The Drosophila muscle LIM protein, Mlp84B, is essential for cardiac function

Hypertrophic and dilated cardiomyopathies are major causes of human heart failure and mortality. Familial forms of these pathologies have led to investigation of the genetic causes, and there is a growing body of evidence that suggests a role for mutations in cardiac cytoskeletal genes(Chien, 2000; Seidman and Seidman, 2001). Cytoskeletal components located in the Z-disc have previously been recognized as critical for normal cardiac function. The Z-discs define the lateral boundaries of sarcomeres and constitute anchoring sites for actin and titin filaments, crosslinked by α-actinin. It is now well established that Z-disc components are essential for mechanical stretch sensing and sarcomere integrity (Clark et al., 2002). Loss or mutation of Z-disc proteins such as α-actinin-associated LIM protein (Pashmforoush et al.,2001), Cypher/ZASP (Vatta et al., 2003; Zhou et al.,2001), muscle LIM protein/CRP3(Arber et al., 1997; Geier et al., 2003; Knöll et al., 2002; Mohapatra et al., 2003), titin(Gerull et al., 2002) and Tcap, also known as telethonin (Hayashi et al., 2004; Moreira et al.,2000), all lead to hypertrophic or dilated cardiomyopathy in mouse models as well as in human patients.

Loss of Drosophila MLP (Mlp84B) leads to developmental arrest at the pupal stage. Specifically, the muscle-dependent morphogenetic movements necessary for pupation are severely compromised in Mlp84B mutants(Clark et al., 2007). These defects may be explained by muscle weakness, and the observation that the few mlp84B^–/– adult escapers fail in a flight test is consistent with this hypothesis (Clark et al., 2007). However, the cause of lethality remains unclear. Mlp84B shares many features with vertebrate MLPs including muscle-specific expression and localization at the Z-disc I-band interface(Stronach et al., 1999). Since vertebrate studies have demonstrated a critical requirement for MLP in cardiac function (Arber et al., 1997)and the MLP/CRP3 residues mutated in human patients are conserved in Drosophila Mlp60A and Mlp84B (Fig. 1A), we hypothesized that Mlp84B is essential for normal cardiac function in Drosophila. Here we show that Mlp84B is expressed in the Drosophila heart from late embryonic stages to adulthood and that mlp84B-deficient Drosophila display bradycardia and heart rhythm abnormalities. Thus, Drosophila provides a new model system in which genetic and physiological tools can be applied in combination to investigate the cardiac stretch-sensing response in vivo.

w¹¹¹⁸ flies (Drosophila melanogaster) were used as the wild-type control. Crosses were done at 25°C unless otherwise specified. Expression of mlp84B RNA interference (RNAi) was achieved using the standard Gal4/UAS system (Brand and Perrimon, 1993). The drivers used were Daughterless(Da)-Gal4 and tinCΔ4-Gal4(Lo and Frasch, 2001). UAS-mlp84B-RNAi flies were obtained from a genome-wide RNAi collection (B. Dickson, Institute of Molecular Pathology, Vienna, Austria). The sequence of the RNAi construct was as follows:ACCGATGGCTTGACCGAGGATCAGATCAGCGCCAACAGGCCCTTCTATAACCCGGACACCACGTCAATTAAGGCCCGTGACGGCGAAGGCTGTCCCCGGTGCGGAGGAGCCGTATTCGCCGCCGAGCAACAGCTGTCCAAGGGCAAGGTGTGGCACAAGAAGTGCTACAACTGCGCCGACTGCCACCGGCCATTGGACTCGGTCCTGGCCTGCGATGGACCCGATGGCGACATCCACTGCCGCGCCTGCTACGGC. mlp84B^–/– flies were generated by combining the mlp84B^P8 P-element excision allele with the Df(3R)dsx2M deficiency (for details, see Clark et al., 2007). This combination results in the complete absence of Mlp84B protein(Clark et al., 2007). These flies were crossed and raised at 18°C to optimize the yield of adult escapers. A mlp84B transgene was reintroduced in the Df(3R)dsx2M fly line to be used as a rescue of the mlp84Bdeficiency when crossed to mlp84B^P8(Clark et al., 2007). The mlp84B-GFP (green fluorescent protein) transgene was constructed as follows. The mlp84B transgene consists of a 7 kb genomic fragment containing the entire mlp84B transcription unit, plus 2 kb of upstream sequence and 1.5 kb of downstream sequence. A BamHI site was engineered just before the stop codon of the mlp84B open reading frame (ORF). The GFP ORF was amplified by PCR using primers to introduce a BamHI site at both the 5′- and 3′-ends and allow for cloning into the mlp84B transgene in-frame with the mlp84BORF. The cypher-GFP protein trap line (line G189)(Morin et al., 2001) was a gift from X. Morin to K.A.C. These flies were crossed to flies bearing the mlp84B^P8 mutant allele to generate cypher-GFP;mlp84B^P8. cypher-GFP;mlp84B^–/– flies were then obtained by combining cypher-GFP;mlp84B^P8 with Df (3R)dsx2M.

Total RNA was isolated from whole flies or dissected hearts using an equal number of 1 week old males and females with TRIzol Reagent (Invitrogen,Carlsbad, CA, USA) according to standard protocol, then DNAseI digestion was performed to ensure elimination of genomic DNA. RNA was reverse transcribed using the Superscript First Strand kit from Invitrogen and diluted 1:5 prior to real-time quantitative PCR. The nucleotide sequences of the PCR primers used were as follows: mlp84B forward 5′-ACGTCAATTAAGGCCCGTGAC-3′ and reverse 5′-AGGACGGCCATCTGGGAACTGG-3′; actin79B forward 5′-ATCCGCAAGGATCTGTATGC-3′ and reverse 5′-ACATCTGCTGGAAGGTGGAC-3′(Akasaka et al., 2006). Real-time quantitative PCR was performed using a LightCycler (Roche Diagnostics, Indianapolis, IN, USA) rapid thermal cycler. Amplification was carried out as recommended by the manufacturer using the Light Cycler-DNA Master SYBR Green I mix containing 1 μmol l^–1 of appropriate primer mix, added with 2 μl of cDNA. The amplification program included an initial denaturation step at 95°C for 8 min, and 45 cycles of denaturation at 95°C for 5 s, annealing at 60°C for 6 s and extension at 72°C for 15 s. Melting curves were used to determine the specificity of PCR products, which was confirmed by conventional gel electrophoresis. Relative mlp84B expression represents the average of triplicates normalized to actin79B (see Akasaka et al., 2006). Error bars represent the standard error of the mean. Student's t-test was used to analyze statistical significance. P values corresponded to two-tailed tests, and P<0.01 was considered statistically significant.

Lifespan was determined by collecting a minimum of 100 male and 100 female flies for each genotype, distributed in food vials (standard yeast/molasses/cornmeal) containing 25 flies or fewer of the same sex. Flies were maintained at 25°C and transferred to a fresh vial every 2–3 days. The number of dead flies was scored at each vial change.

Embryos were heat-fixed and processed for immunofluorescence as previously described (Clark et al., 2003). Third instar larvae were immobilized using insect pins, filleted in calcium-free Ringer solution and fixed in –20°C methanol for 10 min. Processing of the fixed fillets for immunofluorescence was done as for embryos. One week old flies were dissected to expose the heart and fixed for 30 min in 4% paraformaldehyde. The following antibodies were used at the concentrations indicated: rabbit anti-Mlp84B (B50)(Stronach et al., 1996) 1:500,rat anti-α-actinin (Technix, Cambridge, UK) 1:100 (for larval heart stainings) and mouse anti-α-actinin (a gift from J. Saide)(Saide et al., 1989) 1:20 (for adult heart stainings). Secondary antibodies were FITC-conjugated anti-mouse and Cy3-conjugated anti-rabbit IgG (Sigma, St Louis, MO, USA). Samples were mounted in Vectashield (Vector Labs, Burlingame, CA, USA). Images were acquired on an MRC 1024 SP BioRad laser point scanning confocal microscope using LaserSharp 2000 software (Bio-Rad, Hercules, CA, USA). Three to four consecutive confocal sections were acquired in the z-axis with a step of 2 μm and projected in ImageJ software(http://rsb.info.nih.gov/ij).

Image analysis of semi-intact preparations was carried out as described previously (Ocorr et al.,2007). Ten to 15 flies per genotype were anesthetized with Flynap(Carolina, Burlington, NC, USA) and dissected in an oxygenated saline solution mimicking adult hemolymph [based on previous studies(Singleton and Woodruff, 1994; Wang et al., 2003)]. The artificial hemolymph used in this study contained 108 mmol l^–1 Na⁺, 5 mmol l^–1K⁺, 2 mmol l^–1 Ca²⁺, 8 mmol l^–1 MgCl₂, 1 mmol l^–1NaH₂PO₄, 4 mmol l^–1 NaHCO₃,10 mmol l^–1 sucrose, 5 mmol l^–1 trehalose and 5 mmol l^–1 Hepes (pH 7.1). All internal organs and abdominal fat were removed in order to expose the heart. Roughly equal numbers of males and females were used, except for the mlp84B^–/– genotype, where the analyzed sample comprised three female and 13 male flies owing to the very low number of female adult escapers. Prior to imaging, dissected flies were allowed to recover for 20–30 min at room temperature in oxygenated saline solution. Sequences of cardiac contractions were recorded for 30 s (wild-type flies) or 1 min (all other genotypes) at room temperature on a Leica DM LFSA microscope(Bannockburn, IL, USA) equipped with a ×10 water-immersion lens. Movies were taken at rates of 100–200 frames s^–1 using a fast-acquisition Hamamatsu EM-CCD digital camera (Irvine, CA, USA) controlled by SimplePCI software (Compix, Inc., Hamamatsu Corporation). M-modes were generated using a MatLab-based image analysis program written by M. Fink (M. Fink, W. Giles, R.B. and K.O., unpublished data). Briefly, a 1 pixel-wide region is defined in a single frame of the movie that encompasses both edges of the heart tube; identical regions are then cut from all of the frames in the movie and aligned horizontally, providing an edge trace that documents the movement of the heart tube edges in the y-axis over time in the x-axis. Measurements of the heart period and diastolic interval were obtained as output from the MatLab-based program. Results represent the average of the median heart period or diastolic interval in each recording. Error bars represent the standard error of the mean. Two-tailed tests were performed to determine statistical significance, and P<0.01 was considered statistically significant. For the histogram representation, the frequency of interval duration was plotted for all heart beats in one given genotype using a bin of 0.02 s. The cutoff was fixed at 2 s for heart period,1.4 s for diastolic interval and 0.6 s for systolic interval. Longer intervals were pooled and represented at the cutoff value. Asystole was defined by a diastolic interval longer than 1 s, which is 3 times the average diastolic interval observed in 5 week old wild-type flies (0.33 s). Tachyarrhythmia was defined by a systolic interval longer than 0.5 s (approximately twice the average systolic interval observed in 5 week old wild-type flies, which was 0.22 s) or a diastolic interval shorter than 0.05 s (about half the shortest median diastolic interval observed in 1 week old wild-type flies, which was 0.12 s). The frequency of asystole and tachyarrhythmia was calculated by adding the asystole/tachyarrhythmia events observed in all recordings for each genotype and dividing this by the number of flies analyzed.

To gain insight into the possible requirements for Mlp84B during development, we examined its expression at various stages. Mlp84B was undetectable in the embryonic heart until late stages, where it was less strongly expressed compared with body wall muscles(Fig. 1B). A Mlp84B::GFP fusion protein was strongly expressed in the heart tube of 3rd instar larvae(Fig. 1C) and localized to sarcomeres as well as cell nuclei. Staining with Mlp84B antiserum likewise revealed robust expression in the regularly arranged sarcomeric units constituting the cardiac myofibrils (Fig. 1D). Mlp84B was found to co-localize with α-actinin at the Z-line of cardiac sarcomeres (Fig. 1E–G).

We then examined Mlp84B expression in the adult. The adult fly heart is a tube composed of transverse or spiral muscle fibers that constitute the myocardium, apposed by a layer of ventrolateral longitudinally oriented myofibrils (Fig. 2A,B)(Molina and Cripps, 2001).α-Actinin and Mlp84B immunostaining revealed co-localization at the Z-disc of sarcomeres in the longitudinal myofibrils. α-Actinin localized in a tight band (Fig. 2C)whereas Mlp84B was found in a broader domain, forming a doublet(Fig. 2D). The merged image shows that the α-actinin-positive band lies in the center of the Mlp84B doublet (Fig. 2E). This pattern is the same as that seen in larval and adult body wall muscle(Clark et al., 2007) (and data not shown) as well as murine cardiomyocytes(Arber et al., 1997).

In order to examine heart function in mlp84B-deficient Drosophila, we expressed mlp84B RNAi either ubiquitously or specifically in the heart. mlp84B RNAi produced viable flies,probably due to incomplete knockdown. RNAi efficiency was tested in whole flies using the ubiquitous Daughterless (Da)-Gal4 as well as specifically in the heart using the cardiac driver tinCΔ4-Gal4 (Lo and Frasch, 2001). In both cases, mlp84B relative expression was strongly decreased (by 74% and 65%, respectively, Fig. 3A,B). To determine whether mlp84B knockdown in the heart impairs adult viability, we conducted lifespan studies. Compared with wild-type or controls(UAS-mlp84B-RNAi outcrossed to wild-type), the RNAi-expressing male flies had a dramatically reduced lifespan(Fig. 3C). This finding points to a role for Mlp84B in heart performance and aging, and prompted us to examine cardiac structure and function in mlp84B mutants.

Body wall muscle structure is not significantly altered in mlp84B^–/– larvae(Clark et al., 2007). Similarly, MLP knock-out mice show no obvious skeletal muscle phenotype, but develop severe dilated cardiomyopathy a few weeks after birth(Arber et al., 1997). We thus sought to assess cardiac myofibril cytoarchitecture in mlp84B-deficient 1 week old adult Drosophila. In mlp84B^–/– adult hearts, anti-α-actinin staining revealed some sarcomeric disorganization in the ventral longitudinal myofibrils as evidenced by a disrupted Z-disc arrangement(Fig. 4A) compared with wild-type (Fig. 2C), including Z-disc misalignments and gaps. Normal sarcomeric architecture was restored in the mlp84B^–/– rescue genotype(Fig. 4C). To visualize the inner cardiac myofibrils that were only weakly stained by the anti-α-actinin antibody, we introduced the protein trap cypher-GFP (Morin et al.,2001) in the mlp84B^–/– background. Sarcomeric structure did not appear to be affected by the absence of Mlp84B in the inner, spiral muscle fibers that constitute the myocardium proper(Fig. 4E). Thus, mlp84B deficiency in Drosophila does not reflect the severe sarcomere disorganization found in MLP^–/–murine hearts, and lethality due to the absence or down-regulation of Mlp84B cannot be readily explained by sarcomeric architecture defects.

For technical reasons, it is difficult to assess cardiac function in larvae. However, we were able to use an imaging-based cardiac performance assay to investigate the consequences of the absence or reduced levels of Mlp84B in the adult heart. Partial dissection of the abdomen in a solution mimicking hemolymph allows for recording of the cardiac contractions and measurement of various heartbeat parameters(Ocorr et al., 2007). We determined the heart period as well as systolic and diastolic intervals in mlp84B^–/– or RNAi-expressing flies as compared with wild-type and controls (UAS-mlp84B-RNAi outcrossed to wild-type). Because mlp84B^–/– flies have a shortened lifespan, we only measured heartbeat intervals in 1 week old flies. Drosophila expressing mlp84B RNAi in the heart were examined at 1 week and 5 weeks of age in order to uncover a potential increased deterioration of cardiac function with age. The most striking observation was an increased duration of the heart period (heart beat length consisting of systolic plus diastolic interval), and this was mainly due to a prolongation of the diastolic interval (relaxation phase; Fig. 5). In 1 week old mlp84B^–/– flies, we observed a significantly longer heart period than in wild-type age-matched controls(Fig. 5A). This effect was partially reversed in the rescue line. Flies expressing mlp84B RNAi(mlp84B knockdown) in the heart were not significantly different from controls at 1 week of age (not shown); however, 5 week-old mlp84Bknockdown flies exhibited a significantly prolonged heart period compared with controls (Fig. 5B). Significantly, the prolonged heart period seen in 5 week old mlp84Bknockdown flies was due to an increase in the diastolic interval similar to that observed for the 1 week old mlp84B-deficient flies(Fig. 5C,D).

We also examined the distribution of heart periods as well as systolic and diastolic intervals, which highlights the deterioration of the heartbeat regularity (Fig. 6, and Fig. S1 in supplementary material). Both heart period and diastolic interval distributions were altered in mlp84B^–/– flies(compare the position, amplitude and sharpness of the frequency peak in mlp84B^–/– versus wild-type and rescue flies, Fig. 6A). The distribution of systolic intervals appeared to be unaffected. In 5 week old mlp84BRNAi-expressing flies, the frequency peak was shifted towards longer durations and the distribution of heart periods and diastolic intervals was more scattered than in age-matched wild-type and UAS-mlp84B-RNAioutcrossed to wild-type controls (Fig. 6B). This reveals that in addition to a decreased heart rate,diminished levels of Mlp84B resulted in a more irregular heartbeat due to an increased variability in the diastolic interval duration (see also supplementary material Fig. S1). Systolic intervals were slightly longer in 5 week old mlp84B RNAi-expressing flies than in controls, but this difference was not statistically significant (not shown). These data suggest that partial or total loss of Mlp84B causes a significant diastolic dysfunction.

As another index of heart function, we monitored mlp84B^–/– mutant and mlp84B RNAi cardiac-knockdown flies for heart rhythm abnormalities. Specifically, we examined the occurrence of tachyarrhythmia (systolic interval longer than 0.5 s or diastolic interval shorter than 0.05 s; Fig. 7A, arrows) and the occurrence of asystole events (diastolic interval longer than 1 s; Fig. 7A, arrowheads). No tachyarrhythmia was observed in 1 week old flies of any of the genotypes. At 5 weeks, the occurrence of cardiac tachyarrhythmia was 10 times higher in 5 week old flies expressing mlp84B RNAi in the heart compared with wild-type and control flies (Fig. 7B). Wild-type Drosophila never exhibited asystoles at 1 week of age. In contrast, 1 week old mlp84B^–/– flies displayed on average 4.2 events of asystole per minute(Fig. 7C). The occurrence of asystole was reduced to about half in rescue flies(Fig. 7C). Similarly, 5 week old Drosophila expressing mlp84B RNAi in the heart had a dramatically higher incidence of asystole (7.7 events per minute) compared with the same age wild-type or control flies(`Fig. 7D). In summary, the observed disturbances in cardiac rhythmicity are consistent with the altered distribution of heart periods and diastolic intervals as dramatically illustrated in the histograms of Fig. 6 (see also supplementary material Fig. S1).

The identification of the precise molecular mechanisms that link the cytoskeletal defect in the cardiac Z-disc to the development of cardiomyopathy is critical for the design of therapeutic strategies to prevent disease progression, and there is a need for models in which to study candidate genes. Drosophila is well suited to test genetic interactions, but so far there has been a lack of heart function assays in this organism. A recently developed imaging assay allows for measuring contractile properties of the adult heartbeat in semi-intact adult flies(Ocorr et al., 2007). Using this technique, we examined the role of Mlp84B, the Drosophilacounterpart of MLP, in cardiac function. mlp84B null mutants die at the pupal stage or as young adults, but RNAi knockdown of mlp84Bspecifically in the heart produced viable flies, allowing us to assess cardiac function (which is technically difficult in larvae and pupae). By conducting lifespan studies, we observed that male Drosophila with reduced cardiac expression of mlp84B have a shortened lifespan as compared with control flies, which is consistent with a role for Mlp84B in cardiac function. A detailed analysis of heartbeat parameters in Drosophilawith decreased cardiac mlp84B expression showed a prolongation of the diastolic interval leading to a slower cardiac rhythm in both male and female flies, as well as a greatly elevated occurrence of tachyarrythmia and asystole compared with control flies. This phenotype was subtle in 1 week old Drosophila but significant at 5 weeks of age, suggesting an aggravation of cardiac dysfunction with muscle use and/or age. One week old mlp84B^–/– flies displayed similar perturbations in diastolic function and a high incidence of asystole, which were partially rescued by reintroducing the mlp84B transgene in the mutant background. Taken together, these data establish an essential role for Mlp84B in heart function.

In mice, deletion of the MLP gene induces dilated cardiomyopathy and heart failure that develops after 4 weeks of postnatal life. Among other defects, MLP^–/– animals display severe sarcomeric disarray (Arber et al.,1997), a slower heart rate, and diastolic dysfunction defined by slower relaxation time (Minamisawa et al.,1999). When examining heart function in adult flies with no or reduced mlp84B expression, we found a reduction in heart rate that was the result of a significantly prolonged diastolic interval, which is reminiscent of the mouse phenotype. Although we do not know the reason for the additional abnormalities in myofiber morphology in mutant mouse hearts, a possible clue may be that the mouse heart beats much faster than the fly heart(10 versus 2 beats s^–1, respectively), making it more prone to structural defects. Another explanation could be that various secondary pathomechanisms such as upregulation of proteases and tissue remodeling contribute to myofibrillar disarray in MLP^–/– mice and human patients with mutated MLP. The heart rhythm defects observed in mlp84B null or mlp84B knockdown flies may actually represent the primary defect caused by the absence of MLP, making Drosophila a useful model to investigate the primary mechanisms of the disease without the bias of the secondary remodeling seen in mice.

We also examined the heart rhythm in aging flies with reduced cardiac mlp84B expression. Remarkably, we observed that both mlp84Bnull and mlp84B knockdown flies have a high incidence of asystole and that mlp84B knockdown also causes cardiac tachyarrhythmia in 5 week old flies. These anomalies could be interpreted as manifestations of diastolic dysfunction, perhaps because reduced levels of Mlp84B impair stretch sensing at the Z-disc of cardiac myofibrils, which may result in a failure to maintain a constant relaxation time. Consequently, the diastolic interval duration can be either prolonged or reduced, which is detected as asystole and tachyarrythmia in our analysis. It would be interesting to determine whether muscle tension is also altered in mlp84B mutants.

The severe structural sarcomeric defects characteristic of the MLP^–/– murine heart were not observed in flies. It is possible that in Drosophila other proteins, such as Mlp60A (Stronach et al.,1996), compensate for Mlp84B deficiency to maintain sarcomere integrity. Another feature of MLP^–/– mice is a dramatically enlarged heart as compared with wild-type(Arber et al., 1997). We measured systolic and diastolic diameters in mlp84B^–/– or mlp84B RNAi-expressing Drosophila versus control animals and did not observe any significant increase at the ages examined. It might be that the dilatation phenotype develops in older mlp84B-deficient flies, which would need to be investigated with conditional mutants, since the partial heart-specific RNAi knockdown does not seem to be sufficient. Cardiac chamber dilatation in mutant flies has previously been reported in flies with mutations in the troponin I and tropomyosin genes or expressing a mutatedδ -sarcoglycan human gene in the heart(Wolf et al., 2006). Deletion of the MLP gene in mice also results in impaired systolic function,characterized by a severe decrease in fractional shortening(Arber et al., 1997). In our mlp84B mutants, we did not detect any significant change in fractional shortening. However, it should be noted that young MLP^–/– mice that have not yet developed cardiomyopathy display alterations in passive myocardial properties and relaxation time, but heart diameter and systolic characteristics are normal at that stage, including fractional shortening(Lorenzen-Schmidt et al.,2005).

Based on the phenotype of young MLP^–/–mice, it has been proposed that the progression to heart failure in the MLP deficiency model may be driven by diastolic dysfunction and abnormal passive properties rather than systolic dysfunction(Lorenzen-Schmidt et al.,2005). MLP interacts with Tcap, and mutations in the Tcap binding region of MLP have been found in a subset of patients with dilated cardiomyopathy (Knöll et al.,2002). Interestingly, Tcap also associates with the cytoplasmic domain of the stretch-dependent potassium channel β-subunit minK(Furukawa et al., 2001). The diastolic interval defects that we observed in Mlp84B-deficient flies may thus be due to impaired stretch-dependent signaling on this channel, in turn affecting the delayed rectifier potassium current I(K_s). Another candidate for linking MLP to the cardiomyopathy phenotype is the giant protein Titin, whose interaction with Tcap is required for sarcomeric integrity (Gregorio et al.,1998), suggesting an important role of the Z-disc complex titin–Tcap–MLP in mechanical signaling and regulation of passive myocardial properties. Titin may play a key role in this complex, since it contributes over 80% of passive force during stretch(Wu et al., 2000) and appears to be involved in mechanical signaling(Granzier and Labeit, 2004). In support of this hypothesis, titin mutations have been reported to be associated with a dilated cardiomyopathy phenotype in zebrafish(Xu et al., 2002) and humans(Gerull et al., 2002). In Drosophila, the counterpart of the titin gene is sallimus (sls), also referred to as D-titin or kettin. Loss-of-function sls mutants die as late embryos with severely disrupted muscle organization(Hakeda et al., 2000; Machado and Andrew, 2000; Zhang et al., 2000). mlp84B^–/– larvae that are also heterozygous for a sls mutation display a dramatic loss of sarcomeric integrity whereas mlp84B^–/– or sls^+/– mutants show no visible muscle structure phenotype, demonstrating a genetic interaction(Clark et al., 2007). It will be interesting to investigate the role of sls in Drosophilaheart function, as a functional MLP–titin interaction may be conserved from flies to humans.

We are grateful to Barry Dickson and Georg Dietzl (Institute of Molecular Pathology, Vienna, Austria) for the gift of the UAS-mlp84B-RNAi Drosophila line and to Judith Saide for providing the anti-α-actinin antibody. Laurent Perrin is acknowledged for critical reading of the manuscript. This work was supported by grants from the National Institutes of Health (NHLBI and NIA) to R.B.