Fishing for the genetic basis of cardiovascular disease

With a frequency of nearly 1%, heart defects are the most common congenital anomalies in humans. Siblings have a high probability of sharing congenital heart disease, indicating a genetic contribution, but in most cases there is no Mendelian inheritance. The zebrafish offers the possibility to dissect pathways controlling heart development; components of these pathways can then be analyzed in human patients.

For example, the zebrafish mutant heart-strings (hst) is caused by a mutation in the transcription factor tbx5, which leads to abnormalities in cardiac differentiation and fin development and, when mutated in humans, causes Holt-Oram syndrome with cardiac septation defects and limb abnormalities (Garrity et al., 2002; Mori and Bruneau, 2004). This shows that even though zebrafish do not have an atrial or ventricular septum, both of which are affected in the human disease, the respective gene still fulfills essential functions in zebrafish heart development.

slipjig (sli) mutants harbor a mutation in the zebrafish homologue of the transcription factor Foxn4, and display structural atrioventricular canal malformation accompanied by atrioventricular conduction defects. Foxn4 is expressed in the atrioventricular canal and binds to a tbx2 enhancer domain to drive transcription of tbx2b in the atrioventricular canal (Chi et al., 2008). Atrioventricular canal defects are rather frequent in humans but, genetically, are poorly understood. Future studies will show whether Foxn4 also plays a role in human atrioventricular canal defects.

One particularly interesting zebrafish mutant is gridlock (grl), which carries a mutation in the hey2 gene leading to impaired blood flow in the tail owing to arterial blockade in the anterior trunk. This makes it a suitable model for the human congenital disease aortic coarctation, which is characterized by severe stenosis or obliteration of the distal aortic arch around the insertion of the ductus arteriosus (Weinstein et al., 1995). Most importantly, grl was the first zebrafish mutant to be successfully used in an in vivo small-molecule screen. In this screen, a class of compounds was identified that suppress the coarctation phenotype by upregulation of vascular endothelial growth factor (VEGF) (Peterson et al., 2004). This approach offers tremendous opportunities for future drug development to treat human heart disease.

Cardiac muscle cells in humans are post-mitotic and cannot be regenerated adequately if lost through events such as myocardial infarction. A growing field in cardiology focuses on the regeneration of functional cardiac myocytes from cardiac stem cells through transdifferentiation, or by inducing cardiac muscle cells to re-enter the cell cycle. Interestingly, zebrafish hearts are able to regenerate after injuries (Poss et al., 2002). The zebrafish mutant island beat (isl), which harbors a mutation in the alpha1 C L-type calcium channel subunit (C-LTCC) leading to a silent ventricle as a result of impaired calcium signaling, fails to add an adequate number of cells to the cardiac ventricle, causing the ventricles to be hypoplastic (Rottbauer et al., 2001). Another pathway controlling cardiac growth was discovered with the help of the zebrafish mutant liebeskummer (lik). Reptin, a transcriptional co-repressor of β-catenin-dependent transcription, is activated in lik mutants leading to augmentation of cardiac myocyte numbers, so-called cardiac hyperplasia. Reptin signaling is antagonized by the transcriptional cofactor pontin, which regulates β-catenin signaling in cardiomyocytes (Rottbauer et al., 2002). Aside from the coarctation phenotype described above, grl mutants have also been shown to develop hyperplastic hearts. Gridlock, a basic helix-loop-helix transcription factor, forms a complex with the cardiac transcription factor Gata5 and represses Gata5 transcriptional activity leading to inhibition of myocardial proliferation (Jia et al., 2007). By contrast, the hearts of the zebrafish mutants heart of glass (heg), santa (san) and valentine (val) do not show concentric growth because they fail to add additional myocardial cell layers, even though the total cell number is not altered. As a result, these mutants have hypocontractile, monolayered, giant cardiac ventricles (Mably et al., 2003; Mably et al., 2006). Further dissection of these pathways could allow us to reverse the final exit of cardiac myocytes from the cell cycle and help us to engineer functional neo-myocardium.

Human cardiomyopathies are diseases that primarily affect the myocardium. The two most prevalent forms are dilated cardiomyothapy (DCM) and hypertrophic cardiomyopathy (HCM). Dilated cardiomyopathy is a condition that, in humans, is characterized by decreasing cardiac pump function, leading to congestive heart failure. HCM is defined as hypertrophy of the myocardium in the absence of any other diseases that cause myocardial hypertrophy, such as high blood pressure or storage diseases. Whereas HCM is caused exclusively by genetic mutations, DCM is considered to be multifactorial with a strong genetic influence. Several zebrafish mutants have been identified that display heart phenotypes resembling the features of the human disease.

In the zebrafish mutant dead beat (ded), cardiac contractility initially resembles that of wild-type fish, but then decreases from 48-60 hpf. The gene responsible for this phenotype is the zebrafish homologue of phopholipase C gamma 1 (PLCG1). PLCG1 is known to associate with the VEGF-receptor Flt-1 in endothelial cells. Interestingly, using the zebrafish mutant ded, this pathway is now known to be crucial for cardiac pump function in cardiac myocytes in a cell-autonomous fashion. Conservation of this pathway throughout evolution into mammals was confirmed using cultured rat cardiac myocytes (Rottbauer et al., 2005).

One mechanism allowing for the development of DCM is defective cardiac stretch sensing (Knoll et al., 2002). The zebrafish mutant main squeeze (msq) is characterized by a loss of cardiac contractility from 60-72 hpf. Positional cloning identified a mis-sense mutation in the homologue of the integrin-linked kinase (ilk) gene, leading to loss of kinase activity and failure to bind to the ILK-associated protein β-parvin/affixin. Loss of contractility in msq mutant zebrafish is preceded by a loss of atrial natriuretic factor (anf) expression, a gene that is known to be expressed in a stretch-dependent manner. This led to the hypothesis that ILK might be part of a potential cardiac stretch sensor that recognizes an increase in preload by augmented stretching of the heart, and then transmits these signals to the contractile apparatus in order to increase contractility (Bendig et al., 2006). Indeed, based on the msq findings, ILK mutations have been identified in human patients suffering from DCM, demonstrating the usefulness of the unbiased, zebrafish forward genetics approach for the identification of truly novel candidate genes for human disease (Knoll et al., 2007; Dahme et al., 2008).

Several zebrafish mutants have been identified that never establish normal ventricular contractility and, when analyzed ultrastructurally, do not form cardiac sarcomeres. pickwick (pik) heart failure mutants harbor a mutation in a cardiac-specific exon of the gene encoding the giant filament protein titin. Mutations in the cardiac-specific exon of titin are also found in human patients suffering from DCM (Gerull et al., 2002; Xu et al., 2002). The mutant tell tale heart (tel), which has a similar phenotype to the pik mutant, also fails to assemble functional cardiac sarcomeres and has been shown to be caused by a mutation in the cardiac regulatory myosin light chain gene mlc2 (Rottbauer et al., 2006). Mutations in the human gene encoding ventricular MLC-2 (MYL2) are found in patients suffering from HCM, a condition were cardiac myocytes are enlarged leading to thickening of the cardiac ventricular walls. Indeed, HCM is considered to be mainly a disease of inefficient sarcomerogenesis, because most mutations found in human HCM patients affect genes encoding sarcomeric proteins. silent heart (sil) is another example of a mutant that displays ventricular acontractility owing to impaired myofibrillogenesis; sil harbors a mutation in the cardiac troponin T gene (Sehnert et al., 2002). Cardiac Troponin T (TNNT2) mutations are also a major cause of HCM in humans.

Arrhythmias are defined as diseases with an abnormal heart rhythm. A major limitation for arrhythmia research, in vivo, is the difference in the molecular underpinnings of the cardiac action potential between humans and classic model organisms, such as mice. Mice have a heart rate of 300-600 beats per minute (bpm), whereas human hearts have a rate of 60-100 bpm. As a consequence, much of the heart cycle, especially repolarization – meaning the reestablishment of resting conditions, is accomplished by a completely different set of ion channels in humans compared with in mice. Interestingly, zebrafish have a heart rate of 80-160 bpm and repolarization is achieved by similar mechanisms as in humans. As zebrafish embryos are transparent, arrhythmias can be detected by simple observation of the heartbeat. For example, atrioventricular block, where only every second or third atrial contraction is conducted to the ventricle or where atrioventricular conduction is completely blocked, can be observed in the embryos. Another common arrhythmia phenotype, and the most common arrhythmia in humans, is atrial fibrillation, which is characterized by uncoordinated contraction of atrial cardiac myocytes.

Two zebrafish mutants have been isolated that harbor mutations in the main repolarizing ion channel gene kcnh2 (erg). A mutant with complete loss of kcnh2 function displays complete atrioventricular block and ventricular asystole (Arnaout et al., 2007). This mutant has been proposed as a model for human long QT syndrome, a repolarization disorder associated with sudden cardiac death. By contrast, reggae (reg) mutants have a kcnh2 gain-of-function mutation leading to accelerated repolarization and, thus, can be considered a model for human short QT syndrome (Hassel et al., 2008). Interestingly, reg mutants present with paroxysmal atrial fibrillation, a condition that has been found in human short QT syndrome (Schimpf et al., 2008). Slow mo (smo) is a mutant that is characterized by bradycardia (a slow heart rate). The underlying mechanism is severe reduction in the cardiac pacemaker ‘funny current’, I_f (Baker et al., 1997; Warren et al., 2001). Interestingly, an I_f-channel blocker (ivabradin), which leads to a reduction of the I_f-current, has been introduced recently into clinical practice to treat conditions where tachycardia (a fast heart rate) is undesirable. Interestingly, unlike most zebrafish mutants, reg and smo mutants can survive into adulthood allowing additional experimental approaches to be carried out in these zebrafish, such as ECG recordings, isolation of cardiomyocytes and determination of action potentials – all of which are more easily achieved in adult zebrafish than in embryonal zebrafish. Another zebrafish arrhythmia mutant is tremblor (tre), which harbors a mutation in the gene that encodes the cardiac-specific sodium-calcium-exchanger (NCX) and leads to defective calcium extrusion during diastole and a subsequent intracellular calcium overload and atrial fibrillation (Ebert et al., 2005; Langenbacher et al., 2005). Zebrafish island beat (isl) mutants have defective L-type calcium currents which, aside from the reduced number of ventricular cardiac myocytes, also leads to atrial fibrillation (Rottbauer et al., 2001). Zebrafish mutants with atrial fibrillation have mutations in various genes that are involved in ion transport and, therefore, recapitulate the complexity of human disease, where mutations in multiple genes are known to contribute to atrial fibrillation (Tsai et al., 2008).

Several hundred cardiac mutants have been identified through various zebrafish mutagenesis screens. The saturation mutagenesis approach, which proposes that every gene of the zebrafish genome should be altered for analysis, suggests that the genetic basis of cardiovascular function could be completely dissected in a simplified system. As discussed here, many of the zebrafish gene programs are conserved between species and the essential genes are also candidates for human CVD. Eventually, cardiac genes that have been recognized in zebrafish will be analyzed, either by direct sequencing or by single nucleotide polymorphism (SNP)-based whole genome analysis. Small molecule screens and similar techniques should provide a detailed molecular understanding of the underlying pathophysiology of human CVD and help in the development of novel therapies. So, let’s keep fishing for novel cardiovascular disease genes!