Original Article

Glycosylated Chromogranin A in Heart FailureCLINICAL PERSPECTIVE

Implications for Processing and Cardiomyocyte Calcium Homeostasis

Anett Hellebø Ottesen, Cathrine R. Carlson, William E. Louch, Mai Britt Dahl, Ragnhild A. Sandbu, Rune Forstrøm Johansen, Hilde Jarstadmarken, Magnar Bjørås, Arne Didrik Høiseth, Jon Brynildsen, Ivar Sjaastad, Mats Stridsberg, Torbjørn Omland, Geir Christensen, Helge Røsjø
https://doi.org/10.1161/CIRCHEARTFAILURE.116.003675
Circulation: Heart Failure. 2017;10:e003675
Originally published February 16, 2017

Jump to

Abstract

Background—Chromogranin A (CgA) levels have previously been found to predict mortality in heart failure (HF), but currently no information is available regarding CgA processing in HF and whether the CgA fragment catestatin (CST) may directly influence cardiomyocyte function.

Methods and Results—CgA processing was characterized in postinfarction HF mice and in patients with acute HF, and the functional role of CST was explored in experimental models. Myocardial biopsies from HF, but not sham-operated mice, demonstrated high molecular weight CgA bands. Deglycosylation treatment attenuated high molecular weight bands, induced a mobility shift, and increased shorter CgA fragments. Adjusting for established risk indices and biomarkers, circulating CgA levels were found to be associated with mortality in patients with acute HF, but not in patients with acute exacerbation of chronic obstructive pulmonary disease. Low CgA-to-CST conversion was also associated with increased mortality in acute HF, thus, supporting functional relevance of impaired CgA processing in cardiovascular disease. CST was identified as a direct inhibitor of CaMKIIδ (Ca2+/calmodulin-dependent protein kinase IIδ) activity, and CST reduced CaMKIIδ-dependent phosphorylation of phospholamban and the ryanodine receptor 2. In line with CaMKIIδ inhibition, CST reduced Ca2+ spark and wave frequency, reduced Ca2+ spark dimensions, increased sarcoplasmic reticulum Ca2+ content, and augmented the magnitude and kinetics of cardiomyocyte Ca2+ transients and contractions.

Conclusions—CgA-to-CST conversion in HF is impaired because of hyperglycosylation, which is associated with clinical outcomes in acute HF. The mechanism for increased mortality may be dysregulated cardiomyocyte Ca2+ handling because of reduced CaMKIIδ inhibition.

Introduction

Paracrine factors and hormones are important biomarkers and key targets for therapy in heart failure (HF). Accordingly, testing novel hormonal substances as biomarkers and potential therapeutic strategies may prove valuable to advance our understanding of the pathophysiology of HF.

See Clinical Perspective

Chromogranin A (CgA) is a 48-kDa prohormone that is produced in many tissues throughout the body, including in neuroendocrine and myocardial cells.1–3 Previous studies have found circulating CgA levels to be associated with prognosis in patients with acute coronary syndromes,4–6 HF,7–9 and severe sepsis.10 Furthermore, the association between CgA levels and clinical outcome has also been demonstrated in multivariable models that have adjusted for other risk variables.4–6,8–10 Thus, CgA seems to reflect pathophysiology not measured by the established risk indices in cardiovascular disease (CVD). Although originally considered a surrogate marker of adrenergic tone,11 previous studies in patients with CVD have not found significant correlations between CgA and catecholamine levels.5,6,8 In contrast, the literature supports that CgA influences myocardial function via the production of short cleavage fragments. The fragment catestatin (CST; CgA352-372) may be of particular relevance for CVD. CST has been found to reduce myocardial β-adrenergic and endothelin-1 signaling, to affect myocardial contractility, to attenuate myocardial ischemia–reperfusion injury,12 and to have a direct effect on Ca2+ handling by interaction with calmodulin (CaM) in noncardiac cells.13 However, currently no information is available in the literature concerning the processing of CgA to shorter peptide fragments in HF. In addition, it is not established whether CgA fragments like CST may directly influence cardiomyocyte function. Although previous studies have reported that CST modulates cardiomyocyte Ca2+ handling only via indirect mechanisms,14 this finding is surprising given the direct effect of CST on Ca2+ handling in noncardiac cells,13 the influence by other chromogranin–secretogranin proteins on cardiomyocyte Ca2+ handling,15 and the ability of CST to directly interact with CaM.13 CaM is an upstream activator of CaMKIIδ (Ca2+/CaM-dependent protein kinase IIδ), which is an established regulator of cardiomyocyte Ca2+ handling.16 Because overactive myocardial CaMKIIδ signaling constitutes a central mechanism of HF, direct regulation of this pathway by CST would link CgA and CST directly to CVD. Accordingly, the aims of this study were to characterize CgA processing in acute HF and to explore whether CST directly modulates cardiomyocyte Ca2+ handling via CaMKIIδ inhibition.

Methods

Details can be found in the Appendix in the Data Supplement. Animal experiments conformed to the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85-23, revised 1996). The clinical study was performed according to the Declaration of Helsinki, approved by the Regional Ethics Committees, and all participants provided written informed consent before study commencement.

Postinfarction HF Mouse Model and Immunoblotting

Experimental myocardial infarctions were induced by permanent ligation of the main left coronary artery, and the presence of HF was verified by echocardiography 1 week later.3,17 Biopsies from the infarcted and noninfarcted left ventricle (LV), right ventricle, and other organs were collected. Deglycosylation was performed by adding different combinations of enzyme cocktails. Short CST fragments were assessed by using a high percentage gel and subjecting the lower part of the membrane to longer exposure.

Clinical Study

We included 314 patients who were hospitalized because of acute dyspnea at Akershus University Hospital, Lørenskog, Norway, into the ACE 2 study (Akershus Cardiac Examination) (Figure I in the Data Supplement). Patients were characterized as acute HF and acute exacerbation of chronic obstructive pulmonary disease (COPD) by an adjudication committee of 2 experts working independently.18 CgA levels were measured by ELISA and CST levels by an in-house radioimmunoassay.19 We calculated the percentage of CgA not converted to CST (CgA-to-CST conversion) in individual patients by this formula: ([CgA−CST]/CgA)×100. High-sensitivity troponin T, NT-proBNP (N-terminal pro-B-type natriuretic peptide), and norepinephrine levels were measured as previously reported.15

Effect by CST on Ca2+ Handling

CST-CaMKIIδ interactions were examined by immunoprecipitation, pull-down experiments, and surface plasmon resonance. CaMKIIδ activity was assessed by in vitro kinase activity assays. Effects of CST on CaMKIIδ, ryanodine receptor 2, and phospholamban (PLB) phosphorylation were studied in Langendorff-perfused hearts. BayK was included in the perfusate to promote L-type Ca2+ channel opening and, thereby, increase CaMKIIδ activity. Adult ventricular cardiomyocytes were isolated as previously reported.15 Contractions and Ca2+ transients were elicited by field stimulation, Ca2+ sparks were measured in resting cells, and cytosolic [Ca2+] was detected by confocal microscopy and whole-cell photometry. Cells were preincubated and perfused with 45 nmol/L or 4.5 μmol/L CST.

Statistics

Clinical data are presented as mean±SD and median (quartile 1–3) for biomarkers because of non-normal distributions and experimental data as mean±SEM. Differences between groups were examined by Student’s t test or the Mann–Whitney U test, and the Chi-square test and correlation coefficients were calculated by the Spearman rank test. Data with multiple groups were assessed by the Kruskal–Wallis test. We present patients stratified according to biomarker quartiles in Kaplan–Meier plots, and the association with mortality was examined by the log-rank test. Variables associated with mortality in the acute HF patients were also examined by single variable, and multivariable Cox proportional hazard regression analysis with hazard ratios presented with 95% confidence intervals. Variables associated with mortality in single variable analysis were included in the multivariable model (forward selection of variables). P values <0.05 were considered significant.

Results

CgA Is Hyperglycosylated in the Failing Myocardium

Immunoblotting demonstrated high molecular weight (hmw) CgA bands in myocardial biopsies of HF, but not sham-operated animals (Figure 1A). Short CgA fragments were observed in sham-operated animals, but to a much lesser degree in HF. The hmw CgA bands were also observed in the right ventricle in HF, but not other organs examined, neither in HF nor in sham-operated animals (Figure I in the Data Supplement). The short CST fragment was also found to be highly reduced in LV of HF animals compared with CST levels in the LV of sham-operated animals (Figure 1B). Adding enzymes that cleaved N- or O-linked glycosylation modifications yielded a shift in the myocardial hmw CgA bands (Figure 1C). Combinations of enzymes that cleave O-linked glycosylation also reduced the appearance of hmw CgA bands and yielded an increase in short CgA fragments. The increase in short CgA fragments was most prominent when all enzymes were added in combination (Figure 1C, right lane).

Figure 1.
Figure 1.

Posttranslational modifications and processing of chromogranin A (CgA) in heart failure (HF). A, Immunoblotting demonstrates high molecular weight (hmw) CgA bands in myocardial biopsies of HF, both from the viable area and from the infarcted area. No hmw bands are observed in sham-operated animals. Short CgA fragments including catestatin (CST) were observed in sham-operated animals, but to a much lesser degree in HF (sham, n=15; HF (viable), n=16; HF (infarct), n=6). B, By immunoblotting, the short CST fragment was found to be highly reduced in HF animals compared with that in sham-operated animals (n=2). C, Adding enzymes that cleaved N- or O-linked glycosylation modifications yielded a shift in the myocardial hmw CgA bands. Some combination of enzymes also reduced the appearance of hmw CgA bands and yielded an increase in short CgA fragments (n=6). The increase in short fragments was most prominent when all enzymes where added.

Circulating CgA Levels Predict Mortality in Acute HF and Reduced CgA-to-CST Conversion Is Associated With a Poor Outcome

In total, 314 patients were included in the ACE 2 Study, and the final diagnosis was acute HF in 143 patients (46%) and acute exacerbation of COPD in 84 patients (27%), while 87 patients (28%) were hospitalized because of non–HF-, non–COPD-related dyspnea (Figure II in the Data Supplement). Among the patients classified as hospitalized for acute HF, 51 patients (36%) were considered to have HF with preserved LV ejection function (Table I in the Data Supplement). Admission CgA levels, but not CST levels, were higher in the acute HF patients compared with the patients with non–HF-related dyspnea (Table I in the Data Supplement). The proportion of CgA molecules not converted to CST molecules was also higher in acute HF patients than in the patients with non-HF dyspnea: median 57% versus median 41%, respectively, P=0.001 (Table I in the Data Supplement). Baseline CgA levels in the acute HF patients were inversely correlated with estimated creatinine clearance and diastolic blood pressure on admission, positively correlated with high-sensitivity troponin T and NT-proBNP levels and not significantly correlated with norepinephrine levels (Table II in the Data Supplement). CST levels were inversely correlated with age, but not correlated to other clinical or laboratory variables (Table II in the Data Supplement).

During median follow-up of 817 days, 66 patients (46%) with acute HF died, 35 patients (42%) with acute COPD died, and 13 patients (15%) with non–HF and non–COPD-related dyspnea died (Figure II in the Data Supplement). To include relatively homogenous patient categories with similar mortality rates during follow-up, we selected the patients hospitalized with HF or COPD for further studies. Patients with the highest CgA levels on admission for acute HF had a significantly higher risk of mortality during follow-up compared with the other HF patients (Figure 2A; P=0.001 by the log-rank test). Within 200 days of follow-up, 50% of HF patients with 4th quartile CgA levels had died compared with <10% of HF patients with CgA levels in the 1st quartile. CgA was also found to be a strong prognostic marker in Cox regression analysis, including in multivariable analysis that included NT-proBNP, high-sensitivity troponin T, and norepinephrine levels: hazard ratio (ln CgA per 1 SD increase) 1.51 (95% confidence interval, 1.16–1.95), P=0.002 (Table). In contrast, baseline CST levels did not predict mortality in patients hospitalized with acute HF (Table and Figure 2B). Pertinent to this point, patients with low CgA-to-CST conversion had a worse outcome compared with the other HF patients (Figure 2C). CgA levels, CST levels, and CgA-to-CST conversion were not associated with mortality in the patients hospitalized with acute exacerbation of COPD (Figure III in the Data Supplement).

View this table:
Table.

Predictors of Mortality During Follow-Up in Patients With Acute HF

Figure 2.
Figure 2.

Associations between chromogranin A (CgA) levels, catestatin (CST) levels, and CgA-to-CST conversion in patients with acute heart failure (HF). Stratifying patients according to admission (A) CgA quartiles, but not (B) CST quartiles, provided excellent risk stratification in the patients with acute HF. C, Patients with low CgA-to-CST conversion had worse outcome compared with the other HF patients (n=143).

CST Interacts Directly With CAMKIIδ and Reduces CAMKIIδ Activity

Because we found low CgA-to-CST conversion to be associated with mortality in acute HF, but not in acute exacerbation of COPD, we aimed to explore whether CST may directly affect cardiomyocyte function. The short CgA fragment CST shows sequence similarity to the autoregulatory domain of CaMKIIδ containing the CaM-binding site, which suggests that CST might bind directly to the substrate binding site in the catalytic domain of CaMKIIδ (Figure 3A). As predicted by bioinformatics, we identified a CST–CaMKIIδ interaction by immunoprecipitation (Figure 3B) and pull-down experiments (Figure 3C) using recombinant His–CaMKIIδ (T287D) and biotin–CST. The direct interaction between CST and CaMKIIδ was also demonstrated by surface plasmon resonance analyses with a KD=(5±3)×10–7 mol/L, a ka=(2±1)×103 mol−1 L−1 s−1, and a kd=(8±3)×10−4s−1 (Figure 3D).

Figure 3.
Figure 3.

Catestatin (CST) is an endogenous CaMKIIδ (Ca2+/calmodulin-dependent protein kinase IIδ) inhibitor via direct binding. A, CST from different species exhibits sequence similarities to the calmodulin (CaM)-binding region in CaMKIIδ; the CaM-binding motif 1-5-10 is conserved in all species shown except mouse (red box). Black boxes indicate identical or functionally similar amino acids (DNA Star, Lasergene). RKLK, Stars denotes the core pseudosubstrate site in CaMKIIδ (RKLK).20 CST–CaMKIIδ interaction was demonstrated by (B) immunoprecipitation (n=4) and (C) pull-down experiments using recombinant His-CaMKIIδ (T287D; active) and biotin–CST (n=4). D, The direct binding of CST–CaMKIIδ was verified by surface plasmon resonance analysis (n=5). The binding response (gray) was overlaid with the fit of a 1:1 interaction model (black). E, CST reduced autophosphorylation of Thr286-CaMKII in BayK-perfused hearts (BayK, n=5; CST+BayK, n=4). *P≤0.05, examined by Student’s t test. F, CaMKII activity was decreased by CST in a dose-dependent manner measured by an in vitro kinase assay (n=6–10). CN21a indicates positive control. Negative results are because of inhibition of autophosphorylation of CaMKIIδ. HRP indicates horseradish peroxidase. ***P≤0.001, examined by analysis of variance (ANOVA) and Kruskal–Wallis test.

Based on the interaction between CST and CaMKIIδ, we investigated whether CST directly affects CaMKIIδ activity. CaMKIIδ can be autophosphorylated at Thr286 in the regulatory domain of CaMKIIδ, which leads to prolonged kinase activity.16 We observed that CST reduced the autophosphorylation in BayK-perfused mouse hearts (Figure 3E), whereas CST did not reduce the basal level of autophosphorylation (Figure IVA in the Data Supplement). CST was also observed to strongly inhibit CaMKIIδ activity in a dose-dependent manner in an in vitro kinase assay (Figure 3F).

CST Reduces CaMKIIδ-Dependent Phosphorylation of the Ryanodine Receptor 2 and PLB

CaMKIIδ is known to phosphorylate proteins involved in cardiomyocyte Ca2+ homeostasis, and we examined whether CST could influence ryanodine receptor 2 phosphorylation and SERCA via phosphorylation of PLB in isolated perfused mouse hearts. CST reduced both the basal level and the BayK-induced increase in Ser2814- ryanodine receptor 2 phosphorylation (Figure 4A; Figure IVB in the Data Supplement), which is regulated by CaMKIIδ.16 This strong inhibitory effect was not observed at the Ser2808- ryanodine receptor 2 phosphorylation site (Figure 4B; Figure IVC in the Data Supplement), which is regulated by protein kinase A.20 We also found that CST reduced the basal level and the BayK-induced increase in phosphorylation of PLB at Thr17-PLB (Figure 4C; Figure IVD in the Data Supplement) and reduced the BayK-induced increase in phosphorylation of PLB at Ser16-PLB but not the basal level (Figure 4D; Figure IVE in the Data Supplement), which are regulated by CaMKIIδ and protein kinase A, respectively.16

Figure 4.
Figure 4.

Catestatin (CST) reduced CaMKIIδ (Ca2+/calmodulin-dependent protein kinase IIδ)–dependent phosphorylation of the ryanodine receptor and phospholamban. CST reduced (A) BayK-induced Ser2814-RyR2 phosphorylation, (B) but not Ser2808-RyR2 phosphorylation. CST reduced (C), BayK-induced CaMKII-dependent phosphorylation of Thr17-PLB, and (D) BayKinduced Ser16-PLB phosphorylation (BayK, n=5; CST+BayK, n=4). *P≤0.05, ***P≤0.001, examined by Student’s t test. RyR2 indicates ryanodine receptor 2.

CST Modulates Cardiomyocyte Ca2+ Homeostasis

We next examined the effects of CST on Ca2+ handling in isolated cardiomyocytes. As expected of a CaMKIIδ inhibitor,21 we observed that CST (4.5 μM) prominently reduced the dimension of Ca2+ sparks, including the full duration at half-maximum, full width at half-maximum intensity, and the time to peak of the Ca2+ spark (Figure 5A). Ca2+ spark frequency and Ca2+ wave frequency were also markedly reduced (Figure 5A). A similar reduction in Ca2+ spark dimensions and the frequency of Ca2+ sparks and waves was observed when the experiment was repeated with a 100-fold lower CST concentration (45 nmol/L; Figure 5B), which is in the range of circulating CgA levels in the HF patients. Ca2+ transients and contractions were larger across a range of stimulation frequencies (1, 0.5, 2, and 5 Hz) (Figure 6A and 6B). Kinetics of both contraction and relaxation were more rapid in CST-treated cells than in controls (Figure 6A), an effect likely explained by faster declining Ca2+ transients (Figure 6B). Frequency-dependent acceleration of relaxation and Ca2+ transient decline was significantly attenuated in the CST treatment group (Figure 6A and 6B). Consistent with reduced sarcoplasmic reticulum Ca2+ leak (sparks; Figure 5), we observed that CST treatment increased sarcoplasmic reticulum Ca2+ content (Figure 6C) without altering rates of Ca2+ extrusion from the cell or Ca2+ reuptake into the sarcoplasmic reticulum (Figure V in the Data Supplement).

Figure 5.
Figure 5.

Catestatin (CST) reduced diastolic Ca2+ leak. CST reduced Ca2+ sparks in all dimensions and reduced Ca2+ spark and wave frequency using either (A) 45 nmol/L or (B) 4.5 μmol/L of CST (Ca2+ sparks: Ctr, nhearts=3, ncells=12; CST, 4.5 μM: nhearts=3, ncells=9; Ca2+ sparks: Ctr, nhearts=9, ncells=49; CST, 45 nmol/L: nhearts=8, ncells=25). *P≤0.05, **P≤0.01, ***P≤0.001, examined by Student’s t test.

Figure 6.
Figure 6.

Catestatin (CST) augmented cardiomyocyte calcium transients and contractions. A, CST treatment induced larger and faster contractions of cardiomyocytes across a range of stimulation frequencies (1, 0.5, 2, and 5 Hz). ***P≤0.001, examined by analysis of variance (ANOVA) and Kruskal–Wallis test. B, CST increased the magnitude of cardiomyocyte Ca2+ transients and reduced the half relaxation time at all frequencies. ***P≤0.001, examined by ANOVA and Kruskal–Wallis test. C, Sarcoplasmic reticulum Ca2+ content was increased by CST (contractions: Ctr, nhearts=3, ncells=5–12; CST, nhearts=2, ncells=5–8; Ca2+ transients: Ctr, nhearts=3, ncells=13; CST, nhearts=4, ncells=7; SR content, 10 mmol/L caffeine: Ctr, nhearts=5, ncells=17; CST, nhearts=4, ncells=15). *P≤0.05, examined by Student’s t test.

Discussion

The main results of this study are (1) that CgA is hyperglycosylated in the failing myocardium, (2) that the ratio between circulating CgA-to-CST levels, but not CST levels alone, provides prognostic information in patients with acute HF, and (3) that CST has cardioprotective effects via direct CaMKIIδ inhibition. Accordingly, as CaMKII activity is increased in HF,16 myocardial CgA hyperglycosylation with reduced CgA-to-CST processing should be detrimental because of impaired local CaMKII control. Indeed, we found that acute HF patients with low CgA-to-CST conversion had a worse outcome compared with HF patients with higher CgA to CST conversion.

CgA has also previously been recognized as a biomarker that provides prognostic information in patients with CVD. Earlier studies have examined CgA in patients with ACS,4–6 HF,7–9 and severe sepsis,10 and in agreement with our present data, these studies found CgA to provide additional prognostic information to established risk indices, including biomarkers and LV function.4–6,8–10 The complementary information provided by CgA is also reflected by only modest correlations between CgA and BNP (B-type natriuretic peptide) or NT-proBNP levels4,6,7,9,10 and no correlation between CgA and catecholamine levels in patients with CVD.5,6 We now advance the present knowledge regarding CgA in CVD by demonstrating CgA hyperglycosylation in the failing myocardium and that this may have direct functional consequences via reduced CST-mediated CaMKII inhibition. Based on the knowledge from other proteins, myocardial glycosylation is found to be increased in the failing myocardium and represents a mechanism for posttranslational protein modification. Pertinent to this point, processing of glycosylated proteins can be impaired by sugar groups blocking the binding of proteases to cleavage sites. A prominent example of this is pro-BNP, which is extensively glycosylated in the middle and N-terminal end of the molecule,18 and where processing to BNP and NT-proBNP is impaired by the glycosylation. As CgA processing takes place at several dibasic cleavage sites along the full-length molecule,2 extensive CgA glycosylation could influence CgA processing, and our results support this model because short CgA and CST fragments are increased after exposing myocardial biopsies to deglycosylation enzymes. Hence, a substantial proportion of CgA molecules, at least in the failing myocardium, could be glycosylated and, thus, inaccessible for further processing. Additional mechanism for CST removal may be degradation pathways of CgA to fragments other than CST or factors that independently affect the breakdown or stability of CST, which both will have to be assessed in additional studies.

Although no group previously has reported cardiac-specific glycosylation of CgA, we think our results are supported by the literature. A previous report has found circulating CST levels to be reduced in patients with subclinical and established HF compared with subjects free from HF,22 which supports reduced CgA-to-CST processing as CgA levels are known to be increased in HF patients8 (as also demonstrated in our clinical cohort). Another group has also reported impaired CgA-to-CST processing in myocardial biopsies from aging mice.23 However, these authors did not report immunoblot bands >100 kDa, and thus, whether the reduction in CST was paralleled by increments in hmw CgA bands is unknown. Still, this study23 and another more recent study24 both observed CgA immunoblot bands above the expected full-length molecular weight of 74 kDa in myocardial tissue samples. In contrast, no such bands were evident in adrenal gland tissue samples. These groups also suggested that the myocardial hmw CgA bands could be because of glycosylation, but they did not follow-up on this hypothesis. Finally, using high-performance liquid chromatography to characterize plasma CgA fragments in stable HF patients, Pieroni et al1 also found peaks with molecular weights above the peak considered representative for full-length-CgA, which could represent the release of glycosylated CgA molecules into the circulation.

We found reduced CgA-to-CST conversion to be associated with a poor outcome in acute HF, but not in patients with acute exacerbation of COPD, which supports a role for CST in cardiomyocyte pathophysiology. Based on our data of CST as a CaMKII inhibitor and previous data that found CST to reduce myocardial ischemia–reperfusion injury,12 increase cardiomyocyte cell survival,25 and attenuate myocardial β-adrenergic and endothelin-1 signaling,26 CST seems to be a counter-regulatory paracrine ligand of relevance in CVD. Given the paramount importance of hyperactive CaMKIIδ signaling in CVD,16 the primary role of CST may be as part of a paracrine/autocrine protective feedback loop that could antagonize overactive myocardial CaMKIIδ signaling. However, our data also indicate that this system may be malfunctioning in a proportion of HF individuals because of myocardial CgA hyperglycosylation. Accordingly, although we previously have reported increased CgA mRNA levels4 and increased prohormone convertases 1/3 and prohormone convertases 2 activity in the failing myocardium,17 which are the principal proteases for CgA cleavage,27 CgA glycosylation could block the binding of these proteases to the full-length molecule and, thus, reduce CST production. Moreover, although a recent report found isoprenalin- and endothelin-infusion to increase myocardial CgA processing in healthy hearts,24 whether the same increased processing would occur in failing hearts with extensive CgA glycosylation is unknown.

CaMKIIδ is a nodal kinase in the regulation of cardiomyocyte Ca2+ handling.16 CaMKIIδ is known to be hyperactive in several cardiac pathologies, including HF, ischemia, remodeling, cardiomyocyte cell death, and arrhythmias.16 Hence, CaMKIIδ inhibition is recognized as an interesting target for attenuation of disease progression.16,28 CST exhibits the CaM-binding motif 1-5-10 and shows homology to the sequence surrounding the core pseudosubstrate site (RKLK) in CaMKIIδ,29 indicating that CST could also bind to CaMKIIδ. The ability of CST to interact directly with CaMKIIδ is consistent with prior work demonstrating that CST and other granin-derived fragments can bind to CaM in noncardiac cells13 and to CaM and CaMKIIδ in cardiomyocytes.15 Our data on Ca2+ handling in the presence of CST are also consistent with those on CST acting as CaMKIIδ inhibitor because CaMKIIδ inhibition has been found to reduce Ca2+ spark frequency and magnitude, augment sarcoplasmic reticulum Ca2+ content, and increase the magnitude of Ca2+ transients.21,30 As expected with CaMKIIδ inhibition, faster transient decay during CST treatment was also associated with more rapid contraction kinetics.31 CST also blunted frequency-dependent acceleration of relaxation and Ca2+ transients, a phenomenon reported to result from activation of SERCA at high frequencies, after CaMKIIδ-dependent phosphorylation of PLB at Thr17.32 The observed effects of CST on cardiomyocyte Ca2+ handling are in agreement with the effects of other CaMKIIδ inhibitors like KN-93 and AIP32 and our recent report on the granin protein secretoneurin.15 Hence, a possible integrated model could be that (1) the most severely ill HF patients increase CgA production as a compensatory mechanism, (2) myocardial hyperglycosylation reduces the effect of this paracrine system by blocking CgA-to-CST conversion, and (3) reduced CST production is detrimental because of lack of local CaMKIIδ control. Based on this model, targeting myocardial CgA hyperglycosylation to increase local CST production could prove valuable as a future strategy to improve outcome in HF. Still, we acknowledge that additional work is needed to decipher the mechanism, whereby the glycosylation occurs and whether interventions to reduce CgA glycosylation would lead to increased myocardial CST levels in subjects with HF. Pertinent to this point, we also found CST to reduce Ser16-PLB phosphorylation, which is not recognized as a CaMKII phosphorylation site. Accordingly, CST could have effects also outside of CaMKII inhibition, including effects via the nitric oxide-cGMP pathway, analogously to data previously demonstrated,26,33 or CST could compete with PLB for phosphorylation. Thus, there is a need for experimental studies to establish additional processes that CST could regulate in the failing myocardium, besides CaMKII inhibition, which was the focus of the current work.

In conclusion, we found CgA-to-CST ratio, but not CST alone, to be a valuable prognostic biomarker in acute HF. Moreover, we demonstrate increased myocardial CgA glycosylation and impaired CgA processing in HF, which should be considered detrimental because CST reduces diastolic Ca2+ leak via direct CaMKIIδ inhibition. Thus, although CgA production seems to increase and probably represent a counter-regulatory mechanism in HF, this system may malfunction because of augmented myocardial CgA glycosylation.

Acknowledgments

We gratefully acknowledge the Clinical Trial Unit, Division of Medicine, Akershus University Hospital (Lorentzen, Bakkelund, and Jørgensen) for assistance with blood sampling in the HF study. We thank the Division of Medicine, Akershus University Hospital, and all participating investigators and study nurses. We also acknowledge expert animal care from the Section of Comparative Medicine, Oslo University Hospital, Ullevål.

Sources of Funding

This project was funded by the Norwegian National Association for Public Health (A.H. Ottesen), the Anders Jahre Trust for Promotion of Science (GC), and the Research Council of Norway (Drs Carlson and Omland). Drs Omland and Røsjø have received funding relating to the current work from Akershus University Hospital, the University of Oslo, the South-Eastern Norway Regional Health Authority, and the K.G. Jebsen Family Foundation. Euro Diagnostica AB (Malmö, Sweden) supported the study by performing CgA analyses free of charge. Roche Diagnostics also provided us with reagents at reduced costs. Additionally, Dr Omland received grants and personal fees from Abbott Diagnostics and AstraZeneca and personal fees from Roche Diagnostics, Abbott Diagnostics, and Novartis. Dr Røsjø has received personal fees from Novartis. The sponsors played no role in study design, data collection, management, and analysis, or preparation, review, and approval of the article.

Disclosures

None.

  • Received February 4, 2016.
  • Accepted January 11, 2017.

References

  1. 1.↵
    1. Pieroni M,
    2. Corti A,
    3. Tota B,
    4. Curnis F,
    5. Angelone T,
    6. Colombo B,
    7. Cerra MC,
    8. Bellocci F,
    9. Crea F,
    10. Maseri A
    . Myocardial production of chromogranin A in human heart: a new regulatory peptide of cardiac function. Eur Heart J. 2007;28:1117–1127. doi: 10.1093/eurheartj/ehm022.
    OpenUrlAbstract/FREE Full Text
  2. 2.↵
    1. Bartolomucci A,
    2. Possenti R,
    3. Mahata SK,
    4. Fischer-Colbrie R,
    5. Loh YP,
    6. Salton SR
    . The extended granin family: structure, function, and biomedical implications. Endocr Rev. 2011;32:755–797. doi: 10.1210/er.2010-0027.
    OpenUrlCrossRefPubMed
  3. 3.↵
    1. Røsjø H,
    2. Husberg C,
    3. Dahl MB,
    4. Stridsberg M,
    5. Sjaastad I,
    6. Finsen AV,
    7. Carlson CR,
    8. Oie E,
    9. Omland T,
    10. Christensen G
    . Chromogranin B in heart failure: a putative cardiac biomarker expressed in the failing myocardium. Circ Heart Fail. 2010;3:503–511. doi: 10.1161/CIRCHEARTFAILURE.109.867747.
    OpenUrlAbstract/FREE Full Text
  4. 4.↵
    1. Jansson AM,
    2. Røsjø H,
    3. Omland T,
    4. Karlsson T,
    5. Hartford M,
    6. Flyvbjerg A,
    7. Caidahl K
    . Prognostic value of circulating chromogranin A levels in acute coronary syndromes. Eur Heart J. 2009;30:25–32. doi: 10.1093/eurheartj/ehn513.
    OpenUrlAbstract/FREE Full Text
  5. 5.↵
    1. Omland T,
    2. Dickstein K,
    3. Syversen U
    . Association between plasma chromogranin A concentration and long-term mortality after myocardial infarction. Am J Med. 2003;114:25–30.
    OpenUrlCrossRefPubMed
  6. 6.↵
    1. Estensen ME,
    2. Hognestad A,
    3. Syversen U,
    4. Squire I,
    5. Ng L,
    6. Kjekshus J,
    7. Dickstein K,
    8. Omland T
    . Prognostic value of plasma chromogranin A levels in patients with complicated myocardial infarction. Am Heart J. 2006;152:927.e1–927.e6. doi: 10.1016/j.ahj.2006.05.008.
    OpenUrlCrossRefPubMed
  7. 7.↵
    1. Røsjø H,
    2. Masson S,
    3. Latini R,
    4. Flyvbjerg A,
    5. Milani V,
    6. La Rovere MT,
    7. Revera M,
    8. Mezzani A,
    9. Tognoni G,
    10. Tavazzi L,
    11. Omland T
    ; GISSI-HF Investigators. Prognostic value of chromogranin A in chronic heart failure: data from the GISSI-Heart Failure trial. Eur J Heart Fail. 2010;12:549–556. doi: 10.1093/eurjhf/hfq055.
    OpenUrlCrossRefPubMed
  8. 8.↵
    1. Ceconi C,
    2. Ferrari R,
    3. Bachetti T,
    4. Opasich C,
    5. Volterrani M,
    6. Colombo B,
    7. Parrinello G,
    8. Corti A
    . Chromogranin A in heart failure; a novel neurohumoral factor and a predictor for mortality. Eur Heart J. 2002;23:967–974. doi: 10.1053/euhj.2001.2977.
    OpenUrlAbstract/FREE Full Text
  9. 9.↵
    1. Dieplinger B,
    2. Gegenhuber A,
    3. Struck J,
    4. Poelz W,
    5. Langsteger W,
    6. Haltmayer M,
    7. Mueller T
    . Chromogranin A and C-terminal endothelin-1 precursor fragment add independent prognostic information to amino-terminal proBNP in patients with acute destabilized heart failure. Clin Chim Acta. 2009;400:91–96. doi: 10.1016/j.cca.2008.10.012.
    OpenUrlCrossRefPubMed
  10. 10.↵
    1. Røsjø H,
    2. Nygård S,
    3. Kaukonen KM,
    4. Karlsson S,
    5. Stridsberg M,
    6. Ruokonen E,
    7. Pettilä V,
    8. Omland T
    ; FINNSEPSIS Study Group. Prognostic value of chromogranin A in severe sepsis: data from the FINNSEPSIS study. Intensive Care Med. 2012;38:820–829. doi: 10.1007/s00134-012-2546-8.
    OpenUrlPubMed
  11. 11.↵
    1. Takiyyuddin MA,
    2. Cervenka JH,
    3. Sullivan PA,
    4. Pandian MR,
    5. Parmer RJ,
    6. Barbosa JA,
    7. O’Connor DT
    . Is physiologic sympathoadrenal catecholamine release exocytotic in humans? Circulation. 1990;81:185–195.
    OpenUrlAbstract/FREE Full Text
  12. 12.↵
    1. Angelone T,
    2. Mazza R,
    3. Cerra MC
    . Chromogranin-A: a multifaceted cardiovascular role in health and disease. Curr Med Chem. 2012;19:4042–4050.
    OpenUrlCrossRefPubMed
  13. 13.↵
    1. Zhang D,
    2. Shooshtarizadeh P,
    3. Laventie BJ,
    4. Colin DA,
    5. Chich JF,
    6. Vidic J,
    7. de Barry J,
    8. Chasserot-Golaz S,
    9. Delalande F,
    10. Van Dorsselaer A,
    11. Schneider F,
    12. Helle K,
    13. Aunis D,
    14. Prévost G,
    15. Metz-Boutigue MH
    . Two chromogranin a-derived peptides induce calcium entry in human neutrophils by calmodulin-regulated calcium independent phospholipase A2. PLoS One. 2009;4:e4501. doi: 10.1371/journal.pone.0004501.
    OpenUrlCrossRefPubMed
  14. 14.↵
    1. Bassino E,
    2. Fornero S,
    3. Gallo MP,
    4. Ramella R,
    5. Mahata SK,
    6. Tota B,
    7. Levi R,
    8. Alloatti G
    . A novel catestatin-induced antiadrenergic mechanism triggered by the endothelial PI3K-eNOS pathway in the myocardium. Cardiovasc Res. 2011;91:617–624. doi: 10.1093/cvr/cvr129.
    OpenUrlAbstract/FREE Full Text
  15. 15.↵
    1. Ottesen AH,
    2. Louch WE,
    3. Carlson CR,
    4. Landsverk OJ,
    5. Kurola J,
    6. Johansen RF,
    7. Moe MK,
    8. Aronsen JM,
    9. Høiseth AD,
    10. Jarstadmarken H,
    11. Nygård S,
    12. Bjørås M,
    13. Sjaastad I,
    14. Pettilä V,
    15. Stridsberg M,
    16. Omland T,
    17. Christensen G,
    18. Røsjø H
    . Secretoneurin is a novel prognostic cardiovascular biomarker associated with cardiomyocyte calcium handling. J Am Coll Cardiol. 2015;65:339–351. doi: 10.1016/j.jacc.2014.10.065.
    OpenUrlFREE Full Text
  16. 16.↵
    1. Anderson ME,
    2. Brown JH,
    3. Bers DM
    . CaMKII in myocardial hypertrophy and heart failure. J Mol Cell Cardiol. 2011;51:468–473. doi: 10.1016/j.yjmcc.2011.01.012.
    OpenUrlCrossRefPubMed
  17. 17.↵
    1. Røsjø H,
    2. Stridsberg M,
    3. Florholmen G,
    4. Stensløkken KO,
    5. Ottesen AH,
    6. Sjaastad I,
    7. Husberg C,
    8. Dahl MB,
    9. Øie E,
    10. Louch WE,
    11. Omland T,
    12. Christensen G
    . Secretogranin II; a protein increased in the myocardium and circulation in heart failure with cardioprotective properties. PLoS One. 2012;7:e37401. doi: 10.1371/journal.pone.0037401.
    OpenUrlCrossRefPubMed
  18. 18.↵
    1. Røsjø H,
    2. Dahl MB,
    3. Jørgensen M,
    4. Røysland R,
    5. Brynildsen J,
    6. Cataliotti A,
    7. Christensen G,
    8. Høiseth AD,
    9. Hagve TA,
    10. Omland T
    . Influence of glycosylation on diagnostic and prognostic accuracy of N-terminal pro-B-type natriuretic peptide in acute dyspnea: data from the Akershus Cardiac Examination 2 Study. Clin Chem. 2015;61:1087–1097. doi: 10.1373/clinchem.2015.239673.
    OpenUrlAbstract/FREE Full Text
  19. 19.↵
    1. Stridsberg M,
    2. Hellman U,
    3. Wilander E,
    4. Lundqvist G,
    5. Hellsing K,
    6. Oberg K
    . Fragments of chromogranin A are present in the urine of patients with carcinoid tumours: development of a specific radioimmunoassay for chromogranin A and its fragments. J Endocrinol. 1993;139:329–337.
    OpenUrlAbstract/FREE Full Text
  20. 20.↵
    1. Shan J,
    2. Betzenhauser MJ,
    3. Kushnir A,
    4. Reiken S,
    5. Meli AC,
    6. Wronska A,
    7. Dura M,
    8. Chen BX,
    9. Marks AR
    . Role of chronic ryanodine receptor phosphorylation in heart failure and β-adrenergic receptor blockade in mice. J Clin Invest. 2010;120:4375–4387. doi: 10.1172/JCI37649.
    OpenUrlCrossRefPubMed
  21. 21.↵
    1. Sag CM,
    2. Wadsack DP,
    3. Khabbazzadeh S,
    4. Abesser M,
    5. Grefe C,
    6. Neumann K,
    7. Opiela MK,
    8. Backs J,
    9. Olson EN,
    10. Brown JH,
    11. Neef S,
    12. Maier SK,
    13. Maier LS
    . Calcium/calmodulin-dependent protein kinase II contributes to cardiac arrhythmogenesis in heart failure. Circ Heart Fail. 2009;2:664–675. doi: 10.1161/CIRCHEARTFAILURE.109.865279.
    OpenUrlAbstract/FREE Full Text
  22. 22.↵
    1. Zhu D,
    2. Wang F,
    3. Yu H,
    4. Mi L,
    5. Gao W
    . Catestatin is useful in detecting patients with stage B heart failure. Biomarkers. 2011;16:691–697. doi: 10.3109/1354750X.2011.629058.
    OpenUrlPubMed
  23. 23.↵
    1. Biswas N,
    2. Curello E,
    3. O’Connor DT,
    4. Mahata SK
    . Chromogranin/secretogranin proteins in murine heart: myocardial production of chromogranin A fragment catestatin (Chga(364-384)). Cell Tissue Res. 2010;342:353–361. doi: 10.1007/s00441-010-1059-4.
    OpenUrlCrossRefPubMed
  24. 24.↵
    1. Pasqua T,
    2. Corti A,
    3. Gentile S,
    4. Pochini L,
    5. Bianco M,
    6. Metz-Boutigue MH,
    7. Cerra MC,
    8. Tota B,
    9. Angelone T
    . Full-length human chromogranin-A cardioactivity: myocardial, coronary, and stimulus-induced processing evidence in normotensive and hypertensive male rat hearts. Endocrinology. 2013;154:3353–3365. doi: 10.1210/en.2012-2210.
    OpenUrlCrossRefPubMed
  25. 25.↵
    1. Penna C,
    2. Pasqua T,
    3. Amelio D,
    4. Perrelli MG,
    5. Angotti C,
    6. Tullio F,
    7. Mahata SK,
    8. Tota B,
    9. Pagliaro P,
    10. Cerra MC,
    11. Angelone T
    . Catestatin increases the expression of anti-apoptotic and pro-angiogenetic factors in the post-ischemic hypertrophied heart of SHR. PLoS One. 2014;9:e102536. doi: 10.1371/journal.pone.0102536.
    OpenUrl
  26. 26.↵
    1. Mazza R,
    2. Gattuso A,
    3. Mannarino C,
    4. Brar BK,
    5. Barbieri SF,
    6. Tota B,
    7. Mahata SK
    . Catestatin (chromogranin A344-364) is a novel cardiosuppressive agent: inhibition of isoproterenol and endothelin signaling in the frog heart. Am J Physiol Heart Circ Physiol. 2008;295:H113–H122. doi: 10.1152/ajpheart.00172.2008.
    OpenUrlAbstract/FREE Full Text
  27. 27.↵
    1. Eskeland NL,
    2. Zhou A,
    3. Dinh TQ,
    4. Wu H,
    5. Parmer RJ,
    6. Mains RE,
    7. O’Connor DT
    . Chromogranin A processing and secretion: specific role of endogenous and exogenous prohormone convertases in the regulated secretory pathway. J Clin Invest. 1996;98:148–156. doi: 10.1172/JCI118760.
    OpenUrlCrossRefPubMed
  28. 28.↵
    1. Shah AM,
    2. Mann DL
    . In search of new therapeutic targets and strategies for heart failure: recent advances in basic science. Lancet. 2011;378:704–712. doi: 10.1016/S0140-6736(11)60894-5.
    OpenUrlCrossRefPubMed
  29. 29.↵
    1. Yang E,
    2. Schulman H
    . Structural examination of autoregulation of multifunctional calcium/calmodulin-dependent protein kinase II. J Biol Chem. 1999;274:26199–26208.
    OpenUrlAbstract/FREE Full Text
  30. 30.↵
    1. Huke S,
    2. Desantiago J,
    3. Kaetzel MA,
    4. Mishra S,
    5. Brown JH,
    6. Dedman JR,
    7. Bers DM
    . SR-targeted CaMKII inhibition improves SR Ca2+ handling, but accelerates cardiac remodeling in mice overexpressing CaMKIIδC. J Mol Cell Cardiol. 2011;50:230–238. doi: 10.1016/j.yjmcc.2010.10.014.
    OpenUrlCrossRefPubMed
  31. 31.↵
    1. Guo T,
    2. Zhang T,
    3. Ginsburg KS,
    4. Mishra S,
    5. Brown JH,
    6. Bers DM
    . CaMKIIδC slows [Ca]i decline in cardiac myocytes by promoting Ca sparks. Biophys J. 2012;102:2461–2470. doi: 10.1016/j.bpj.2012.04.015.
    OpenUrlCrossRefPubMed
  32. 32.↵
    1. DeSantiago J,
    2. Maier LS,
    3. Bers DM
    . Frequency-dependent acceleration of relaxation in the heart depends on CaMKII, but not phospholamban. J Mol Cell Cardiol. 2002;34:975–984.
    OpenUrlCrossRefPubMed
  33. 33.↵
    1. Angelone T,
    2. Quintieri AM,
    3. Brar BK,
    4. Limchaiyawat PT,
    5. Tota B,
    6. Mahata SK,
    7. Cerra MC
    . The antihypertensive chromogranin a peptide catestatin acts as a novel endocrine/paracrine modulator of cardiac inotropism and lusitropism. Endocrinology. 2008;149:4780–4793. doi: 10.1210/en.2008-0318.
    OpenUrlCrossRefPubMed

CLINICAL PERSPECTIVE

Paracrine factors and hormones are important biomarkers and key targets for therapy in heart failure (HF). Chromogranin A (CgA) is a 48-kDa prohormone that is produced in many tissues throughout the body, including in neuroendocrine and myocardial cells. Circulating CgA levels have previously been found to be associated with mortality in patients with myocardial dysfunction, but currently no information is available regarding the processing of CgA to shorter, functionally active CgA fragments during HF development. Whether the CgA fragment catestatin (CST) may directly influence cardiomyocyte function is also not known. In this work, we provide the first characterization of CgA processing in HF and demonstrate that CST directly modulates cardiomyocyte Ca2+ handling via direct CaMKIIδ (Ca2+/calmodulin-dependent protein kinase IIδ) inhibition. We also found myocardial CgA-to-CST conversion to be impaired because of hyperglycosylation in HF individuals, and low CgA-to-CST conversion was associated with increased mortality in patients hospitalized with acute HF. In contrast, we found no evidence of myocardial hyperglycosylation in subjects free from HF or impaired CgA-to-CST processing in patients hospitalized with acute exacerbation of chronic obstructive pulmonary disease. Accordingly, our clinical and experimental data link CgA and CST directly to important cardiovascular pathophysiology, and the primary role of CST may be as part of a paracrine/autocrine protective feedback loop that could antagonize overactive myocardial CaMKIIδ signaling. However, our data also indicate that this system may be malfunctioning in HF because of myocardial CgA hyperglycosylation; this should be explored in additional experimental and clinical studies.

View Abstract

This Issue

Jump to

Article Tools

Share this Article

  • Email
  • Glycosylated Chromogranin A in Heart FailureCLINICAL PERSPECTIVE
    Anett Hellebø Ottesen, Cathrine R. Carlson, William E. Louch, Mai Britt Dahl, Ragnhild A. Sandbu, Rune Forstrøm Johansen, Hilde Jarstadmarken, Magnar Bjørås, Arne Didrik Høiseth, Jon Brynildsen, Ivar Sjaastad, Mats Stridsberg, Torbjørn Omland, Geir Christensen and Helge Røsjø
    Circulation: Heart Failure. 2017;10:e003675, originally published February 16, 2017
    https://doi.org/10.1161/CIRCHEARTFAILURE.116.003675
    Permalink:
    del.icio.us logo Digg logo Reddit logo Twitter logo CiteULike logo Facebook logo Google logo Mendeley logo

Related Articles

Cited By...

Subjects