|Home | About | Journals | Submit | Contact Us | Français|
The GATA transcription factor encoded by pannier (pnr) is a critical regulator of heart progenitor formation in Drosophila. Mutations in GATA4, the mammalian homolog of pnr, have also been implicated in causing human cardiac disease in a haploinsufficient manner. Mouse models of Gata4 loss-of-function and gain-of-function studies underscored the importance of Gata4 in regulating cardiac progenitor cells specification and differentiation. However, it is not known whether pnr/Gata4 is directly involved in establishing and maintaining adult heart physiology because of the lethality associated with defective heart function and redundancy among various GATA factors in vertebrates. Here, we took advantage of the Drosophila heart model to examine the function of pnr in adult heart physiology. We found that pnr heterozygous mutants result in defective cardiac performance in response to electrical pacing of the heart as well as in elevated arrhythmias. Adult-specific disruption of pnr function using a dominant-negative form pnrEnR revealed a cardiac autonomous requirement of pnr in regulating heart physiology. Moreover, we identified Tbx20/neuromancer (nmr) as a potential downstream mediator of pnr in regulating cardiac performance and rhythm regularity, based on the observation that overexpression of nmr genes, but not of tinman, partially rescues the adult defects in pnr mutants. We conclude that pnr is not only essential for early cardiac progenitor formation, along with tinman and T-box factors, but also plays an important role in establishing and/or maintaining proper heart function, which is partially through another key regulator Tbx20/nmr.
Members of the GATA family of transcription factors are among the first to be expressed in the developing heart. The family comprises six members, characterized by a highly conserved DNA binding domain consisting of two zinc fingers that directly bind to DNA, a potent transcriptional activation domain and domains that mediate interaction with cofactors (reviewed in 1–3). On the basis of sequence similarity and expression profiles, the GATA family can be divided into two subfamilies: GATA-1, -2, -3 that are prominently expressed in hematopoietic cells and GATA-4, -5, -6 that are mainly expressed in mesodermal and endodermal derivatives (4). One of these family members, GATA4, has been extensively characterized as an essential regulator of cardiac development and differentiation (2,5–7). A mouse model using tetraploid embryo complementation shows that Gata4−/− embryos exhibit hypoplastic ventricles and a loss of the proepicardium, which was confirmed by a Cre-loxP-based approach using Nkx2-5-Cre (8–10). Clinical studies have shown an association between GATA4 and human congenital heart disease (CHD) (11,12). Specifically, a heterozygous missense mutation adjacent to the C-terminal zinc finger of GATA4, which is required for DNA binding and interaction with cofactors, was identified in family members with CHD, especially cases with atrial septal defects (12).
GATA4 is also expressed in the adult heart. In addition to its role in maintaining differentiation gene expression, GATA4 also induces gene expression in response to hypertrophic stimuli (2,9,13–15). Overexpression of GATA4 in vitro in cell culture or in vivo using transgenic mice induced cardiomyocyte hypertrophy (9,14). Conversely, expression of dominant-negative GATA4 or antisense GATA4 mRNAs blocked hypertrophic responses (14,16). In addition, GATA4 regulates adult cardiomyocyte apoptosis and survival (9,17). However, it is not known whether GATA4 continues to play a role in maintaining normal heart physiology during adulthood. Such studies in mammals are hampered by the complexity of mammalian hearts and the redundancy between GATA4–6.
Drosophila has recently emerged as an attractive model system to study the heart, not only during embryonic development, but also to investigate the adult heart and its physiology (18–20). Numerous innovations in a variety of heart assays, including external electrical pacing, electrical recording, optical coherence tomography and high-speed image analysis of heart contractions, greatly facilitated functional studies in the fly's adult heart (21–27). By using these methods, important findings on the roles of ion channels, sarcomeric proteins, dystrophin glycoprotein complex and longevity pathway components in adult cardiac performance were made (22,23,28–32). Combined with powerful genetic tools available in the fly, screens aiming to identify molecular-genetic pathways regulating cardiac contractility and rhythmicity can be conducted, providing new cardiac regulators for examination in mammalian systems (27,33).
In Drosophila, there are three GATA factors: pannier, serpent and grain, all of which play important roles during development (34,35). serpent functions primarily in hematopoiesis, gut and fat body formation, whereas grain is required for filzkorper (structural specializations of the eighth abdominal segment in Drosophila larva) and head skeleton morphogenesis. As GATA4 plays a crucial role in vertebrate heart development and is strongly implicated in human cardiac disease, the fly counterpart pnr is also critical for Drosophila heart development (36–39). During embryonic stages, pnr is required for formation of all tinman-expressing cardiac progenitor cells. Loss of pnr results in defective differentiation of both myocardial and pericardial cells (36–38). In contrast, pan-mesodermal overexpression of pnr induces ectopic cardiac progenitors. Expression of a dominant-negative pnr, pnrEnR, in the mesoderm precisely mimics the heart defects of the pnr mutants (36).
In this paper, we investigate the adult requirement for pnr in the heart. We found that pnr heterozygous mutant hearts exhibit a higher susceptibility to pacing stress, suggesting the dosage of pnr is critical for proper performance of the adult heart. By using a dominant-negative transgene of pnr in combination with the temperature-sensitive Gal4–Gal80 system (40), we restricted pnr expression to the adult heart and muscles, which revealed an adult-specific requirement of pnr in maintaining normal heart function. To further explore the role of pnr in the adult heart, we determined in detail how the dynamics of cardiac contraction is affected in heterozygous pnr mutants. pnr/+ hearts beat more slowly and are arrhythmic. Furthermore, we identified Tbx20/nmr as a candidate downstream mediator of pnr in maintaining normal adult cardiac performance and heart rhythm. Our findings suggest that the early cardiogenic transcriptional regulators are re-employed in the adult heart. Thus, mutations in these factors may contribute to human heart disease during adult stages.
Previous studies showed that the cardiogenic transcription factors tinman and nmr also function in the adult heart (41,42). Therefore, we hypothesized that pnr may also contribute to maintain adult cardiac function. To test this notion, we first measured cardiac performance in pnr mutants by using an electrical pacing protocol to provoke cardiac dysfunction, which is defined as the proportion of flies that exhibit ‘arrest/fibrillation’ (a.k.a. electrical ‘heart failure’) as a consequence of the applied pacing stress (23,27). Heterozygous flies for pnrVX6, an amorphic allele containing a C-terminal deletion of the gene (43), exhibit a dramatic increase in arrest/fibrillation upon electrical pacing (Fig. 1A). Heterozygotes of pnr1 (43), which contains a nucleotide substitution (G1034A, likely causing a truncated protein), or a third allele, pnrVX1e (causing a smaller C-terminal deletion), similarly showed an increase in pacing-induced arrest/fibrillation (Fig. 1A). As an alternative approach, we utilized UAS–Gal4 system to specifically inhibit pnr function in the adult heart by crossing the cardiac-specific driver tinCΔ4-Gal4 to a well-established dominant-negative form of pnr, UAS-pnrEnR (36). The progeny of these flies also exhibit a significant increase in arrest/fibrillation, which recapitulates the phenotypes of heterozygous pnr loss-of-function alleles (Fig.1A).
It has previously been shown that the percentage of wild-type flies exhibiting pacing-induced cardiac arrest/fibrillation increases with age (27; see also Fig. 1D). To further test whether the defective cardiac performance was indeed due to the cardiac/mesodermal loss of pnr, we conducted rescue experiments using various cardiac and mesodermal drivers. The increased arrest/fibrillation rate of pnrVX6/+ and pnr1/+ was partially rescued by mesodermal or heart expression of a wild-type UAS-pnr transgene (Fig. 1B and C). This suggests that the observed phenotype is indeed due to lack of pnr in heart and/or mesoderm (Fig. 1B and C). When we examined the rescue experiments at different time points, we found a similar extent of rescue in 1- and 3-week-old flies but not longer at 5 weeks of age (Fig. 1E and F). Taken together, we conclude that partial loss of pnr compromises adult heart function, which implies that the dosage of pnr is crucial for maintaining proper heart performance (in young flies).
The progressive increase in cardiac arrest/fibrillation rate of pnr/+ when flies were aged suggested that pnr may be required at adult stages in addition to its role during embryogenesis. To distinguish between an embryonic and an adult requirement for pnr, we disrupted pnr in the heart using a dominant-negative form during adult stages only (using the TARGET system: tub-Gal80-ts;tinCΔ4-Gal4 or tub-Gal80-ts;24B-Gal4 crossed to UAS-pnrEnR, see Materials and Methods) (40). In the progeny of these flies, we observed a dramatic increase in the rate of pacing-induced cardiac arrest/fibrillation (Fig. 2A), which was further exasperated with age (Fig. 2A). Curiously, in tub-Gal80-ts;tinCΔ4-Gal4/UAS-pnrEnR flies, the heart's susceptibility to pacing is very high already at a young age, whereas in tub-Gal80-ts;24B-Gal4/UAS-pnrEnR, flies cardiac dysfunction is lower at young ages, but quickly increase with time.
Because cardiac dysfunction may influence the organism's life expectancy (44–46), we examined whether partial loss of pnr affects fly lifespan. Haploinsufficiency of pnr gave arise to a reduction in lifespan, which can be partially rescued by expressing pnr in the mesoderm of pnr heterozygotes (twi 24B-Gal4, 24B-Gal4, Dmef2-Gal4; Fig. 2B). This suggests that pnr heterozygotes affect both adult heart function and lifespan. We also expressed dominant-negative pnr using the primarily cardiac-expressed tinCΔ4-Gal4 driver (47), which resulted in a severely truncated lifespan (Fig. 2B). We speculate the reduced lifespan is primarily due to compromised heart performance caused by disruption of normal Pnr protein function in the heart, reminiscent of reduced cardiac nmr or tinman function (41,42). These data suggest that pnr function is also critical in the adult to modulate cardiac performance and longevity in Drosophila.
To further explore how pnr regulates heart function, we conducted image-based contraction analysis in pnr mutant flies. High-speed movies taken from dissected and exposed hearts provide detailed information of heart contractility (23). M-mode displays of these movies illustrate the dynamics and rhythmicity of heart wall movements (see Materials and Methods). Loss of one copy of pnr causes an increase in irregular heart rate with progressive bradycardia compared with the controls (Fig. 3A–B′). This irregularity (arrhythmicity) is mainly due to the increased variability of diastolic interval lengths of the heart period (Fig. 3A–D). We quantified the severity of arrhythmia as ‘arrhythmicity index’, which is defined as the standard deviation of the heart period normalized to the median of each fly [Fig. 3E; see ref (23) for details]. Thus, hearts with reduced pnr function beat more slowly and are arrhythmic. To further test whether full pnr function was required within the heart/mesoderm for normal cardiac rhythm and rate, we performed rescue experiments by expressing wild-type pnr in the mesoderm of pnr heterozygotes. Both arrhythmias and slow heart beats (due to increased diastolic intervals) were reversed, comparable to wild-type levels (Fig. 3D and E), indicating that the observed heart function abnormalities were indeed due to cardiac/mesodermal loss of pnr function. The arrhythmicity we observed with pnr heterozygous mutants is reminiscent of the phenotype of adult cardiac nmr mutants (42), which raises the possibility of a connection between pnr and nmr in maintaining adult heart function (see below).
Recent studies show that the early cardiogenic transcription factors encoded by tinman and nmr are required for adult heart function (41). In addition, tinman and nmr exhibit a strong genetic interaction in regulating cardiac function. In order to test whether pnr also exhibits a genetic interaction with tinman or nmr, we generated trans-heterozyous mutants for pnr and tinman or pnr and nmr. The pacing-induced heart arrest/fibrillation rate of pnr/+;tin/+ or pnr/+;nmr/+ was similar to the single heterozygous controls (Fig. 4A and B), suggesting that reducing levels of tinman or nmr do not further aggravate the already much compromised cardiac performance of pnr/+.
In a complementary set of experiments, we tested whether overexpressing tinman or nmr was able to rescue the adult heart defects of pnr/+. Because in embryonic pnr null mutant hearts tinman and nmr are dramatically downregulated (36–39,48), we reasoned that tinman and/or nmr may act downstream of pnr in maintaining adult heart function. Mesodermal overexpression of tinman did not ameliorate cardiac dysfunction of pnr heterozygotes (Fig. 4C). In contrast, however, nmr2 expression in heart and muscles dramatically rescued the high arrest/fibrillation rate of reduced pnr function (Fig. 4D). Overexpression of nmr2 in heart using the heart-specific tinCΔ4-Gal4 driver also resulted in considerable rescue of cardiac dysfunction in pnr/+ (Fig. 4D). Rescue by nmr2 but not by tinman of pnr heterozygotes was also observed in the development of cardiac arrhythmias (Fig. 5). The normalizing effect of nmr overexpression in pnr/+ is also illustrated by the shortening of median length of the heart period, due to a decreased diastolic interval (Fig. 5A–E), and a reversal of the elevated arrhythmias. In sum, nmr overexpression in the mesoderm reversed the slow and irregular heart beat due to reduced pnr function. In contrast, tinman expression was ineffective in improving the compromised heart function of pnr heterozygotes (Fig. 5C, E, F). These data suggest that nmr is critical in directly or indirectly mediating essential aspects of pnr function in the control of adult cardiac physiology.
Since maintaining expression of both tinman and nmr depends on pnr (37,44), but only overexpression of nmr is capable of restoring the heart function defects of pnr heterozygotes, we tested the hypothesis that nmr expression in pnr/+ hearts is affected more dramatically than that of tinman. For this purpose, we performed quantitative real-time PCR (qRT-PCR) on mRNA isolated from heterozygous hearts of three pnr alleles (see Materials and Methods). Indeed, we found that nmr1 RNA is downregulated over 90% and nmr2 65–80% in the hearts of three different pnr heterozygotes (Fig. 6A and B), consistent with the interpretation that nmr is a critical target of pnr in establishing normal cardiac function in adult flies. In contrast, tinman RNA is only decreased by 20–50%, thus much less than the nmr genes (Fig. 6C). Since maintenance of tinman expression depends on nmr function (48), the moderate reduction in tinman RNA may primarily be due to decreased nmr expression.
To further explore the possible mechanisms underlying that pnr regulates adult cardiac function, we qRT-PCR screened mRNA isolated from pnr/+ hearts. Nearly 30 genes coding for potential effectors or signaling pathways that were suspected to potentially function downstream of pnr in adult fly hearts were tested. We found that the expressions of nine of these genes, in addition to nmr and tinman, were significantly downregulated in at least two different pnr heterozygous mutants (Fig. 6D–L). One set of misregulated factors are encoded by structural genes. Dystrophin, an ECM protein that regulates the integrity of myofibril structure (29), was downregulated in all three pnr heterozygotes (Fig. 6D). RNA expression levels of myosin heavy chain (MHC) and collagen 18 were reduced prominently in the absence of one copy of pnr (Fig. 6E and F). The other gene set encodes potassium- or calcium-type ion channels or exchangers, including KCNQ, dSUR, shaker, Na/Ca-exchange protein (calx), slowpoke, Calcium ATPase at 60A (SERCA)(Fig. 6G–L). Notably, the potassium channel genes, KCNQ and dSUR, were dramatically downregulated in all pnr heterozygous mutants (by 80–90%), consistent with elevated cardiac arrhythmias and pacing-induced cardiac dysfunction of KCNQ and dSUR mutants, respectively (23,28). We noticed that decreased expression of some candidate genes is only observed in two of three pnr alleles, which may be because different pnr mutations could affect some target genes' expression slightly differently, or the reduction in gene expression is small to begin, thus not always reaching statistical significance. Overall, these data suggest that partial loss of pnr in the adult fly heart dramatically affects the gene network of cardiogenic transcription factors that control muscle fiber and ion channel gene expression, which in turn may be the cause of the observed heart function abnormalities.
GATA factors constitute an important family of transcription factors that orchestrate cardiac specification and differentiation. Specifically, GATA4, which is enriched in the cardiac mesoderm, was shown to act in a combinatorial fashion with other cardiac transcription factors such as Mef2, SRF, Nkx2-5 and Hand1/2 in defining cardiac lineages (2,5–7). It is also known that Gata4 is re-employed in adult heart upon hypertrophy stimuli (2,9,13–15). However, it was not known whether Gata4 is also involved in regulating adult heart contractility and rhythm. In this study, we identified novel roles for GATA factor Pnr in establishing adult cardiac function. By using conditional and cardiac-specific manipulation, we found that pnr/Gata4 continues to have a regulatory role in the adult fly heart, and its partial depletion dramatically interferes with cardiac performance. Consistent with recent studies on the roles of tinman/Nkx2-5 and nmr/Tbx20 (41,42), our findings further demonstrate the importance of the re-employment of early cardiogenic genes in the adult heart. Thus, the genetic network of the transcription factors responsible for early heart formation (20,49,50) seems to be used again in regulating the establishment and maintenance of adult heart function. Because of the high conservation, we speculate that Gata4 may also have a role in maintaining adult heart function in vertebrates. Remarkably, the heart phenotype of pnr partial loss-of-function is different from those of tinman/Nkx2-5 and nmr/Tbx20 heterozygotes, in which cardiac function is not impaired (42). This suggests that the precise dosage of pnr is critical for the fly heart, as Gata4 is in mammals, since haploinsufficiency of GATA4 causes CHD in human.
Interestingly, we found that pnr1/− flies exhibit the strongest compromise in cardiac performance, which is consistent with the previous observation of homozygous pnr1 embryos having a more extreme phenotype than a deficiency for the locus (43), suggesting it acts in a partly dominant-negative fashion. It would be interesting to see whether expression of a pnr1 transgene (G1034A) in a wild-type background causes a partial pnr heart phenotype as pnr1/−.
Gata4 has been widely studied in several model organisms and was shown to play a crucial role in heart development. Emerging evidence indicates that Gata4 continues to be expressed in the adult and its deficiency prevents stress-induced cardiac hypertrophy and causes heart failure (2,9,13–15). Pressure overload of the heart in vivo as well as stimulation of cardiac myocytes in vitro activates expression of Gata4 and other genes responding to hypertrophic stimuli. Since a ‘hypertrophic’ response has not yet been observed in the fly heart it is possible that we have discovered other physiological requirements for this transcription factor. It will be interesting to see, however, whether a Gata4-mediated hypertrophic response in the mammalian heart includes the activation of the structural genes that we found to be downregulated with reduced pnr function in flies. In sum, Pnr/Gata4 seems to play several, possibly conserved roles in the developing as well as in the adult heart.
In an attempt to identify downstream targets that may be regulated by pnr in the adult heart, we found that nmr/Tbx20 is critical, since overexpression of nmr2 in heart/muscle tissue in pnr heterozygotes significantly restores normal heart function. This is consistent with our previous finding that nmr is no longer expressed in the cardiogenic region in pnr mutant embryos (48). Interestingly, we found consensus GATA binding sites in the promoter regions of the nmr genes, suggesting that pnr might directly regulate nmr expression. Further analysis is required to test whether these sites are functionally required for nmr expression and are bound by Pnr protein. We have previously shown that nmr/Tbx20 genetically interacts with tinman/Nkx2-5 in regulating cardiac physiology (42). In this study, we found that, unlike nmr, tinman overexpression in heart cannot rescue pnr mutant defects, and double heterozygous mutants for tinman and pnr are not performing worse than single pnr/+ mutants. These data are consistent with the idea that the relationship of pnr with tinman different from that with nmr. On the basis of our investigation of several critical cardiogenic transcription factors in the adult heart (tinman/Nkx2-5, nmr/Tbx20, pnr/Gata4; see 41,42 and this study), we conclude these factors are re-employed in the adult and are critical for maintaining normal heart function.
The following mutant stocks were used: pnrVX6/TM3 (43), pnr1/TM3 (43), pnrVX1e/TM3 (43), tin346/TM3-ftzlacZ (51), tinEC40/TM3 (52), Df(3R)tinGC14 (52), nmr1614 (48), Df(2L)Exel6012 (53, this Exelixis deficiency deletes both nmr1 and nmr2). Description of the pnr and other alleles is provided at http://flybase.org/reports/FBgn0003117.html. Wild-type control flies are w1118. Overexpression of transgenes was achieved using the UAS–Gal4 system (47). The following Gal4 and UAS lines were used: twi-Gal4 (twi>) (54), 24B-Gal4 (24B>) (47), the double combination twi-Gal4;24B-Gal4 (twi24B>; pan-mesodermal expression) (55), tinCΔ4-Gal4 (56) (cardiac-specific), Dmef2-Gal4 (57), UAS-pnr (58), UAS-tin (59), UAS-nmr1 and UAS-nmr2 (48).
The electrical pacing was conducted as previously described (27,32). Briefly, 50–100 flies were paced with a square wave stimulator at 40 V and 6 Hz for 30 s and scored for heart failure rate, defined as the percentage of flies that enter either cardiac arrest or fibrillation during or immediately after pacing.
Procedures used were as in Ocorr et al. (23). Briefly, flies were anaesthetized and the head, ventral thorax and ventral abdominal cuticle were removed. All internal organs except the heart were removed as well as any abdominal fat. Dissections were done under oxygenated saline (108 mm Na+, 5 mm K+, 2 mm Ca2+, 8 mm MgCl2, 1 mm NaH2PO4, 4 mm NaHCO3, 10 mm sucrose, 5 mm trehalose, 5 mm HEPES, pH 7.1). These semi-intact preparations were allowed to equilibrate with oxygenation for 10–20 min prior to filming. These procedures were carried out at room temperature. Image analysis of heart contractions was performed using a Hamamatsu EM-CCD digital camera on a Leica DM LFSA microscope using direct immersion lenses. High-speed movies taken at rates of 100–200 frames per second were acquired and contrast enhanced using Simple PCI imaging software (Compix, Inc.). M-modes were generated using Simple PCI software and MatLab-based image analysis (23).
The longevity measurements were performed as previously described (32). Briefly, 100–200 males and 100–200 virgin females were aged at 25°C. Flies were transferred to fresh food on alternative days and dead flies were counted after transfer.
Total RNA was extracted from 20 dissected hearts of 1-week-old adult females by using TRIzol (Invitrogen). After lysis with TRIzol followed by phase separation, the aqueous phase was subjected to RNA extraction (Mini RNA isolation kit, Zymo Research, Orange, CA, USA) with on-colomn DNase I treatment (QIAGEN). After RNA extraction with RNase-free water, the first-strand cDNA was immediately transcribed with SuperScript III (Invitrogen) by using oligo(dT) primer, followed by the second-strand synthesis with DNA polymerase I, E. coli DNA ligase and RNase H. qRT-PCR was carried out using the LightCycler FastStart DNA Master PLUS SYBR Green I kit (Roche). The primer sets were as follows:
The following gene expressions did not show any consistent and/or significant difference among the genotypes we tested: InR signaling: InR, Pten, Akt, Foxo, TOR; transporters and exchangers: alalar1, cacophony, Inositol 1,4,5,-tris-phosphate receptor (Itp-r83A) and Na+-driven anion exchanger 1 (Ndae1); potassium channels: seizure, Inwardly rectifying potassium channel (Ir), eag-like K+ channel and others: cdc42, bves, β3-tubulin, p53, smt3, sima, PKC.
L.Q. was supported by a predoctoral fellowship from the American Heart Association. This work was funded by grants from NHLBI of the National Institutes of Health to R.B.
We thank the Bloomington stock center for sending fly stocks. We thank Timothy Crawley for his technical assistance with the heart function experiments, Nikki Alayari and Takeshi Akasaka for their help with qRT-PCR, and Karen Ocorr for her help in the initial characterization of pannier heterozygotes. We also thank Anthony Cammarato for critical reading of the manuscript.
Conflict of Interest statement. None declared.