Obesity is strongly correlated with lipotoxic cardiomyopathy, heart failure and thus mortality. The incidence of obesity has reached alarming proportions worldwide, and increasing evidence suggests that the parents’ nutritional status may predispose their offspring to lipotoxic cardiomyopathy. However, to date, mechanisms underlying intergenerational heart disease risks have yet to be elucidated. Here we report that cardiac dysfunction induced by high-fat-diet (HFD) persists for two subsequent generations in Drosophila and is associated with reduced expression of two key metabolic regulators, adipose triglyceride lipase (ATGL/bmm) and transcriptional cofactor PGC-1. We provide evidence that targeted expression of ATGL/bmm in the offspring of HFD-fed parents protects them, and the subsequent generation, from cardio-lipotoxicity. Furthermore, we find that intergenerational inheritance of lipotoxic cardiomyopathy correlates with elevated systemic H3K27 trimethylation. Lowering H3K27 trimethylation genetically or pharmacologically in the offspring of HFD-fed parents prevents cardiac pathology. This suggests that metabolic homeostasis is epigenetically regulated across generations.
Heart failure is the main cause of mortality worldwide and is now further aggravated by the growing obesity epidemic. Considerable efforts have been made to identify the genetic basis of obesity. However, genome-wide association studies have rarely established causality between genetic variants and obesity or type 2 diabetes, despite the clearly established importance of heritability in the development of these ailments1,2,3. Thus, other forms of variation are likely involved, such as epigenetic factors, that may be able to confer changes to the genomic material to be inherited without altering the genetic code. In fact, compelling evidence indicates that the deleterious effects of maternal and paternal obesity are strongly linked to an increased risk for developing obesity-related pathologies, including, type 2 diabetes and cardiovascular disease in the offspring4,5,6,7,8,9. Based on epidemiological studies, Barker and others10 formulated the hypothesis that nutritional quality during gestation may result in what is known as fetal programming, thereby predisposing the offspring to metabolic and cardiovascular disorders later in life. Several animal model studies, conducted in fruit flies, mice, and non-human primates, provided compelling evidence for this hypothesis. These studies collectively show that progeny of parents exposed to a high-fat diet (HFD) and/or high-sugar diet (HSD) during gestation are more susceptible to developing obesity and obesity-related diseases as adults11. However, the mechanistic basis for this phenomenon remains elusive.
To investigate whether parental exposure to a HFD increases the risk of obesity and cardiac lipotoxicity in the progeny, we used the fruit fly Drosophila as a model system. Drosophila with its short generation time is well suited for transgenerational studies and is well established as a powerful tool for discovering conserved gene networks in cardiac development, function, and metabolism12,13,14. In Drosophila, both high-fat and high-sugar diets increase fat storage and perturb metabolic homeostasis that are classic hallmarks of metabolic syndrome and diabetic cardiomyopathy15,16,17. Additionally, recent studies suggest that these extreme diets may have long-lasting effects over multiple generations18,19.
Here, we establish a Drosophila model to study the intergenerational inheritance of HFD-induced lipotoxic cardiomyopathy and present first evidence of a mechanism underlying this phenomenon. Our data indicate that reduced expression levels of adipose triglyceride lipase (ATGL; brummer (bmm) in Drosophila) and peroxisome proliferator-activated receptor-γ coactivator 1α (PGC-1α, PGC-1 in Drosophila) are closely associated with intergenerational lipotoxic cardiomyopathy. Furthermore, we find that cardio-protection can be achieved by cardiac or systemic bmm overexpression, which can lead to re- and even pre-programming of the metabolic state of a developing or adult individual. Finally, we demonstrate that intergenerational inheritance of HFD-induced lipotoxic cardiomyopathy coincides with elevation of systemic levels of trimethylated H3K27 (H3K27me3). Importantly, overexpression of the H3K27me3 demethylase, UTX, or pharmacological inhibition of the H3K27 methyltransferase, polycomb repressive complex 2 (PRC2) enzymatic subunit EzH2, could reverse the elevated H3K27me3 levels and prevent intergenerational transmission of parental HFD-induced cardiac lipotoxicity.
HFD-induced cardiac lipotoxicity is inherited
To investigate the consequences of a parental HFD, we evaluated cardiac and metabolic function in three generations of female offspring fed a normal food diet (NFD, diagram in Fig. 1a). Strikingly, we found that the effects of a HFD on heart function were evident not only in the HFD-fed parents themselves15, but also in their first- and second-generation progeny raised on NFD. Both progeny generations displayed dramatic detrimental changes in heart function, including decreased diastolic and systolic diameters, reduced fractional shortening, reduced heart beat length (heart period), and increased incidence of non-contractile myocardial cells, partial conduction blocks, and dysfunctional inflow valves, called ostia (Fig. 1b–g). The impaired heart phenotypes returned to normal in the third generation of offspring on a NFD (Fig. 1b–d, g).
HFD also caused increased fat accumulation in the parents, as observed previously15,16, which persisted in the progeny of HFD-fed parents, manifested as higher fat content in embryos at stages 1–15 (early to late stage embryogenesis) and subsequently as bigger and heavier adult offspring (Fig. 2b–c). The increase in body mass is likely because HFD augments growth-promoting Target of rapamycin signaling15. However, first progeny generation adults did not exhibit an increase in fat content relative to body weight (Fig. 2d). In addition, second-generation progeny of HFD-fed parents raised on a NFD presented no significant increase in fat content at embryo stages 1–15 or as adults, and were no longer significantly heavier (Fig. 2b–d). Interestingly, the offspring’s increase in fat accumulation at late embryonic stages was not due to increased maternal contribution, since early embryos (stages 1–4, before the start of zygotic transcription) exhibited no difference in fat content when parents or grandparents were fed a HFD, compared to embryos from NFD-fed parents (Fig. 2a).
HFD induces intergenerational reduction of ATGL expression
Previous work suggested that HFD reduces the expression of bmm in Drosophila15. Therefore, we asked whether parental exposure to a HFD affected bmm transcript levels in subsequent generations. We found no differences in the levels of maternally deposited bmm transcripts when early, pre-zygotic embryos from HFD-fed parents or grandparents were analyzed, in comparison to embryos from NFD-fed flies across generations (Fig. 2e). In addition, bmm transcript levels of embryos from HFD-fed parents or grandparents were also unchanged later in development (Fig. 2f). In contrast, we found that parental HFD decreased total adult bmm transcript levels for two generations, returning to normal only in the third generation (Fig. 2g).
The persistent effects of a HFD in one generation on future generations can be explained in two ways: (1) an ‘intergenerational’ effect that lasts for two generations following exposure of the parents (G0) to the initial dietary insult, presumably due to the direct exposure (within the parent) of the first-generation (G1) oocyte/egg and its germplasm (future G2); and (2) a ‘transgenerational’ effect that persists up to or beyond the third generation11. Our data suggest that parental HFD induces an ‘intergenerational’ effect (Figs. 1g and 2g), since both heart dysfunction and reduction of bmm transcript levels persist for only two generations after parental exposure to a HFD.
Cardiac lipotoxicity caused by PGC-1 depletion is inherited
In order to validate the intergenerational model of HFD-induced lipotoxic cardiomyopathy, we genetically induced lipid accumulation and tested the effects on heart function and metabolism in the next generation. For this purpose we used the previously characterized heterozygous PGC-1/spargel mutant flies (PGC-1XP)16. PGC-1 is a potent transcriptional co-activator that mediates mitochondrial activity in response to insulin signaling20. Reduction of PGC-1 function mimics the effects of a HFD on heart function and lipid content by acting downstream of bmm in Drosophila16. Specifically, heart dysfunction phenotypes of PGC-1XP heterozygotes (and of another PGC-1 allele) have previously been found to be the same as those of heart-specific PGC-1 knockdown16. Additionally, heart dysfunction of PGC-1XP heterozygotes was rescued by a genomic construct (PGC-1GR20), as reported by Diop et al.16, thus providing strong evidence that cardiomyopathy observed in these PGC-1XP heterozygotes is due to loss of PGC-1 function.
Indeed, we found in control flies that HFD reduced transcript levels of PGC-1 for two generations, suggesting that reduced PGC-1 expression is also associated with intergenerational inheritance of HFD-induced lipotoxic cardiomyopathy (Fig. 3a), and may be sufficient for its persistent effect across generations. Thus, we proceeded to study the effect of parental PGC-1 depletion on heart function of the offspring. PGC-1XP heterozygous mutants (PGC-1XP/+) were outcrossed to control flies and then intercrossed (G0). Their progeny (G1) was assessed for heart function and lipid content (Fig. 3b). We found that parental PGC-1 heterozygosity had a similar detrimental effect on heart function in the genetically wild-type progeny (+/+ for PGC-1), as on acute HFD or their PGC-1XP/+ siblings; they all exhibited increased incidence of dysfunctional ostia, non-contractile myocardial cells, partial conduction block resulting in asynchronous contractions (Fig. 3c), and decreased heart period (Fig. 3d).
As reported previously, we found that PGC-1XP/+ flies have increased fat content, similar to HFD-fed flies16, but +/+ flies did not exhibit significant differences in triglyceride levels compared to NFD-fed flies. However, HFD triggered a stronger increase in fat content in these +/+ flies from PGC-1XP/+ parents (6th bar, Fig. 3e) compared to control flies (2nd bar, Fig. 3e). These results provide genetic evidence to support the intergenerational inheritance of lipotoxic cardiomyopathy induced by a parental obesogenic metabolic state.
bmm expression reverts inherited cardiac lipotoxicity
Cardiac lipotoxicity has been found to be prevented by lipolysis induction upon genetically elevating bmm expression in Drosophila15, and manipulating ATGL in mammals and humans21,22. Since the intergenerational HFD-induced heart phenotypes are accompanied by reduced bmm expression, we used the GAL4/UAS system23 to overexpress bmm at different stages of development. We found that transgenic expression of bmm in the adult and embryonic myocardium, using the cardiac GMH5-Gal4 driver24, or the embryonic cardiac mesoderm TinD-Gal4 driver25, protected flies from HFD-associated cardiac dysfunction (Fig. 4a, b and Supplementary Figure 3a-e; also see Supplementary Figure 1 and Supplementary Figure 2 for genetic controls and driver specificity, respectively). Remarkably, expression of bmm in myocardial progenitors and adult cardiomyocytes protected the adult heart not only from the adverse effects of the parental HFD, but also from acute exposure to a HFD (Fig. 4b). In contrast, transgenic expression of bmm in adult and/or embryonic cardiomyocytes did not prevent the systemic increase in triacylglyceride (TAG) levels upon acute exposure to HFD (Fig. 4c). This suggests that autonomous expression of bmm in cardiomyocytes counteracts the effects of HFD and parental HFD in the heart. Moreover, it also implies that cardiac progenitors can be autonomously re- and even pre-programmed in the early embryo to protect against pathological heart phenotypes later in life.
Adipose bmm expression reverts inherited cardiac lipotoxicity
In Drosophila, the major site for lipid storage is the fat body, which may systemically modulate cardiac lipotoxicity. We first validated the specificity of commonly used fat body drivers. By performing RNAscope® against Gal4 and immunohistochemistry we found that both Lsp2-Gal426 and ppl-Gal427 drive strong expression in fat body of young flies; and while Lsp2-Gal4 is specific for fat body, ppl-Gal4 showed a mild expression in pericardial cells (Supplementary Figure 2c,d,k and m). Indeed, expression of bmm in the larval and adult fat body, using these two different drivers (Lsp2-Gal4 and ppl-Gal4), prevents heart dysfunction following parental or acute exposure to a HFD (Fig. 4d, e; and Supplementary Figure 3f-j. See also Supplementary Figure 1a-c for genetic controls). In line with these results, specific overexpression of bmm using Lsp2-Gal4 or ppl-Gal4 resulted in a moderate decrease of fat content and prevented HFD-induced fat accumulation (Fig. 4f). These results corroborate the association of metabolic imbalance and obesity to increased risk for heart dysfunction.
Early embryonic bmm expression is cardioprotective
We further tested the intergenerational metabolic reprogramming model by determining whether genetic manipulation of the very early embryo (G1) and the paternal germline of the next generation (G2) was sufficient to protect the adult G1 and G2 progeny from the adverse effects of a parental (or acute) HFD. For this purpose, we expressed bmm with a germline-specific driver, nos-Gal4, which resulted in bmm expression in the embryo at the mid-blastoderm stage (zygotic transcription start) through gastrulation as well as in the germ cells that give rise to the following generation28 (Fig. 5a, b and Supplementary Figure 4a, b). We found that bmm expression in the blastoderm embryo (G1) and its germline (future G2) protected G1 adult progeny from deleterious cardiac lipotoxicity due to parental exposure to a HFD (Fig. 5c; Supplementary Figure 1 and Fig. 4c-g). Importantly, this early embryonic G1 expression of bmm also conferred protection from HFD when the adult G1 flies were fed a HFD acutely (Fig. 5c, 6th and 8th bars). This suggests that early embryonic bmm expression protects the adult heart from both acute and parental HFD lipotoxicity. Next, male flies expressing bmm under the control of nos-Gal4 driver were outcrossed to w1118 and fed a NFD or a HFD (Fig. 5b). Expression of bmm in the early embryo in G1 and the germ cells that become the G2 progeny conferred cardiac protection to acute or parental HFD in G2 flies (Fig. 5e and Supplementary Figure 4h-l). Similar re- and pre-programming effect was achieved when bmm expression was restricted to Drosophila adipose tissue during embryonic to adult stage in the offspring (G1) of HFD-fed parents, using Lsp2-Gal4 driver. Expression of bmm during early development protected the following generation (G2) from the effects of parental or acute exposure to a HFD (Supplementary Figure 5a-d).
In line with these results, bmm expression using either nos-Gal4 or Lsp2-Gal4 drivers both prevented HFD-induced fat accumulation in G1 and G2 flies (Figs. 4f, 5d, f and Supplementary Figure 5e). These data show that augmenting lipolysis in the early embryo or in adipose tissue is sufficient to re- or pre-program metabolism to shield the adult heart against parental, acute, or future HFD.
Lipotoxic cardiomyopathy inheritance correlates with H3K27me3
Changes in gene expression could be transmitted from one generation to the next through inheritance of epigenetic alterations29,30,31,32. Given the emerging relationship between nutrient availability and epigenetic regulation of gene expression9,33,34, we sought to investigate the relationship between intergenerational lipotoxic cardiomyopathy inheritance, reduced expression levels of ATGL/bmm and PGC-1, and epigenetic repression of gene expression. In Drosophila, maternal HFD and HSD are known to modify the expression of metabolic genes in adult offspring18,35 and paternal HSD can modify the offspring’s chromatin state and gene transcription in a H3K9/K27me3-dependent manner19. In relation to heart function, Ubiquitously transcribed tetratricopetide repeat gene on the X chromosome (UTX), a highly conserved H3K27 demethylase, has been shown to promote a cardiac-specific gene program36, while the catalytic subunit of the PRC2 complex, the H3K27 methyltransferase EzH2, is required to stabilize postnatal cardiac gene expression by promoting gene repression37. Thus, we first monitored global levels of H3K27me3 to explore the involvement of the H3K27me3 repressive marks in the intergenerational inheritance of lipotoxic cardiomyopathy. Indeed, acute exposure of adult flies to a HFD caused a significant increase in whole-body H3K27me3 levels that persisted in the next NFD-fed generation (Fig. 6a and Supplementary Figure 8a).
To determine whether H3K27me3 is a key epigenetic regulator of intergenerational HFD-related pathology, we reduced H3K27me3 levels by overexpressing UTX. Adipose-restricted expression of Drosophila UTX (dUTX38) using Lsp2-Gal4 not only prevented the HFD-induced increase in H3K27me3 levels but also significantly increased bmm messenger RNA (mRNA) levels, even under acute or parental HFD conditions (Fig. 6a, b). Interestingly, the elevated levels of bmm expression due to dUTX overexpression in animals fed NFD are mildly reduced by an acute HFD but not a parental HFD (Fig. 6b, right panels). Therefore, acute exposure to a HFD may not lower bmm transcription solely by regulating H3K27me3 levels. However, it is important to note that bmm transcript levels in HFD-fed dUTX expressing flies were similar to the levels in NFD-fed control flies (Fig. 6b).
In order to study the effect of manipulating H3K27me3 levels on heart function, we next used heart-specific drivers to overexpress UTX. Cardiac-specific (GMH5-Gal4) or cardiac progenitor-specific (TinD-Gal4) dUTX expression effectively protected against cardiomyopathies due to acute or parental HFD exposure (Fig. 6c and Supplementary Figure 6a-e), as was observed by overexpressing bmm using these same drivers. It should be noted that dUTX expression in the embryonic and adult myocardium, but not in cardiac progenitors, prevented systemic fat accumulation upon acute HFD feeding (Fig. 6d). This contrasts with GMH5-driven bmm expression, which had no effect on HFD-induced systemic fat accumulation (Fig. 4c). Taken together, these findings suggest that genetic manipulations in adult myocardial cells differ in their potential to affect systemic metabolism. Moreover, Lsp2-Gal4- and ppl-Gal4-driven adipose-restricted expression of dUTX also provided protection against HFD-associated heart dysfunction and HFD-induced fat accumulation across generations (Fig. 6e, f and Supplementary Figure 6f-j). In line with an increase in bmm transcript levels, Lsp2>UTX flies presented a mild reduction in fat content in NFD conditions when compared to NFD control flies.
Finally, we found that nos-Gal4-driven expression of dUTX specifically in the early blastoderm embryo (G1) and future germline (G2) preserved all aspects of heart function and prevented excess fat accumulation following acute or parental exposure to a HFD (Fig. 6g, h and Supplementary Figure 7), similar to observations in animals with nos-Gal4-driven overexpression of bmm (Fig. 5c–i). These data support the conclusion that dUTX overexpression and thus reduction of H3K27me3 levels, as early as in the germline, has a profound effect on metabolic state and cardiac function later in life.
EzH2 inhibition prevents HFD-induced lipotoxic cardiomyopathy
Since inhibition of H3K27me3 by dUTX overexpression protects the fly from the deleterious effects of both acute and parental HFD exposure, we also examined whether reducing H3K27me3 levels pharmacologically, by feeding the flies with an inhibitor of the H3K27 methyltransferase subunit EzH2 of the PRC2, is also protective39. We found that exposure of adult flies to the EzH2 inhibitor, MC1945, prevented the global increase in H3K27me3 levels observed upon acute and parental HFD conditions in control dimethyl sulfoxide (DMSO)-treated flies (Fig. 7a, and Supplementary Figure 8a). Importantly, flies exposed to MC1945 exhibited an increase in bmm expression under NFD and no significant reduction in transcript levels upon acute or parental HFD (Fig. 7b), which is strikingly similar to the phenotype induced by adipose-restricted dUTX expression (compare to Fig. 6b). Notably, pharmacological inhibition of H3K27me3 production with 25 μM of MC1945 (selected from dose–response studies; Supplementary Figure 8b-e) had no detectable effect on cardiac function or TAG content of NFD-fed flies, but significantly reduced the percentage of flies with heart dysfunctions triggered by an acute HFD and fully protected the flies from fat accumulation upon HFD exposure (Fig. 7c, d). Next, we collected the progeny from flies that had been exposed to HFD or NFD containing EzH2 inhibitor (or DMSO). As expected, EzH2 inhibitor (in G0) also protected the offspring from the detrimental effects of parental exposure to a HFD on heart function (Fig. 7e and Supplementary Figure 8f-j). Remarkably, however, such parental EzH2 inhibition also reduced or eliminated heart dysfunctions when their progeny were exposed to an acute HFD (in G1) (Fig. 7e and Supplementary Figure 8f-j). In line with these results, parental exposure to EzH2 inhibitor also prevented triglyceride accumulation in the progeny that was fed an acute HFD (Fig. 7f). While further experiments are required to determine the direct regulation of bmm transcript levels and the specific gene targets of EzH2 and dUTX, we provide multiple lines of evidence to demonstrate that exposure to HFD can elevate H3K27me3 across the genome and that this effect persists into the next generation, thereby mediating the inheritance of lipotoxic cardiomyopathy.
Several studies have established a critical role for TAG hydrolysis in cardiac metabolism and function, in both healthy and diseased hearts21,40,41. Here we show that acute HFD (5 days of food supplemented with 30% coconut oil) induces lipotoxic cardiomyopathy that can be inherited by the next two generations, via the parental germlines, even when the offspring are raised on NFD. Similar to what has been described by epidemiological studies on the offspring of obese pregnant women42, we found in our fly model that parental HFD led to first generation progeny with increased adult body weight and increased fat content in late-stage embryos. This was no longer the case in second generation offspring, and adult progeny did not exhibit an increase in fat content relative to body weight in either generations. In contrast, metabolic reprogramming across generations was particularly evident in the systemic reduction in the transcript levels of ATGL/bmm lipase and its downstream target PGC-1/spargel, a key regulator of energy metabolism. We further validated the intergenerational lipotoxic cardiomyopathy model by genetically reducing PGC-1 expression in the parents (PGC-1xp heterozygotes), which causes similar cardiac lipotoxicity as HFD exposure16. This was sufficient to alter the +/+ offspring’s metabolic state, leading to lipotoxic cardiomyopathy later in life, even though these flies carry two wild-type copies of PGC-1. Of note, the partial reduction of PGC-1 induced by acute and parental HFD in control flies and PGC-1XP heterozygous mutant flies is likely having a profound effect on mitochondrial biogenesis that could be underlying the observed heart dysfunction.
The experiments presented here indicate that HFD and reduced PGC-1 expression levels have the ability to modify the offspring’s metabolism leading to heart dysfunction. Thus, we speculated that parental HFD-dependent metabolic reprogramming and associated lipotoxic cardiomyopathy in the progeny could be prevented by increasing ATGL/bmm levels. Indeed, targeted transgenic expression of bmm is able to reset the altered metabolic state induced by parental HFD, and thus protects the progeny from cardiac lipotoxicity. Remarkably, induction of bmm expression in the early embryo is sufficient to render the adult progeny (G1), as well as the following generation (G2), resistant to acute HFD.
The presented data support the hypothesis, first formulated by Barker43, that the quality of nutrition during gestation leads to fetal programming that functions as a key determinant in establishing predisposition and/or susceptibility to metabolic and cardiovascular disorders later in life. While this hypothesis is sustained by several epidemiological studies, including the Dutch Hunger Winter studies44, and many animal models with different environmental stressors45, the underling mechanisms on the intergenerational inheritance of lipotoxic cardiomyopathy induced by parental HFD remain mostly unknown. The Drosophila model established here indicates that the inheritance of altered histone modifications is a key mechanism in the propagation of lipotoxic cardiomyopathy across generations.
Histone modifications are essential in regulating chromatin packaging and gene expression for proper development and cell function. In turn, metabolites serve as co-factors for chromatin modifying enzymes, allowing protein activity and gene expression to match the specific energy requirements41. We found that HFD, and potentially its maladaptive fluctuation in metabolites, can lead to changes in chromatin structure and gene expression that are transmitted to the next generation. Indeed, a major finding presented here is the involvement of the PRC2 complex in regulating metabolism and heart function in response to a HFD, via H3K27me3 gene repression across generations. The increase in H3K27me3 global levels in the adult flies upon acute or parental HFD supports the hypothesis that HFD causes changes in the epigenome that have lifelong consequences. Our results are in line with previous findings in Caenorhabditis elegans and mice, which provide evidence that PRC2-mediated epigenetic modifications in the germ cells can be transmitted to embryos by sperm and/or oocytes46,47. In addition, we found that reduction of H3K27me3, either by overexpression of UTX or inhibition of EzH2, can prevent the deleterious effects of a parental HFD on heart function and metabolism. Of note, at this point we cannot rule out that UTX overexpression and EzH2 inhibition might also have H3K27me3-independent effects.
Overall, using the genetic model of Drosophila, we provide first evidence that early embryonic and tissue-specific modulation of lipolysis in myocardial progenitors, adipose tissue, and the germline leads to tissue-specific and/or systemic metabolic reprogramming or pre-programming that persists into adulthood. Importantly, the imposed metabolic state appears to be inherited by the next two generations as either a predisposition to metabolic imbalance and cardiac dysfunction, or a protection from HFD insult when bmm is overexpressed. Moreover, we provide evidence that metabolic re-programming that leads to fat accumulation and cardiac lipotoxicity correlates with overall levels of H3K27me3 epigenetic marks, which can be reversed by genetic or pharmacological reduction of H3K27me3 levels. Our findings shed light on possible causes of obesogenic heritability and early adult onset of cardiovascular disease and diabetes, which appear to have their roots in the diet and overall metabolic state of the parents. Importantly, we demonstrate that targeted genetic or pharmacological interventions in the progeny are able to counteract the detrimental effects on cardiac function of parental dietary insults, a protection that persists even in the subsequent generation against acute HFD. These findings provide new perspectives for tackling metabolic syndrome effects across generations and preventing lipotoxic cardiomyopathies.
We obtained w1118 flies (used as controls in Figs. 3–7) from the Bloomington Drosophila Stock Center. We used multiple drivers for each tissue type. For expression in the fat body we used Lsp2-Gal426 and ppl-Gal427; for the heart, we used GMH5-Gal448 and TinD-Gal425. The UAS-bmm overexpression line was provided by Ronald P. Kühnlein (Max-Plank Institute)49, nos-Gal4 flies28 were a gift from R. Zhou (Sanford Burnham Prebys Medical Discovery Institute), and the UAS-dUTX38 flies were provided by A. Bergmann (University of Massachusetts Medical School). PGC-1XP (CG9809d04518) flies were obtained from Harvard Medical School. All flies were maintained on a NFD composed of yeast, corn starch, and molasses (10% yeast, 12% sugar, and 1.5% agar)50. The HFD was made by mixing 30% (weight/volume) coconut oil with NFD. This formulation was used for all experiments because it gave the strongest phenotype and the most reproducible results15.
Intergenerational HFD and NFD feeding regimen
Male and female flies were collected after eclosion and aged for 5 days on NFD (40 flies per tube). Flies were then randomly placed on NFD or HFD for an additional 5 days. Each population was then placed on NFD (see schematic in Fig. 1a) and allowed to lay eggs for 3 days. The offspring were raised on NFD. After the flies emerged (following pupation), they were aged for 5 days on NFD, and then randomly placed on NFD or HFD for 5 more days. Next, mated females were used to assess heart function, to measure TAG levels, and to extract RNA. Additionally, 20 females and 20 males were placed on NFD and allowed to lay eggs for 3 days to create the subsequent generation. This protocol was repeated for 3 generations (see Fig. 1a). Flies were housed in a 21 °C incubator with 50% humidity and controlled 12 h light/dark cycle. Embryos were collected on agar plates following standard protocols51. In brief, flies were placed in bottles containing an agar plate with 10% concentrated grape juice on the bottom and incubated at 25 °C for the specific time lapses described (2 or 13 h). Embryos were then recovered from the agar plates and processed for TAG assessment or RNA extraction. For the EzH2 inhibition treatment, we ran a dose–response test in order to find the best concentration (see Supplementary Figure 8b-e). Then, 100 μl of DMSO (vehicle) or 25 μM EzH2 inhibitor (MC1945) was added to the surface of the vials containing NFD or HFD standard food.
Heart preparation and analysis
Surgical dissection and exposure of hearts, high-resolution video microscopy, and analysis of movies for cardiac function measurements were performed according to published procedures52. Female flies were dissected in oxygenated artificial hemolymph in order to expose their hearts within the abdomen. High-speed digital movies were recorded using HCImageLive software (Hamamatsu) and then analyzed in a group-blinded fashion for heart rate, arrhythmia, contractility, and fractional shortening using SOHA software52. The following formula was used to determine sample size for a significance level of 5% with a power of 90%: n = 1 + 21 × (standard deviation/d)2, where d = size of difference in means53. For most of the experiments, a sample size of 15–20 flies was adequate.
TAGs were measured as described by Diop et al.50. In brief, whole female flies were placed in empty vials for 30 minutes (min), then transferred to Eppendorf tubes in batches of 12, weighed, and either processed immediately or stored at −80 °C for later processing. Flies were pooled in groups of 3 and homogenized in 600 µl of phosphate-buffered saline (PBS) containing 0.05%Triton X-100 for 5 min at 10,000 rpm using a High Throughput ball bearing Homogenizer (Talboys). The homogenates were centrifuged at 3000 × g for 15 min, and 20 µl of the supernatant was transferred into 200 µl of Triglyceride Reagent (Thermo Electron Corp. catalog number TR22421/2780-250). The reaction mixture was placed on a shaker at 300 rpm and incubated at 37 °C for 10 min. The absorbance at 550 nm was then measured using SpectraMax M2e (Molecular Devices Corp.) and TAG content was calculated by comparison with a standard curve constructed with solutions of known concentrations of TAG (Pointe Scientific, catalog number T7531-STD). The Triglyceride Reagent contains a Lipoprotein lipase that hydrolyzes the triglycerides present in the sample and produces glycerol, which is processed by a series of reactions ultimately producing a red-colored dye whose absorbance is proportional to the concentration of triglycerides present in the sample54. The protein content of the same samples was measured using Quick Start Bradford reagent (Bio-Rad). TAG levels are expressed as TAG/protein for embryos and TAG/fly weight for adults and normalized to w1118 NFD flies.
The 1–2-week-old female flies were dissected in oxygenated artificial hemolymph to expose the heart and surrounding fat tissue. Samples were fixed for 20 min in 4% paraformaldehyde diluted in PBS, washed 3 times in PBS containing 0.3% Triton (PBS-T), and incubated overnight at 4 °C with the primary antibodies diluted in PBS-T/anti-GFP-102055, 1:1000 (Aves lab), and anti-α-spectrin (3A9)56, 1:50 (DSHB). Tissues were washed 3 times in PBS-T and incubated for 2 h at room temperature with secondary antibodies: Alexa Fluor 647 and Alexa Fluor 555, 1:500 (Invitrogen), and 4′,6-diamidino-2-phenylindole (DAPI; 0.5 µg ml−1, Invitrogen) in PBS-T. After 3 washes with PBS-T, samples were placed in PBS and subsequently mounted onto glass slides using Prolonged Gold (Invitrogen) as the mounting media. Images were acquired on a Zeiss AxioImager Z1 equipped with an Apotome (Carl Zeiss) and an ORCA-Flash 4.0LT Digital CMSO camera C11440 (Hamamatsu), using Zen 2.3 pro software. Images were analyzed using ImageJ 1.49 m software.
Real-time quantitative PCR (qPCR) analysis
To check the transcriptional expression of bmm and PGC-1 in adult flies, ~12 whole female flies were flash-frozen in liquid nitrogen and homogenized in 700 μl of ice-cold QIAzol Lysis Reagent (Qiagen) using a pellet pestle motor (Kontes) according to the manufacturer’s instructions. To check the transcriptional expression of bmm in embryos, these were recovered at specified time points from agar plates with a brush and dechorionation was performed using 50% bleach57. Next, embryos were placed into Eppendorf tubes containing ice-cold PBS-T and centrifuged for 2 min at 3300 × g. The PBS-T was then completely removed, and embryos were re-suspended in 300 μl of ice-cold QIAzol Lysis Reagent (Qiagen). Total RNA was extracted from adult whole-body or embryos homogenates using the miRNeasy kit (Qiagen). Then, 500 ng of RNA was used to synthesize complementary DNA with the QuantiTect Reverse Transcription (RT) Kit (Qiagen). RT-qPCR was performed according to the manufacturer’s protocol using FastStart Essential DNA Green Master and a LightCycler® 96 System (ROCHE). mRNA levels were calculated relative to w1118 NFD control flies using the ΔΔCt method. rp49 was used as an internal control to normalize the total amount of RNA. Primer sequences were as follows:
PGC-1 F: 5’-CCATTGACACGAGTCATAGAGAC-3’;
PGC-1 R: 5’-AGCATTATCCAAGTCAGCGTAA-3’;
rp49 F: 5’-GCTAAGCTGTCGCACAAATG-3’;
rp49 R: 5’- GTTCGATCCGTAACCGATGT-3’;
bmmA F: 5’-CCTTCACCCACTCAAGCATT-3’;
bmmA R: 5’-GCCGTGGAGCTAAAAGTCTG-3’.
RNA in situ hybridization (RNAscope®)
Probes against Gal4 were generated using RNAscope® (ACD) platform. Female flies were dissected for RNA in situ hybridization as previously described58 and the manufacture’s protocol was followed. Gal4-labeled hearts were mounted immediately after using Prolong Gold (Molecular Probes). Images were acquired using an Apotome.2 (Zeiss), ZEN 2.3 pro software and later analyzed using ImageJ.
Histone extraction and western blot analysis
Approximately 40 flies per sample were pulverized in liquid nitrogen, and histones were extracted. Briefly, each sample was suspended in Triton Extraction Buffer (TEB: PBS containing 0.5% Triton X-100, 2 mM PMSF, and 0.02% NaN3) and incubated on ice for 10 min. During this incubation, samples were vortex mixed every 3 min. Lysates were centrifuged at 6500 × g for 10 min at 4 °C. The formed pellets were washed with TEB, vortex mixed and centrifuged at 6500 × g for 10 min at 4 °C. The recovered pellet was then re-suspended in 0.2 N HCl overnight at 4 °C. Samples were centrifuged at 6500 × g for 10 min at 4 °C and the supernatant was transferred to a clean tube. The protein concentration of each sample was determined with a Micro BCA Protein Assay Kit (ThermoScientific). About 40 μg of protein per sample was resolved by sodium dodecyl sulfate–polyacrylamide gel electrophoresis, transferred to nitrocellulose membranes, and incubated overnight at 4 °C with anti-H3K27me338 (1:1000, Active Motif, catalog number 39155) or anti-H3 (1:1000, Millipore, catalog number 06-755) antibodies in Tris-buffered saline, 0.05% Tween 20 (TBS-T), containing 5% of bovine serum albumin (BSA). After 3 washes of 10 min with TBS-T, blots were incubated for 1 h at room temperature with horseradish peroxidase-conjugated anti-rabbit secondary antibody (Amersham) diluted 1:5000 in TBS-T containing 5% of BSA. Blots were washed again and bands were visualized using enhanced chemiluminescent reagent (ECL Prime Western Blotting Detection Reagent, Amersham). Densitometry quantification was achieved using ImageJ software.
Statistical analyses of TAG, mRNA levels, and H3K27me3 content was performed using one-way analysis of variance (ANOVA) and Dunnett’s multiple comparisons test against NFD among same genotypes or Sidak’s multiple comparisons test when different genotypes but same conditions were compared. Heart function measurements were analyzed using unpaired, two-tailed, Student’s t-test (for two group comparison) or one-way ANOVA followed by Sidak’s multiple comparisons test within same genotypes. Heart dysfunction phenotypes (partial conduction block, non-contractile cells in the heart, ostia defects) were analyzed using the chi-square test and one representative experiment is shown. Flies were assigned a numerical code per genotype/treatment so investigators were blinded when analyzing heart function. Outliers were identified using the ROUT method, Q = 1%. Statistical analysis was performed using GraphPad Prism (GraphPad Software, La Jolla, California, www.graphpad.com). Values are presented as means ± SEM. The p values < 0.05 were considered significant. All data presented here are representative of at least two independent experiments.
Further information on experimental design is available in the Nature Research Reporting Summary linked to this article.
The authors declare that the data supporting the findings of this study are available in the article and its Supplementary Information File. All other relevant data supporting the findings of this study are available from the corresponding author upon reasonable request.
Journal peer review information: Nature Communications thanks the anonymous reviewers for their contribution to the peer review of this work
Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
McCarthy, M. I. Genomic medicine at the heart of diabetes management. Diabetologia 58, 1725–1729 (2015).
Fall, T. & Ingelsson, E. Genome-wide association studies of obesity and metabolic syndrome. Mol. Cell. Endocrinol. 382, 740–757 (2014).
Willyard, C. Heritability: the family roots of obesity. Nature 508, S58 (2014).
Shankar, K. et al. Maternal obesity at conception programs obesity in the offspring. Am. J. Physiol. Regul. Integr. Comp. Physiol. 294, R528–538 (2008).
Barker, D. J. & Martyn, C. N. The maternal and fetal origins of cardiovascular disease. J. Epidemiol. Community Health 46, 8–11 (1992).
Woo, M. & Patti, M.-E. Diabetes risk begins in utero. Cell Metab. 8, 5–7 (2008).
Costa-Silva, J. H., Simões-Alves, A. C. & Fernandes, M. P. Developmental origins of cardiometabolic diseases: role of the maternal diet. Front Physiol. 7, 504 (2016).
Jääskeläinen, A. et al. Intergenerational transmission of overweight among Finnish adolescents and their parents: a 16-year follow-up study. Int. J. Obes. 35, 1289–1294 (2011).
Carone, B. R. et al. Paternally induced transgenerational environmental reprogramming of metabolic gene expression in mammals. Cell 143, 1084–1096 (2010).
Paneth, N. & Susser, M. Early origin of coronary heart disease (the “Barker hypothesis”). BMJ (Clinical research ed.) 310(6977), 411–412 (1995).
Sales, V. M., Ferguson-Smith, A. C. & Patti, M.-E. Epigenetic mechanisms of transmission of metabolic disease across generations. Cell Metab. 25, 559–571 (2017).
Bodmer, R. & Frasch, M. Development and Aging of theDrosophila Heart. In: Heart Development and Regeneration. 2, 47-86 (eds. N. Rosenthal & R. Harvey) (2010).
Ocorr, K., Vogler, G. & Bodmer, R. Methods to assess Drosophila heart development, function and aging. Methods 68, 265–272 (2014).
Diop, S. B. & Bodmer, R. Gaining insights into diabetic cardiomyopathy from Drosophila. Trends Endocrinol. Metab. 26, 618–627 (2015).
Birse, R. T. et al. High-fat-diet-induced obesity and heart dysfunction are regulated by the TOR pathway in Drosophila. Cell Metab. 12, 533–544 (2010).
Diop, S. B. et al. PGC-1/Spargel counteracts high-fat-diet-induced obesity and cardiac lipotoxicity downstream of TOR and brummer ATGL lipase. Cell Rep. 10, 1572–1584 (2015).
Na, J. et al. A Drosophila model of high sugar diet-induced cardiomyopathy. PLoS Genet. 9, e1003175 (2013).
Buescher, J. L. et al. Evidence for transgenerational metabolic programming in Drosophila. Dis. Model Mech. 6, 1123–1132 (2013).
Öst, A. et al. Paternal diet defines offspring chromatin state and intergenerational obesity. Cell 159, 1352–1364 (2014).
Tiefenböck, S. K., Baltzer, C., Egli, N. A. & Frei, C. The Drosophila PGC‐1 homologue Spargel coordinates mitochondrial activity to insulin signalling. EMBO J. 29, 171–183 (2010).
Haemmerle, G. et al. ATGL-mediated fat catabolism regulates cardiac mitochondrial function via PPAR-α and PGC-1. Nat. Med. 17, 1076–1085 (2011).
Pulinilkunnil, T. et al. Myocardial adipose triglyceride lipase overexpression protects diabetic mice from the development of lipotoxic cardiomyopathy. Diabetes 62, 1464–1477 (2013).
Brand, A. H. & Perrimon, N. Targeted gene expression as a means of altering cell fates and generating dominant phenotypes. Development 118, 401–415 (1993).
Wessells, R. J. & Bodmer, R. Screening assays for heart function mutants in Drosophila. BioTechniques 37, 58–60 (2004). 62, 64 passim.
Reim, I. & Frasch, M. The Dorsocross T-box genes are key components of the regulatory network controlling early cardiogenesis in Drosophila. Development 132, 4911–4925 (2005).
Cherbas, L., Hu, X., Zhimulev, I., Belyaeva, E. & Cherbas, P. EcR isoforms in Drosophila: testing tissue-specific requirements by targeted blockade and rescue. Development 130, 271–284 (2003).
Zinke, I., Kirchner, C., Chao, L. C., Tetzlaff, M. T. & Pankratz, M. J. Suppression of food intake and growth by amino acids in Drosophila: the role of pumpless, a fat body expressed gene with homology to vertebrate glycine cleavage system. Development 126, 5275–5284 (1999).
Van Doren, M., Williamson, A. L. & Lehmann, R. Regulation of zygotic gene expression in Drosophila primordial germ cells. Curr. Biol. 8, 243–246 (1998).
Attig, L. et al. Dietary alleviation of maternal obesity and diabetes: increased resistance to diet-induced obesity transcriptional and epigenetic signatures. PLoS ONE 8, e66816 (2013).
Dong, M., Zheng, Q., Ford, S. P., Nathanielsz, P. W. & Ren, J. Maternal obesity, lipotoxicity and cardiovascular diseases in offspring. J. Mol. Cell. Cardiol. 55, 111–116 (2013).
Sullivan, E. L., Smith, M. S. & Grove, K. L. Perinatal exposure to high-fat diet programs energy balance, metabolism and behavior in adulthood. Neuroendocrinology 93, 1–8 (2011).
Remely, M., de la Garza, A. L., Magnet, U., Aumueller, E. & Haslberger, A. G. Obesity: epigenetic regulation – recent observations. Biomol. Concepts 6, 163–175 (2015).
Leung, A., Trac, C., Du, J., Natarajan, R. & Schones, D. E. Persistent chromatin modifications induced by high fat diet. J. Biol. Chem. 291, 10446–10455 (2016).
Radford, E. J. et al. In utero effects. In utero undernourishment perturbs the adult sperm methylome and intergenerational metabolism. Science 345, 1255903 (2014).
Heinrichsen, E. T. et al. Metabolic and transcriptional response to a high-fat diet in Drosophila melanogaster. Mol. Metab. 3, 42–54 (2014).
Lee, S., Lee, J. W. & Lee, S.-K. UTX, a histone H3-lysine 27 demethylase, acts as a critical switch to activate the cardiac developmental program. Dev. Cell 22, 25–37 (2012).
Delgado-Olguín, P. et al. Epigenetic repression of cardiac progenitor gene expression by Ezh2 is required for postnatal cardiac homeostasis. Nat. Genet 44, 343–347 (2012).
Herz, H.-M. et al. The H3K27me3 demethylase dUTX is a suppressor of Notch- and Rb-dependent tumors in Drosophila. Mol. Cell. Biol. 30, 2485–2497 (2010).
Ciarapica, R. et al. The Polycomb group (PcG) protein EZH2 supports the survival of PAX3-FOXO1 alveolar rhabdomyosarcoma by repressing FBXO32 (Atrogin1/MAFbx). Oncogene 33, 4173–4184 (2014).
Kienesberger, P. C., Pulinilkunnil, T., Nagendran, J. & Dyck, J. R. B. Myocardial triacylglycerol metabolism. J. Mol. Cell. Cardiol. 55, 101–110 (2013).
Ritterhoff, J. & Tian, R. Metabolism in cardiomyopathy: every substrate matters. Cardiovasc Res 113, 411–421 (2017).
Yu, Z. et al. Pre-pregnancy body mass index in relation to infant birth weight and offspring overweight/obesity: a systematic review and meta-analysis. PLoS ONE 8, e61627 (2013).
Barker, D. J. Fetal origins of coronary heart disease. BMJ 311, 171–174 (1995).
Painter, R. C. et al. Transgenerational effects of prenatal exposure to the Dutch famine on neonatal adiposity and health in later life. BJOG 115, 1243–1249 (2008).
Blake, G. E. & Watson, E. D. Unravelling the complex mechanisms of transgenerational epigenetic inheritance. Curr. Opin. Chem. Biol. 33, 101–107 (2016).
Gaydos, L. J., Wang, W. & Strome, S. Gene repression. H3K27me and PRC2 transmit a memory of repression across generations and during development. Science 345, 1515–1518 (2014).
Siklenka, K. et al. Disruption of histone methylation in developing sperm impairs offspring health transgenerationally. Science 350, aab2006 (2015).
Wessells, R. J., Fitzgerald, E., Cypser, J. R., Tatar, M. & Bodmer, R. Insulin regulation of heart function in aging fruit flies. Nat. Genet. 36, 1275–1281 (2004).
Grönke, S. et al. Brummer lipase is an evolutionary conserved fat storage regulator in Drosophila. Cell Metab. 1, 323–330 (2005).
Diop, S. B., Birse, R. T. & Bodmer, R. High fat diet feeding and high throughput triacylglyceride assay in Drosophila Melanogaster. J Vis Exp 127, 56029 (2017).
Rothwell, W. F. & Sullivan, W. Drosophila embryo collection. Cold Spring Harb. Protoc. 2007, pdb.prot4825 (2007).
Ocorr, K. et al. Genetic control of heart function and aging in Drosophila. Trends Cardiovasc. Med. 17, 177–182 (2007).
National Research Council (US) Committee on Guidelines for the Use of Animals in Neuroscience and Behavioral Research Research. Sample Size Determination (National Academies Press, Washington, 2003).
Hildebrandt, A., Bickmeyer, I. & Kühnlein, R. P. Reliable Drosophila body fat quantification by a coupled colorimetric assay. PLoS ONE 6, e23796 (2011).
Lin, C.-C. & Potter, C. J. Editing transgenic DNA components by inducible gene replacement in Drosophila melanogaster. Genetics 203, 1613–1628 (2016).
Dubreuil, R., Byers, T. J., Branton, D., Goldstein, L. S. & Kiehart, D. P. Drosophilia spectrin. I. Characterization of the purified protein. J. Cell Biol. 105, 2095–2102 (1987).
Rothwell, W. F. & Sullivan, W. Drosophila embryo dechorionation. Cold Spring Harb. Protoc. 2007, pdb.prot4826 (2007).
Blice-Baum, A. C. et al. Modest overexpression of FOXO maintains cardiac proteostasis and ameliorates age-associated functional decline. Aging Cell 16, 93–103 (2017).
We thank Ronald P. Kühnlein, R. Zhou, and A. Bergmann for generously sharing fly stocks, as well as the DSHB for antibodies. The authors are grateful for expert technical assistance by Hannah Catan, Selma Garcia, Nathan Troup, and Kathryn Reardon. We also thank Sean Oldham for his involvement in the early phases of this work, Karen Ocorr for insightful discussions, and Stan Walls and Anjali Gupta for critically reading the manuscript. This work was supported by grants from the American Heart Association by a Scientist Development Grant to R.T.B. and from the National Institutes of Health (R01 HL054732 and P01 AG033456) to R.B.
About this article
Genes & Nutrition (2019)