency) in SCWNu. Although NM and SR form a contiguous lumenal Ca2+ reservoir [14], there are 9
ACCEPTED MANUSCRIPT
SC
RI
PT
differences between biophysical properties of SCWCy and SCWNu, (e. g., small F/F0, slow TtP and DT50), consistent with the idea that local Ca2+ signaling is highly specialized and differentially modulated [16, 17], Interestingly, reversal of PLB inhibition resulted in similar changes in SCW Cy and SCWNu in terms of F/F0 and t1/2, but significant differences in TtP. These differences between SCWCy and SCWNu further suggested that the Ca2+ uptake and release activities we measured were generated in perinuclear/nuclear regions and were regulated by PLB. We speculate that the significant decrease in TtP might reflect differences in density of T-tubules and in geometry between lumens of SR and NM, the latter is generally thought to have a much smaller lumen than SR. Whether higher levels of intralumenal Ca 2+ within NM would cause increased Ca2+ release rate has not been determined, to our knowledge. Based on these results, we propose that PLB is a unique regulator for nCa2+ signaling, offering a potential missing component to an understanding of nCa2+ store regulation. 4.3 PLB and heart disease.
AC CE P
TE
D
MA
NU
Several PLB mutations have been identified that can cause human heart failure [32-34]. Furthermore, several animal models of disease that result from PLB mutations cannot be explained solely by dysfunction of SR Ca2+ dynamics. For example, a naturally occurring PLB mutant that exhibits a lethal cardiomyopathy, R14Del-PLB, was recently found to exhibit “an abnormal cytoplasmic distribution and perinuclear accumulation [of PLB]” in a patient’s heart (Fig. 1b of ref.[35]). In an earlier study, R14DelPLB was shown to be mis-routed to interact with the Na+,K+-pump in the plasma membranes [36]. These results highlight the complexity and impact of trafficking and distribution of PLB/SERCA2a in perinuclear and NE, yet mechanisms remain very poorly defined. Furthermore, although similar to PLBKO in terms of global SR Ca2+ handling, over-expression of an R14Del-PLB mutation on top of a PLBKO background caused cardiac hypertrophy [36]. Moreover, several mouse models, e.g., over-expression of gain-of-function PLB mutations [37], can generate a severe cardiac hypertrophy by an uncertain molecular mechanism. Louch et al [38] showed that SERCA2b knock-in in SERCA2a-KO/PLB-KO mice normalized global Ca2+ dynamics, but also caused cardiac hypertrophy. They suggested that global SR Ca2+ handling is not the primary cause of hypertrophic signaling. It should be noted that because the regulating role of PLB nCa2+ stores is unknown, none of these studies examined nCa2+ signaling, which is known to contribute significantly to activation of cardiac hypertrophic signaling [39, 40]. Our findings that PLB regulates nCa2+ handling may provide an important mechanistic link between alteration of PLB function and cardiac disease. The localization and role of PLB in NE in disease conditions need to be studied in the future. 4.4 Limitations and future directions. In the present study, the resolution of PLB detection in NE is limited by the use of diffractionlimited light microscopy. Although these light microscopic studies shown here provide an interpretive template for such studies, electron microscopy and super-resolution imaging systems would provide better optical resolution and aid in establishing the precise intracellular sites of PLB concentration. Preliminary super-resolution data using STED imaging (T. Kohl, Z. Chen, and S. Lehnart, manuscript in preparation) indicates that PLB is localized to the NE as opposed to alternate perinuclear membranes. The concentration and calibration differences of fluo-4 based Ca2+ indicators may contribute to the difference in basal fluorescence intensity in SR and in the nuclei. Due to these issues, it is possible that SCWNu is contaminated with SCWCy. Due to the resolution of Ca2+ imaging, we cannot distinguish the Ca2+ activity in perinuclear membranes and inside the nucleus. Although we showed the highly specific the anti-PLB Fab effect on SCWNu, future testing of Ca2+ release directly in isolated cardiac nuclei might add clarity to our findings. Finally, while the current study explored the logical hypothesis that PLB in NE regulates nuclear SERCA, our finding of higher concentration of PLB in NE than in SR hints at the
10
ACCEPTED MANUSCRIPT possibility that PLB may functionally interact with other nuclear proteins. This possibility will be explored in further studies.
RI
Source of Funding
PT
In summary, we have provided compelling new evidence that PLB is present at high levels in the NE of CMs where it can contribute to the modulation of nCa2+ handling. These findings should lead to further studies into PLB biology, PLB effects on nCa2+ signaling in CMs and its possible contribution to nCa2+mediated transcriptional regulation.
NU
SC
This study was supported in part by NIH Grants P01 HL78931, R0171140, a Medtronic-Zipes Endowment (PSC), the Indiana University Health-Indiana University School of Medicine Strategic Research Initiative (PSC) and the Dr. Charles Fisch Cardiovascular Research Award endowed by Dr. Suzanne B. Knoebel of the Krannert Institute of Cardiology (ZC).
MA
Acknowledgments
Disclosures
AC CE P
None.
TE
D
We thank Glen A. Schmeisser, Jian Tan, Xiao Shi (Michelle) and Jin Guo for great technical supports. We also thank Dechun Yin for helpful suggestions.
11
ACCEPTED MANUSCRIPT References
AC CE P
TE
D
MA
NU
SC
RI
PT
[1] Simmerman HK, Jones LR. Phospholamban: protein structure, mechanism of action, and role in cardiac function. Physiol Rev. 1998;78:921-47. [2] MacLennan DH, Kranias EG. Phospholamban: a crucial regulator of cardiac contractility. Nat Rev Mol Cell Biol. 2003;4:566-77. [3] Kranias EG, Hajjar RJ. Modulation of cardiac contractility by the phospholamban/SERCA2a regulatome. Circ Res. 2012;110:1646-60. [4] Lindemann JP, Jones LR, Hathaway DR, Henry BG, Watanabe AM. beta-Adrenergic stimulation of phospholamban phosphorylation and Ca2+-ATPase activity in guinea pig ventricles. J Biol Chem. 1983;258:464-71. [5] Bers DM. Cardiac excitation-contraction coupling. Nature. 2002;415:198-205. [6] Eisner D, Bode E, Venetucci L, Trafford A. Calcium flux balance in the heart. J Mol Cell Cardiol. 2013;58:110-7. [7] Sham JS, Jones LR, Morad M. Phospholamban mediates the beta-adrenergic-enhanced Ca2+ uptake in mammalian ventricular myocytes. Am J Physiol. 1991;261:H1344-9. [8] Santana LF, Kranias EG, Lederer WJ. Calcium sparks and excitation-contraction coupling in phospholamban-deficient mouse ventricular myocytes. J Physiol. 1997;503 ( Pt 1):21-9. [9] Bai Y, Jones PP, Guo J, Zhong X, Clark RB, Zhou Q, et al. Phospholamban knockout breaks arrhythmogenic Ca2+ waves and suppresses catecholaminergic polymorphic ventricular tachycardia in mice. Circ Res. 2013;113:517-26. [10] Chan YH, Tsai WC, Song Z, Ko CY, Qu Z, Weiss JN, et al. Acute reversal of phospholamban inhibition facilitates the rhythmic whole-cell propagating calcium waves in isolated ventricular myocytes. J Mol Cell Cardiol. 2015;80C:126-35. [11] Stehno-Bittel L, Perez-Terzic C, Clapham DE. Diffusion across the nuclear envelope inhibited by depletion of the nuclear Ca2+ store. Science. 1995;270:1835-8. [12] Hardingham GE, Chawla S, Johnson CM, Bading H. Distinct functions of nuclear and cytoplasmic calcium in the control of gene expression. Nature. 1997;385:260-5. [13] Alonso MT, Garcia-Sancho J. Nuclear Ca2+ signalling. Cell calcium. 2011;49:280-9. [14] Wu X, Bers DM. Sarcoplasmic reticulum and nuclear envelope are one highly interconnected Ca2+ store throughout cardiac myocyte. Circ Res. 2006;99:283-91. [15] Wu X, Zhang T, Bossuyt J, Li X, McKinsey TA, Dedman JR, et al. Local InsP3-dependent perinuclear Ca2+ signaling in cardiac myocyte excitation-transcription coupling. J Clin Invest. 2006;116:675-82. [16] Ljubojevic S, Radulovic S, Leitinger G, Sedej S, Sacherer M, Holzer M, et al. Early remodeling of perinuclear Ca2+ stores and nucleoplasmic Ca2+ signaling during the development of hypertrophy and heart failure. Circulation. 2014;130:244-55. [17] Luo D, Yang D, Lan X, Li K, Li X, Chen J, et al. Nuclear Ca2+ sparks and waves mediated by inositol 1,4,5-trisphosphate receptors in neonatal rat cardiomyocytes. Cell calcium. 2008;43:165-74. [18] Bare DJ, Kettlun CS, Liang M, Bers DM, Mignery GA. Cardiac type 2 inositol 1,4,5-trisphosphate receptor: interaction and modulation by calcium/calmodulin-dependent protein kinase II. J Biol Chem. 2005;280:15912-20. [19] Gerasimenko OV, Gerasimenko JV, Tepikin AV, Petersen OH. ATP-dependent accumulation and inositol trisphosphate- or cyclic ADP-ribose-mediated release of Ca2+ from the nuclear envelope. Cell. 1995;80:439-44. [20] Humbert JP, Matter N, Artault JC, Koppler P, Malviya AN. Inositol 1,4,5-trisphosphate receptor is located to the inner nuclear membrane vindicating regulation of nuclear calcium signaling by inositol 1,4,5-trisphosphate. Discrete distribution of inositol phosphate receptors to inner and outer nuclear membranes. J Biol Chem. 1996;271:478-85. [21] Zima AV, Bare DJ, Mignery GA, Blatter LA. IP3-dependent nuclear Ca2+ signalling in the mammalian heart. J Physiol. 2007;584:601-11. 12
ACCEPTED MANUSCRIPT
AC CE P
TE
D
MA
NU
SC
RI
PT
[22] Zima AV, Blatter LA. Inositol-1,4,5-trisphosphate-dependent Ca2+ signalling in cat atrial excitationcontraction coupling and arrhythmias. J Physiol. 2004;555:607-15. [23] Guo A, Cala SE, Song LS. Calsequestrin accumulation in rough endoplasmic reticulum promotes perinuclear Ca2+ release. J Biol Chem. 2012;287:16670-80. [24] Escobar M, Cardenas C, Colavita K, Petrenko NB, Franzini-Armstrong C. Structural evidence for perinuclear calcium microdomains in cardiac myocytes. J Mol Cell Cardiol. 2011;50:451-9. [25] Jorgensen AO, Jones LR. Immunoelectron microscopical localization of phospholamban in adult canine ventricular muscle. J Cell Biol. 1987;104:1343-52. [26] Sleiman NH, McFarland TP, Jones LR, Cala SE. Transitions of protein traffic from cardiac ER to junctional SR. J Mol Cell Cardiol. 2015;81C:34-45. [27] Ferguson DG, Young EF, Raeymaekers L, Kranias EG. Localization of phospholamban in smooth muscle using immunogold electron microscopy. J Cell Biol. 1988;107:555-62. [28] Tadevosyan A, Allen BG, Nattel S. Isolation and study of cardiac nuclei from canine myocardium and adult ventricular myocytes. Methods in molecular biology. 2015;1234:69-80. [29] Chen Z, Akin BL, Jones LR. Mechanism of reversal of phospholamban inhibition of the cardiac Ca2+-ATPase by protein kinase A and by anti-phospholamban monoclonal antibody 2D12. J Biol Chem. 2007;282:20968-76. [30] Chen Z. Competitive displacement of wild-type phospholamban from the Ca-free cardiac calcium pump by phospholamban mutants with different binding affinities. J Mol Cell Cardiol. 2014;76c:130-7. [31] Fricker M, Hollinshead M, White N, Vaux D. Interphase nuclei of many mammalian cell types contain deep, dynamic, tubular membrane-bound invaginations of the nuclear envelope. The Journal of cell biology. 1997;136:531-44. [32] Haghighi K, Kolokathis F, Gramolini AO, Waggoner JR, Pater L, Lynch RA, et al. A mutation in the human phospholamban gene, deleting arginine 14, results in lethal, hereditary cardiomyopathy. Proc Natl Acad Sci USA. 2006;103:1388-93. [33] Haghighi K, Kolokathis F, Pater L, Lynch RA, Asahi M, Gramolini AO, et al. Human phospholamban null results in lethal dilated cardiomyopathy revealing a critical difference between mouse and human. J Clin Invest. 2003;111:869-76. [34] Schmitt JP, Kamisago M, Asahi M, Li GH, Ahmad F, Mende U, et al. Dilated cardiomyopathy and heart failure caused by a mutation in phospholamban. Science. 2003;299:1410-3. [35] Karakikes I, Stillitano F, Nonnenmacher M, Tzimas C, Sanoudou D, Termglinchan V, et al. Correction of human phospholamban R14del mutation associated with cardiomyopathy using targeted nucleases and combination therapy. Nat Commun. 2015;6:6955. [36] Haghighi K, Pritchard T, Bossuyt J, Waggoner JR, Yuan Q, Fan GC, et al. The human phospholamban Arg14-deletion mutant localizes to plasma membrane and interacts with the Na/KATPase. J Mol Cell Cardiol. 2012;52:773-82. [37] Haghighi K, Schmidt AG, Hoit BD, Brittsan AG, Yatani A, Lester JW, et al. Superinhibition of sarcoplasmic reticulum function by phospholamban induces cardiac contractile failure. J Biol Chem. 2001;276:24145-52. [38] Louch WE, Vangheluwe P, Bito V, Raeymaekers L, Wuytack F, Sipido KR. Phospholamban ablation in hearts expressing the high affinity SERCA2b isoform normalizes global Ca2+ homeostasis but not Ca2+-dependent hypertrophic signaling. Am J Physiol Heart Circ Physiol. 2012;302:H2574-82. [39] Nakayama H, Bodi I, Maillet M, DeSantiago J, Domeier TL, Mikoshiba K, et al. The IP3 receptor regulates cardiac hypertrophy in response to select stimuli. Circ Res. 2010;107:659-66. [40] Houser SR, Molkentin JD. Does contractile Ca2+ control calcineurin-NFAT signaling and pathological hypertrophy in cardiac myocytes? Science signaling. 2008;1:pe31.
13
ACCEPTED MANUSCRIPT
RI
PT
Table 1 Characteristics of spontaneous Ca2+ wave (SCW) initiation Ca2+ transients were measured in intact mouse CMs before (Ctl) and after addition of 100nM isoproterenol (ISO) as indicated in Figure 5. Spontaneous Ca2+ waves SCWs were measured in permeabilized mouse CMs before (Ctl) and after addition of anti-PLB Fab (Fab) at [Ca2+]i of 400nM as indicated in Figure 6. The characteristics of Ca2+ transients or SCWs at the cytoplasmic (Cy) and perinuclear (Nu) regions, including F/F0 , DT50, and time-to-peak (TtP) were measured. Data are means ± SEM. (n>9 in at least 4 mice; * indicates p<0.05 vs control by paired t test). Permeabilized Cytosol
ISO
ΔF/F0
3.0±0.2
4.2±0.4*
2.7±0.2
3.7±0.3*
DT50 (ms)
115±8
94±7*
162±9
120±7*
TtP
61.9±5.4
57.4±5.5
96.9±6.3
70.0±6.6*
perinuclear
Ctl
Fab
1.8±0.1
2.6±0.1*
1.0±0.1
1.4±0.1*
82±4
60±3*
151±8
110±6*
71±4
65±2
124±8
84±4*
Ctl
Fab
AC CE P
TE
D
MA
NU
Ctl
Perinuclear/ nuclear Ctl ISO
SC
Intact Cytosol
14
ACCEPTED MANUSCRIPT
SC
RI
PT
Figure 1. Intracellular distribution of PLB in canine cardiomyocytes. Confocal immunofluorescence microscopy images were obtained from canine heart tissue sections. A. Dual-staining images with the monoclonal anti-PLB antibody 2D12 (conjugated with Alexa fluor-488, left panels) and 2A7-A1 (conjugated with Alexa fluor-594, middle panels). Merged images are in the right panels. The regions indicated in the boxes were refocused to show magnified images in the bottom panels. Nuclei were stained with DAPI in blue in all images. B. Intensity profiles of sarcomeres and a nucleus. Region pointed by an arrow in A was analyzed (purple box). Graph shows intensity ratio between NM and SR for PLB and SERCA. * indicates p< 0.05. Bar represents 10 µm. Similar patterns of staining were obtained in at least 10 samples.
AC CE P
TE
D
MA
NU
Figure 2. Species-specific detection of PLB in cardiomyocytes from various species. Confocal immunofluorescence microscopy images were obtained from mouse CMs, rabbit and human heart tissue sections. A. Dual-staining with 2D12-488 and 2A7-A1-594 for mouse CMs, and dual-staining with 2D12594 and 2A7-A1-488 for rabbit and human heart tissue sections. Immunostainings with co-incubation of 2D12 and a PLB peptide 1-31 are shown in the right panels. B. Immunostaining of PLB using speciesspecific anti-PLB antibodies (Abs) in mouse CMs (upper panels) and dog heart tissue sections (lower panels). Abs used were: dog specie-specific PLB monoclonal Ab 1F1, mouse specie-specific PLB monoclonal Ab 1G7, and polyclonal antibody PLB9. Protein A-488 was used to visualize antibody binding signals. Immunostainings with protein A-488 alone but no primary antibody incubation are in right panels. Bar represents 10 µm. Similar patterns of staining were obtained in at least 10 samples in each condition.
Figure 3. Immunoblot analysis of PLB protein levels in isolated cardiac nuclei. 5µg rabbit cardiac SR (rSR) and 15µg nuclei (rNu) were subjected to SDS-PAGE. A. nitrocellulose stained with Amido Black, B. autoradiograms after incubation with antibodies followed by 125I-proteinA. Identical blots were stained with various antibodies (indicated at the bottom) against PLB (2D12, 1G7, PLB9, and 1F1 (canine SR was added as a positive control)), against sarcalumenin (SAR), against lamin A/C (LMNA), and against SERCA (2A7-A1). C. The immunoblot and plot showing the expression ratio of PLB to SERCA. PLB/SERCA in SR is normalized to 1 (n=5, * indicates p<0.05). PLB1 and PLB5, PLB monomer and pentamer, respectively.
Figure 4. Immunofluorescence analysis of PLB in isolated cardiac nuclei. Confocal immunofluorescence microscopy images were obtained from isolated rabbit cardiac nuclei. A. dualstaining with 2D12, conjugated with Alexa fluor-594, and 2A7-A1, conjugated with Alexa fluor-488. B. dual-staining with 2D12 and polyclonal lamin A/C antibodies (Ab), visualized by goat anti-mouse and goat anti-rabbit secondary Abs. Merged and magnified images (indicated by boxes) are shown in two right panels. C. Ab used (From left to right): rabbit and dog species-specific PLB monoclonal Ab 1G7 and 1F1, respectively, visualized with protein A-488; dual-staining with PLB9 and sarcalumenin Ab (SAR); no primary Ab with secondary Abs alone. Blue is DAPI in all images. Bar represents 10 µm. Similar patterns of staining were obtained in at least 10 samples in each condition.
15
ACCEPTED MANUSCRIPT
RI
PT
Figure 5. The effect of isoproterenol on cytoplasmic and perinuclear/nuclear Ca2+ transients in intact mouse CMs. Intact CMs were in Tyrode’s solution with 1.8mM Ca2+. A. Representative confocal line-scan Ca2+ images using Fluo-4-AM were obtained in the same CM before (Ctl) and after addition of 100 nM isoproterenol (ISO). Nucleus is between red lines. Scan-line (dashed white) is over cytosol and nucleus. B. Traces show intensities of fluorescent signals (F) of Ca2+ transients at cytoplasmic and perinuclear/nuclear regions. C-E. Bar graphs show ensemble average cell parameters for all measurements (C, intensity, ΔF/F0; D, time to peak, TtP; E, 50% decay time, DT50) of perinuclear/nuclear (Nu) and cytoplasmic (Cy) regions before and after addition of isoproterenol. See also Table 1 for data. * indicates p<0.05 vs control (average of at least 10 CMs in 6 mice).
AC CE P
TE
D
MA
NU
SC
Figure 6. Effects of anti-PLB Fab on spontaneous Ca2+ waves (SCWs) in cytoplasmic and perinuclear regions of CMs. A. Mouse CM stained by anti-PLB Fab, conjugated with Alexa 594, nuclei (DAPI) are in blue in all images. The right panel shows data from co-incubation of anti-PLB Fab-594 and a PLB peptide 1-31. B. representative confocal line-scan Ca2+ images using Fluo-4 Ca2+ indicator were obtained in the same permeabilized mouse CM (top panels) before (Ctl) or after (Fab) addition of 100 µg/ml anti-PLB Fab. The nucleus is between the red lines. Scan-line (dashed white) is over cytosol and nucleus. Ca2+ concentration was 400 nM. b. Traces show intensities of fluorescent signals (F) of SCWs at cytoplasmic and perinuclear/nuclear regions. c. ensemble averages of traces. d-f. Bar graphs show ensemble parameters (d, intensity, ΔF/F0; e, time to peak, TtP; f, 50% decay time, DT50) of perinuclear/nuclear (Nu) and cytoplasmic (Cy) before and after addition of anti-PLB Fab. See also Table 1 for data. * indicates p<0.05 vs control (average of at least 9 CMs in 4 mice).
16
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
AC CE P
TE
D
MA
Figure 1
17
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
AC CE P
TE
D
Figure 2
18
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
AC CE P
TE
D
MA
Figure 3
19
AC CE P
Figure 4
TE
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
20
Figure 5
AC CE P
TE
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
21
Figure 6
AC CE P
TE
D
MA
NU
SC
RI
PT
ACCEPTED MANUSCRIPT
22
PT
ACCEPTED MANUSCRIPT
AC CE P
TE
D
MA
NU
SC
RI
Graphical abstract
23
ACCEPTED MANUSCRIPT Highlights: We tested if phospholamban (PLB) is in nuclear envelope (NE) of cardiomyocytes (CMs).
PT
Multiple PLB antibodies stained CM perinuclear/nuclear membranes in several species.
RI
Immunoblot/fluorescence assays showed high levels of PLB in NE of isolated CM nuclei.
SC
The Fab fragment of PLB antibody increased perinuclear/nuclear Ca uptake and release.
AC CE P
TE
D
MA
NU
PLB was localized in NE in high concentration and involved in CM nuclear Ca handling.
24