A phenotype-based forward genetic screen identifies Dnajb6 as a sick sinus syndrome gene

Curation statements for this article:
  • Curated by eLife

    eLife logo

    Evaluation Summary:

    This manuscript is of interest to cardiovascular and developmental biologists as it describes Dnajb6 as a novel gene linked with Sick Sinus Syndrome. The claims are mostly supported by observations using zebrafish dnajb6b trap line and Dnajb6 heterozygous mouse models. However, the paper would be strengthened be clarification of some experimental aspects and a discussion of the potential connection of DNAJB6 to the Sick Sinus Syndrome in humans.

    (This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. Reviewer #1 and Reviewer #2 agreed to share their name with the authors.)

This article has been Reviewed by the following groups

Read the full article See related articles

Abstract

Previously we showed the generation of a protein trap library made with the gene-break transposon (GBT) in zebrafish ( Danio rerio ) that could be used to facilitate novel functional genome annotation towards understanding molecular underpinnings of human diseases (Ichino et al, 2020). Here, we report a significant application of this library for discovering essential genes for heart rhythm disorders such as sick sinus syndrome (SSS). SSS is a group of heart rhythm disorders caused by malfunction of the sinus node, the heart’s primary pacemaker. Partially owing to its aging-associated phenotypic manifestation and low expressivity, molecular mechanisms of SSS remain difficult to decipher. From 609 GBT lines screened, we generated a collection of 35 zebrafish insertional cardiac (ZIC) mutants in which each mutant traps a gene with cardiac expression. We further employed electrocardiographic measurements to screen these 35 ZIC lines and identified three GBT mutants with SSS-like phenotypes. More detailed functional studies on one of the arrhythmogenic mutants, GBT411 , in both zebrafish and mouse models unveiled Dnajb6 as a novel SSS causative gene with a unique expression pattern within the subpopulation of sinus node pacemaker cells that partially overlaps with the expression of hyperpolarization activated cyclic nucleotide gated channel 4 (HCN4), supporting heterogeneity of the cardiac pacemaker cells.

Article activity feed

  1. Author Response

    Reviewer #1 (Public Review):

    The authors took advantage of an existing protein-trap resource in zebrafish to identify genes important for normal pacemaker function in adults. They generated a collection of lines with mutation in genes that expressed at reasonably high levels in the heart and assess their ECG. They identified 3 candidates with increased incidence of sinus arrest and focused on validation of dnajb6b. The dnjb6b mutant fish display other defects including enhanced response to atropine and carbacol and bradycardia. They show that dnajb6b is expressed in a subset of cells in the sinus node in zebrafish. In mouse sinus node, DNAJB6 expressing cells have low expression of TBX3 and its target HCN4. In addition, Dnajb6b+/- mice also display similar phenotypes. Analysis of pacemaker function in ex vivo mouse hearts by high-resolution fluorescent optical mapping of action potentials revealed that the number of leading pacemakers in Dnajb6b+/- hearts is decreased in the sinus node, with a concomitant increase in the auxiliary pacemakers. RNAseq analysis of the right atrial tissues detected expression changes in ion channels and genes involved in Ca2+ handling and Wnt signaling. Overall, the results support the conclusion that DNAJB6 is important for proper sinus node function, thus adding it to the short list of sick sinus syndrome genes. However, the manuscript has several weaknesses.

    Weakness:

    The manuscript does not address the mechanism by which decreased DNAJB6B causes sick sinus syndrome. For example, it is unknown if DNAJB6B functions cell autonomously or non-cell autonomously in the sinus node. The RNAseq analysis identified changes in ion channels in the right atrial tissues of 1-year old mice, cellular electrophysiology of the sinus node cells was not assessed.

    The main goal of this research is to prove the feasibility of discovering novel SSS genes in adults via a forward genetic approach in zebrafish. Thus, the major hallmark would be to prove causality and specificity of the candidate genes identified from this screen, such as Dnajb6. Comprehensive mechanistic study would be a focus for future studies.

    Nevertheless, we carried out the following experiments to address the mechanisms. Based on these data, a new section was added to the discussion section (Lines 424-465).

    (1) In mice, we did more antibody immunostaining and confirmed a negative correlation in terms of expression intensity between the Dnajb6 and Tbx3 proteins. We further detected a significantly increased Tbx3 immunostaining signal in the SAN tissues of Dnajb6 heterozygous mice compared to WT controls (new Figure 3D-F).

    (2) In zebrafish, we compared expression patterns of the sqET33-mi59B conduction system reporter line between the GBT411/dnajb6b heterozygous and homozygous mutants. We found the atrio-ventricular canal (AVC) signal became diffused in GBT411/dnajb6b homozygous adult hearts. In addition, the ring-like structure usually seen in the SAN region of WT controls and in the GBT411/dnajb6 heterozygous was largely lost in 3 out of 9 GBT411/dnajb6b homozygous adult hearts examined (new Figure 2).

    Together with the ectopic pacemaker activity detected in the Dnajb6 heterozygous mice (new Figure 5A and 5B), we speculate that Dnajb6 might act as a suppressor of Tbx3 transcription factor in defining cell fate specification into SAN pacemaker myocytes. Since Tbx3 was reported to suppress chamber myocardial differentiation (Mommersteeg et al., Circ Res. 2007;100(3):354-62), upregulation of Tbx3 may thus contribute to enhanced atrial ectopic activity in Dnajb6 heterozygous mice.

    Furthermore, TBX3 has been recently identified as a component of the Wnt/β-catenin-dependent transcriptional complex (Zimmerli et al., eLife. 2020;9:e58123), which is significantly affected in Dnajb6 heterozygous mice (see new Figure 7B-C). This further supports a possible role of TBX3 in both SAN and atrial remodeling.

    (3) Finally, in collaboration with Drs. Grandi, Morotti, and Ni from University of California Davis, we utilized a population-based computational modeling approach to determine the cellular/ionic mechanisms that could underlie the ex vivo observed SSS phenotype in the Dnajb6 heterozygous mice (new Figure 6). We used our previously published model of the mouse SAN myocyte (Morotti et al. Int J Mol Sci. 2021; 22(11):5645) and enhanced it with addition of both sympathetic and parasympathetic stimulations to model the effects of isoproterenol- and carbachol-induced changes in pacemaker activity (i.e., firing rate), respectively. We generated a population of 10,000 mouse SAN myocyte models by random modification of selected model parameters describing maximum ion channel conductances and ion transport rates from the baseline model and assessed isoproterenol- and carbachol-induced effects on each model variant. We then separated this population of models in two subpopulations representing the WT and Dnajb6+/- mice phenotypes: namely, we extracted the model variants that recapitulate changes observed in Dnajb6+/- vs. WT mice, including a reduced firing rate at baseline, an increased response to isoproterenol, and a decreased response to carbachol administration (new Figure 6). This filtering process resulted in n=438 models that correspond to the Dnajb6+/- mice phenotype and n=6,995 models that correspond to the WT phenotype. We analyzed the parameter value differences in these two subgroups to revealed several crucial parameters that are significantly correlated with the observed electrophysiological changes. The analysis revealed a significant decrease in the maximal conductances of the fast (Nav1.5) sodium current, the L-type Ca2+ current (ICa,L), the transient outward, sustained, and acetylcholine-activated K+ currents, the background Na+ and Ca2+ currents, as well as the ryanodine receptor maximal release flux of the Dnajb6+/- vs. WT model variants. We also found a significant increase in the Na+/Ca2+ exchanger (NCX) maximal transport rate, and conductances of the T-type Ca2+ current and the slowly-activating delayed rectifier K+ current. These new studies provide some novel mechanistic insights into the observed SSS phenotype in Dnajb6+/- mice. Importantly, these new in silico experiments add another conceptual level to the phenotype-based screening approach introduced in the current study to identify new genetic factors associated with SAN dysfunction. Direct testing of these mechanisms would require a substantial amount of single SAN cell patch clamp and confocal microscopy experiments which are out of scope of the current manuscript and will be pursued in a follow-up study.

    The manuscript does not address why the zebrafish homozygous mutants are adult viable while the mouse homozygotes are embryonic lethal. The insertion of the GBT411 disrupt dnajb6b(L) but not dnajb6b(S), while the mouse mutation deletes the entire gene. Does this difference partially explain the difference?

    Indeed, the difference between zebrafish and mouse can be partially explained by the fact that only the long isoform of dnajb6b gene, dnajb6b(L), was disrupted in the GBT411 mutant, while both the long-Dnajb6(L) and short-Dnajb6(S) isoforms of Dnajb6 gene was largely deleted in the Dnajb6 knockout mice. However, we think the main reason is probably that functional redundancy in zebrafish but not mouse: zebrafish has two dnajb6 homologues, dnajb6b and dnajb6a, while mouse has only one Dnajb6 homologue. We added these points to the paper (Lines 377-379).

    Reviewer #2 (Public Review):

    In this manuscript, the authors expand upon previous work describing development of a protein trap library made with the gene-break transposon. This library was screened to identify lines displaying gene trap expression in the heart (zebrafish insertional cardiac mutant collection). A pilot screen of these lines using adult ECG phenotypes identifies dnajb6b as a new gene important for cardiac rhythm. Using the GBT/dnajb6b zebrafish line, Ding et al. find a proportion of aged homozygous mutant fish (1.5-2 years) present sinus arrest episodes and reduced heart rate. Treating GBT411/dnajb6b mutant adults with compounds revealed aberrant responses to autonomic stimuli, and sinus arrest episodes were induced following verapamil exposure, providing evidence that GBT411/dnajb6b as an arrhythmia mutant. This conclusion could be better supported by presenting specific ECG parameters to characterize the conduction defect more thoroughly. The authors then report that Dnajb6+/- adult mice recapitulate some of the phenotypes observed in zebrafish, including sinus arrest and AV blocks, as well as impaired (although different) responses to autonomic stimuli. The authors describe that these are features of sick sinus syndrome in the absence of cardiomyopathy phenotypes in either the zebrafish or mouse lines. However, overall cardiac morphology is not well described for either the GBT411/dnajb6b or Dnajb6+/- models.

    We carried out more experiments to examine left ventricular (LV) structure in Dnajb6 heterozygous mice at 1 year of age, using H&E staining, Masson’s trichrome staining, and transmission electron microscopy (TEM) analysis. We now show clearly that there are no significant myocardium structural changes in the LV as well as atrial and SAN tissues of Dnajb6 heterozygous mice (new Supplemental Figures 3 and 5), when the SSS phenotype was already noticeable. However, in the GBT411/dnajb6b heterozygous mutant at ~2 years of age, we detected severe sarcomere structural abnormality in 1 out of 3 fish hearts examined (see Response-only Figure 1). In addition, in a previous publication (Ding et al., Circ Res, 2013:112(40:606-17), we reported evident cardiac remodeling phenotypes in the GBT411/dnajb6b homozygous fish at 12 months of age.

    Together, we have obtained more experimental evidence to strengthen the claim that arrhythmia is not due to cardiomyopathy/structural remodeling in the Dnajb6+/- mice. However, the evidence from fish remains weak. Therefore, we removed the claim that “when structural remodeling/cardiac dysfunction have not yet occurred” in fish and modified our statement in mice accordingly (Lines 372-377, 385-386).

    To further support a role for Dnajb6 in sinoatrial node dysfunction, the authors performed optical mapping of action potentials from isolated mouse atrial tissue. These data reveal that Dnajb6+/- cultures exhibit ectopic pacemakers outside of the sinoatrial node, including within the atrial wall and inter-atrial septum. These data also show prolongation of SAN recovery time at baseline and following autonomic stimulation, further suggesting SAN dysfunction. RNA-sequencing experiments of DNAjb6+/- adult right atrial tissue showed differentially expressed genes encoding Ca2+ handling related proteins, ion channels, and WNT pathway related proteins. As these genes are involved in the cardiac conduction system, the authors suggest these pathways as molecular mechanisms underlying SSS phenotypes in Dnajb6 models.

    Sick sinus syndrome is a relatively rare arrhythmia most commonly found in older populations. Therefore, it has been challenging to establish clinically relevant models and there is a limited understanding of mechanisms of SSS pathogenesis. One particular strength of this manuscript is the ECG phenotype-based forward screen of the gene-breaking transposon (GBT)-based gene trap library in aged animals. This pilot study provides proof-of-concept that this screening approach is well suited to identify regulators of cardiac function in adults and genes linked to adult diseases like SSS.

    Thank you very much for recognizing the major strength of our manuscript!

  2. Evaluation Summary:

    This manuscript is of interest to cardiovascular and developmental biologists as it describes Dnajb6 as a novel gene linked with Sick Sinus Syndrome. The claims are mostly supported by observations using zebrafish dnajb6b trap line and Dnajb6 heterozygous mouse models. However, the paper would be strengthened be clarification of some experimental aspects and a discussion of the potential connection of DNAJB6 to the Sick Sinus Syndrome in humans.

    (This preprint has been reviewed by eLife. We include the public reviews from the reviewers here; the authors also receive private feedback with suggested changes to the manuscript. Reviewer #1 and Reviewer #2 agreed to share their name with the authors.)

  3. Reviewer #1 (Public Review):

    The authors took advantage of an existing protein-trap resource in zebrafish to identify genes important for normal pacemaker function in adults. They generated a collection of lines with mutation in genes that expressed at reasonably high levels in the heart and assess their ECG. They identified 3 candidates with increased incidence of sinus arrest and focused on validation of dnajb6b. The dnjb6b mutant fish display other defects including enhanced response to atropine and carbacol and bradycardia. They show that dnajb6b is expressed in a subset of cells in the sinus node in zebrafish. In mouse sinus node, DNAJB6 expressing cells have low expression of TBX3 and its target HCN4. In addition, Dnajb6b+/- mice also display similar phenotypes. Analysis of pacemaker function in ex vivo mouse hearts by high-resolution fluorescent optical mapping of action potentials revealed that the number of leading pacemakers in Dnajb6b+/- hearts is decreased in the sinus node, with a concomitant increase in the auxiliary pacemakers. RNAseq analysis of the right atrial tissues detected expression changes in ion channels and genes involved in Ca2+ handling and Wnt signaling. Overall, the results support the conclusion that DNAJB6 is important for proper sinus node function, thus adding it to the short list of sick sinus syndrome genes. However, the manuscript has several weaknesses.

    Weakness
    The manuscript does not address the mechanism by which decreased DNAJB6B causes sick sinus syndrome. For example, it is unknown if DNAJB6B functions cell autonomously or non-cell autonomously in the sinus node. The RNAseq analysis identified changes in ion channels in the right atrial tissues of 1-year old mice, cellular electrophysiology of the sinus node cells was not assessed.

    The manuscript does not address why the zebrafish homozygous mutants are adult viable while the mouse homozygotes are embryonic lethal. The insertion of the GBT411 disrupt dnajb6b(L) but not dnajb6b(S), while the mouse mutation deletes the entire gene. Does this difference partially explain the difference?

  4. Reviewer #2 (Public Review):

    In this manuscript, the authors expand upon previous work describing development of a protein trap library made with the gene-break transposon. This library was screened to identify lines displaying gene trap expression in the heart (zebrafish insertional cardiac mutant collection). A pilot screen of these lines using adult ECG phenotypes identifies dnajb6b as a new gene important for cardiac rhythm. Using the GBT/dnajb6b zebrafish line, Ding et al. find a proportion of aged homozygous mutant fish (1.5-2 years) present sinus arrest episodes and reduced heart rate. Treating GBT411/dnajb6b mutant adults with compounds revealed aberrant responses to autonomic stimuli, and sinus arrest episodes were induced following verapamil exposure, providing evidence that GBT411/dnajb6b as an arrhythmia mutant. This conclusion could be better supported by presenting specific ECG parameters to characterize the conduction defect more thoroughly. The authors then report that Dnajb6+/- adult mice recapitulate some of the phenotypes observed in zebrafish, including sinus arrest and AV blocks, as well as impaired (although different) responses to autonomic stimuli. The authors describe that these are features of sick sinus syndrome in the absence of cardiomyopathy phenotypes in either the zebrafish or mouse lines. However, overall cardiac morphology is not well described for either the GBT411/dnajb6b or Dnajb6+/- models.

    To further support a role for Dnajb6 in sinoatrial node dysfunction, the authors performed optical mapping of action potentials from isolated mouse atrial tissue. These data reveal that Dnajb6+/- cultures exhibit ectopic pacemakers outside of the sinoatrial node, including within the atrial wall and inter-atrial septum. These data also show prolongation of SAN recovery time at baseline and following autonomic stimulation, further suggesting SAN dysfunction. RNA-sequencing experiments of DNAjb6+/- adult right atrial tissue showed differentially expressed genes encoding Ca2+ handling related proteins, ion channels, and WNT pathway related proteins. As these genes are involved in the cardiac conduction system, the authors suggest these pathways as molecular mechanisms underlying SSS phenotypes in Dnajb6 models.

    Sick sinus syndrome is a relatively rare arrhythmia most commonly found in older populations. Therefore, it has been challenging to establish clinically relevant models and there is a limited understanding of mechanisms of SSS pathogenesis. One particular strength of this manuscript is the ECG phenotype-based forward screen of the gene-breaking transposon (GBT)-based gene trap library in aged animals. This pilot study provides proof-of-concept that this screening approach is well suited to identify regulators of cardiac function in adults and genes linked to adult diseases like SSS.

  5. Reviewer #3 (Public Review):

    Sick Sinus Syndrome (SSS) is a heart rhythm disorder that often occurs in the older population. The underpinning mechanism of SSS is largely unknown. The Xu lab has previously reported the zebrafish insertion cardiac (ZIC) mutant collection that consists of 35 gene trap lines with cardiac expression. In this manuscript, the authors employed electrocardiography to screen the ZIC collection and found that loss of function of cyth3a, vapal and dnajb6b led to cardiac defects resembling SSS. They then performed more in-depth analyses to examine how dnajb6 loss would impact the heart in both zebrafish and mouse models. The authors made several interesting observations. The authors showed that cardiac cells with high HCN4 and Tbx3 expression have relatively low dnajb6 expression, suggesting differential transcriptional regulations of these genes in the heart. Electrophysiology and optical mapping analyses revealed potential defects in pacemaker activities in the absence of dnajb6 and bulk RNA-seq analysis detected altered expression of multiple ion channels and wnt signaling genes. While these data support a link between dnajb6 and SSS, the manuscript would be greatly strengthened with additional quantitative measurements and mechanistic studies.