High sugar diet–induced fatty acid oxidation potentiates cytokine-dependent cardiac ECM remodeling

The extracellular matrix (ECM) is a dynamic three-dimensional scaffold of fibrous proteins, proteoglycans, and glycoproteins organized in a cell/tissue-specific manner that continually undergoes controlled remodeling. Besides providing physical support, the ECM profoundly influences a diverse array of cellular functions during development and throughout life (Bonnans et al., 2014; Lu et al., 2011; Rozario and DeSimone, 2010). On the flip side, defects in ECM composition cause major developmental defects and strongly contribute to several pathological conditions such as cancer, osteoarthritis, and fibrosis (Bonnans et al., 2014; Frantz et al., 2010; Lu et al., 2011). Understanding the regulatory mechanisms of ECM remodeling is thus essential to comprehend the complex ECM dynamics and its implications in organismal health and disease.

Emerging studies demonstrate that substantial ECM remodeling is associated with increased consumption of high-calorie diets. At times, it could lead to fibrosis, characterized by an excessive or inappropriate accumulation of ECM components in various tissues. For instance, it has been documented that significant changes in cardiac ECM turnover lead to interstitial and perivascular fibrosis due to high dietary sugar (Borghetti et al., 2018; Tate et al., 2017). In both human patients and animal models of diabetes, the roles of diabetes-associated hyperglycemia, advanced glycation end products, reactive oxygen species (ROS), neurohumoral signals, inflammatory cytokines, growth factors, and adipokines have been implicated in the activation of the resident cardiac fibroblasts for the excessive synthesis of cardiac ECM proteins (Russo and Frangogiannis, 2016; Tuleta and Frangogiannis, 2021). Several signaling molecules including TGF β1, Angiotensin II, ROS, and STAT-3 facilitate the process of fibrosis either by enhancing collagen synthesis in the fibroblasts or by inducing their proliferation (Fiaschi et al., 2014; Frangogiannis, 2019, 2021; Russo and Frangogiannis, 2016). Furthermore, other myocardial cell types, including immune cells, cardiomyocytes, endothelial cells, and pericytes, have been documented to contribute to diabetic cardiac fibrosis. Either they are involved in the secretion of profibrotic mediators and fibrogenic growth factors, or they get transdifferentiated into fibroblasts. Nevertheless, the mechanistic basis of high dietary sugar–induced cardiac ECM remodeling and its subsequent contribution to the manifestation of cardiomyopathy remains poorly understood and forms the central focus of this study.

The fruit fly, Drosophila melanogaster, is a versatile model organism that has been extensively used in biomedical research for the last few decades. In particular, the heart of Drosophila provides a simple yet elegant genetic model to examine ECM turnover and fibrosis. The adult heart of a Drosophila, also known as the dorsal tube, is a contractile tubular structure that spans the midline of the abdomen, just beneath the dorsal cuticle (Fig. 1 A). Anteriorly, it extends as the aorta into the thorax. On either side, the heart is flanked by a series of pericardial cells (PCs) that primarily function as nephrocytes (Helmstädter et al., 2017; Zhang et al., 2017). A meshwork of proteoglycans, glycoproteins, and fibrous proteins constitute the cardiac ECM surrounding the heart (Pastor-Pareja and Xu, 2011; Rodriguez et al., 1996; Yasothornsrikul et al., 1997) (Fig. 1 B). The cardiac ECM is instrumental in connecting the cardiomyocytes with the adjoining PCs and muscles, restoring diastolic heart diameter, and participating in synchronization of cardiomyocyte contraction (Rotstein and Paululat, 2016). The fibrous protein Pericardin (Prc), with homologies to mammalian collagen IV (Chartier et al., 2002), is a critical component of the cardiac ECM. Loss of Prc affects cardiac morphogenesis and cardiac integrity, leading to abnormal heart function and reduced lifespan (Chartier et al., 2002; Drechsler et al., 2013; Gera et al., 2022). In adult flies, Prc is synthesized and secreted by the fat cells. Quite intriguingly, an interorgan communication network between the PCs and the fat body regulates the expression of prc in the fat cells. High levels of physiological ROS in the PCs activate the Ask1-JNK/p38 signaling cascade to trigger the expression of the cytokine Unpaired 3 (Upd3). In turn, Upd3 released by the PCs activates JAK-STAT signaling in the fat cells to turn on the expression of prc. Prc released by the fat cells decorates the cardiac ECM (Gera et al., 2022).

Though insect physiology and carbohydrate metabolism bear specific differences with mammals, many studies have evidenced that the molecules and signaling pathways involved in nutrient sensing and sugar homeostasis in Drosophila are evolutionarily conserved and functionally similar to those observed in mammals (Chatterjee and Perrimon, 2021; Mattila and Hietakangas, 2017). Analogous to the mammalian insulin/glucagon endocrine system, Drosophila insulin-like peptides (dILPs) and glucagon-like adipokinetic hormone play a central role in glucose metabolism (Baker and Thummel, 2007). Moreover, not only are the members of the insulin signaling pathway conserved (Garofalo, 2002), but mutations in the genes coding for them disrupt sugar homeostasis (Böhni et al., 1999; Kim and Neufeld, 2015; Murillo-Maldonado and Riesgo-Escovar, 2017). Strikingly, elevated dietary sugar levels manifest several hallmark features of type 2 diabetes (T2D) and lead to progressive heart failure accompanied by fibrosis-like accumulation of Prc in the cardiac ECM of adult flies (Musselman et al., 2011; Na et al., 2013). However, the molecular and genetic basis of high dietary sugar induced increased Prc accumulation remains elusive.

Our results unravel the mechanistic basis of the process involved in remodeling cardiac ECM upon high dietary sugar intake. Based on in vivo molecular genetic analysis, we demonstrate that high dietary sugar causes ROS-independent activation of JNK signaling to promote elevated levels of fatty acid oxidation (FAO) in the PCs. Importantly, we establish the role of high FAO–induced histone acetylation in transcriptional upregulation of the cytokine Upd3, responsible for enhancing prc expression. Thus, besides revealing the molecular genetic basis of high dietary sugar–induced increased accumulation of Prc in the cardiac ECM of Drosophila, the outcome of this study sheds light on an otherwise unappreciated physiological role of FAO in ECM remodeling by modulating cytokine levels with far-reaching implications in diabetic fibrosis.

For our studies, we compared adult flies reared on high sugar diet (HSD; standard cornmeal-sugar-yeast-agar medium brought to a final concentration of 1 M sucrose) to those reared on a normal diet (ND; standard cornmeal-sugar-yeast-agar medium with 0.15 M sucrose). For the first 2 days after eclosion (AE), the control and experimental batches of flies were reared on ND. Subsequently, the experimental flies were transferred to HSD, while the control batches of flies were reared on ND for the rest of their life (Fig. 1 C). The hemolymph glucose level in flies reared on ND ranged between 63.91 ( ±4.8) mg/dl and 96.3 ( ±10.4) mg/dl till day 30AE (Fig. 1 D). In contrast, the flies fed on HSD exhibited a gradual increase in hemolymph glucose level, attaining a more than twofold increase by day 20AE (Fig. 1 D). Additionally, on day 20AE, the flies reared on HSD exhibited hypertrehalosemia (Fig. S1 A), increased triglyceride levels (Fig. S1 B), and insulin resistance (Fig. S1, C and C′) despite an increase in the production of insulin-like peptides, DILP2, DILP3, and DILP5, primarily associated with carbohydrate metabolism (Fig. S1 D). The HSD-fed flies demonstrated shortened average life expectancy (Fig. S1 E), with the median lifespan being 7 days shorter than the ND-fed flies (Fig. S1 F). Importantly, on day 20AE, flies reared on HSD demonstrated an increased amount of Prc accumulation in the cardiac ECM (Fig. 1, E–G) and exhibited irregular beating patterns of the heart (Fig. 1 H and Videos 1 and 2) with a significant increase in arrhythmia index (Fig. 1 I) when compared to the flies reared on ND. Given that by day 20AE, HSD-fed flies demonstrated several hallmark features of T2D, fibrosis-like accumulation of Prc in the cardiac ECM, and cardiac dysfunction, as evidenced earlier (Musselman et al., 2011; Na et al., 2013), for the rest of the study, all analyses were done on day 20AE.

In adult flies, Prc is synthesized and released by the fat cells in a JAK-STAT–dependent manner. The cytokine Upd3, released by the PCs, triggers the JAK-STAT signaling in the fat cells (Fig. 1 J) (Gera et al., 2022). Our investigation revealed that the level of expression of prc is more than threefold higher in the fat cells of HSD-fed flies compared with those reared on ND (Fig. 1 K). In tune with this result, a robust increase in the reporter UAS-GFP expression driven by prc-Gal4 was detected in the fat cells of HSD-fed flies (Fig. 1, L and M). Notably, fat cell–specific attenuation of JAK-STAT signaling by knocking down the expression of either Stat92E (the signal transducer) or Domeless (the receptor) drastically reduced prc expression in the fat cells of HSD-fed flies to a level that was comparable with that observed upon similar genetic manipulations in ND-fed flies (Fig. 1 N and Fig. S1 G).

To ascertain whether the upregulation in prc expression in the fat cells of HSD-fed flies is induced by increased levels of upd3 expression in the PCs, we first checked for the levels of upd3 expression in the PCs. As evident from Fig. 2, A–C, compared with those reared on ND, a marked increase in reporter lacZ expression for upd3 was detected in the PCs of HSD-fed flies. An analogous increase in the transcript levels of upd3 was detected in the RNA isolated from the PCs and cardiac tube of HSD-fed flies (Fig. 2 D). Considering that upd3-lacZ expression is restricted to the PCs, we argue that the enrichment of upd3 transcripts is primarily due to elevated levels of upd3 expression only in the PCs. Next, we wanted to downregulate upd3 specifically in the PCs and check for prc expression in the fat cells. For that purpose, we knocked down upd3 in the PCs of HSD-fed flies by dot-Gal4, which specifically expresses in the PCs (Fig. S2 A; Gera et al., 2022). Downregulation of upd3 in the PCs of HSD-fed flies resulted in a drastic reduction in prc expression in the fat cells (Fig. 2 E and Fig. S2 B). Consistent with these results, a remarkable reduction in the excessive amount of Prc accumulation around the cardiac tube of HSD-fed flies was also detected upon knocking down upd3 in the PCs (Fig. 2, F–I). Together, these results establish that elevated levels of Upd3 from the PCs augment prc expression in a JAK-STAT–dependent manner in the fat cells of HSD-fed flies.

We then sought to determine whether the upd3 knockdown-dependent suppression of increased prc expression could restore normal cardiac function in HSD-fed flies. Though a detectable suppression of the irregularity in the heartbeats of HSD-fed flies was observed upon downregulating pericardial upd3 expression (Fig. 2 J and Videos 3, 4, and 5), the rate of heartbeats was still slower compared with that observed in the ND-fed flies. While an appreciable recovery was detected in arrhythmia index (Fig. 2 K) and systolic interval (Fig. S2 E), the heart period (Fig. S2 C) and the diastolic interval (Fig. S2 D) were still higher in HSD-fed flies upon knocking down upd3 in the PCs. We argue that complete restoration in cardiac function was not seen as the observed cardiac dysfunction in HSD-fed flies could be an outcome of several factors. Accumulation of excess Prc in the cardiac ECM might be one of the contributing factors. Nonetheless, these results establish a causal relationship between increased pericardial Upd3 and cardiac dysfunction associated with enhanced Prc accumulation in the cardiac ECM of HSD-fed flies.

Under normal dietary conditions, high physiological levels of ROS trigger a signaling cascade that involves Ask1 and JNK/p38 to regulate upd3 expression in the PCs (Fig. 3 A). Given that excessive ROS generation and oxidative stress are associated with the pathophysiology of diabetic cardiomyopathy (Kaludercic and Di Lisa, 2020), we conjectured that a further increase in the levels of ROS in the PCs could hyperactivate this signaling cascade to augment upd3 expression in the PCs of HSD-fed flies. Surprisingly, the intensity of dihydroethidium staining (DHE; a redox-sensitive dye that detects superoxide radicals) was drastically reduced in the PCs of HSD-fed flies (Fig. 3, B–D). This drop in ROS level was further confirmed by analyzing the expression of glutathione S-transferase D1-GFP (gstD1-GFP), an in vivo reporter for cellular ROS levels (Sykiotis and Bohmann, 2008). Compared with that observed in the PCs of flies reared on ND, reduced levels of gstD1-GFP expression were detected in the PCs of HSD-fed flies (Fig. S2, F–H). On a similar note, a reduction in the level of phospho-p38 was observed in the PCs of HSD-fed flies (Fig. 3, E–G). However, in sharp contrast, a robust increase in TRE-dsRed (reporter for JNK signaling) expression was detected in the PCs of HSD-fed flies (Fig. 3, H–J). Importantly, attenuating JNK signaling by expressing a dominant negative form of basket (bsk, JNK in flies) resulted in an appreciable restoration in the level of upd3 expression in the PCs of HSD-fed flies, implicating the involvement of JNK signaling in the upregulation of upd3 expression (Fig. 3 K and Fig. S2 I). Analogous results were obtained when the expression of kayak (kay, Fos in flies) was downregulated in the PCs of HSD-fed flies (Fig. 3 K and Fig. S2 I).

Apoptotic signaling kinase 1 (Ask1) is a Jun N-terminal Kinase Kinase that serves as a ROS sensor and gets activated via ROS-mediated dimerization (Sekine et al., 2012). Inactivating Ask1 by knocking down its expression did not lead to any drop in the level of TRE-dsRed expression in the PCs of HSD-fed flies (Fig. 3, L and O). Furthermore, attenuating Ask1 function in the PCs of HSD-fed flies failed to rescue the elevated levels of upd3 transcripts observed in the RNA isolated from the PCS and cardiac tube (Fig. 3 P and Fig. S2 J). Together, these results demonstrate that the upregulation of upd3 expression in the PCs of HSD-fed flies is not mediated by the conventional ROS-Ask1-JNK/p38 signaling cascade. Instead, elevated JNK signaling, triggered in a ROS-independent manner, induces the upd3 expression to potentiate the excessive expression of Prc. Importantly, this kind of regulation appears quite distinct from the previously known roles of ROS and oxidative stress as common denominators associated with cardiac hypertrophy and cardiac fibrosis (Wilson et al., 2018; Zhou et al., 2011).

Our results demonstrated that feeding adult flies on HSD led to insulin resistance, a physiological condition characterized by reduced tissue responsiveness to insulin and low glucose uptake. To analyze the glucose uptake potential of the PCs in HSD-fed flies, we stained them with 2-NBDG, a fluorescent deoxyglucose analog (Zou et al., 2005) that can be taken up by cells through glucose transporters. The PCs of HSD-fed flies demonstrated a threefold reduction in the uptake of 2-NBDG compared with those of flies reared on ND (Fig. 4, A–C) as reported earlier (Na et al., 2015). Next, we used a GLUT1-GFP protein trap reporter to analyze the expression of glucose transporters. As evident from Fig. 4, D–F, compared with the PCs of ND-fed flies, the PCs of HSD-fed flies exhibited a significant drop in GLUT1-GFP expression. Similarly, the PCs of HSD-fed flies exhibited a remarkable drop in the expression of the Hexokinase-GFP (HexA-GFP; Hexokinase is the initial glycolytic enzyme involved in catalyzing the phosphorylation of glucose to glucose-6-phosphate) as compared with that observed in the PCs of ND-fed flies (Fig. 4, G–I). An analogous drop in reporter GFP expression for Phosphoglucoisomerase (Pgi), the glycolytic enzyme responsible for the reversible isomerization of glucose-6-phosphate and fructose-6-phosphate, was also detected in the PCs of HSD-fed flies (Fig. 4, J–L) Together, these results indicate that the process of glycolysis is compromised in the PCs of the HSD-fed flies.

Several studies demonstrated that besides reduced glucose uptake, insulin resistance and hyperglycemic conditions also alter cellular lipid metabolism, leading to the development of dyslipidemia in mammals (Schofield et al., 2016). Given that the basic metabolic pathways and signaling pathways involved in lipid metabolism are evolutionarily and functionally conserved between Drosophila and mammals (Chatterjee and Perrimon, 2021), we were intrigued to investigate the status of lipid metabolism in the PCs of HSD-fed flies. Cellular lipid metabolism in Drosophila is divided into two main processes: one is the formation of triglycerides (TAG) (lipid droplets) called lipogenesis and the other is lipolysis or mobilization of TAG from lipid droplet (Fig. 5 A). Quantitative analyses of the transcript levels of the genes encoding enzymes involved in lipogenesis, which include Glycerol-3-phosphate acyltransferase 4 (Gpat4), Acylglycerol phosphate acyltransferase 4 (Agpat4), and Diacylglycerol acyltransferase (Dgat2) (Fig. 5 A), revealed their enrichment in the RNA isolated from PCs and cardiac tube of HSD-fed flies as compared with that of ND-fed flies (Fig. 5 B). Interestingly, compared with Gpat4 and Agpat4, a robust increase in the levels of Dgat2 transcripts (which codes for the enzyme involved in the conversion of diacylglycerol (DAG) to triacylglycerol was detected (Fig. 5 B). Interestingly, expression of the genes involved in lipolysis, such as Brummer lipase (bmm) and Hormone-sensitive lipase (Hsl) (Fig. 5 A), were also upregulated in HSD-fed condition (Fig. 5 B). In contrast, a significant reduction in the levels of Acetyl-Coenzyme A carboxylase (Acc) and Fatty acid synthase 1 expression (genes involved in de novo fatty acid synthesis; Fig. 5 A) was observed (Fig. 5 B). Together these results indicate high dietary sugar induced significant alterations in cellular lipid metabolism; while the kinetics of both lipogenesis and lipolysis are upregulated, generation of fatty acids by de novo fatty acid synthesis is compromised. Since the rate of de novo fatty acid synthesis is dependent on glycolytic flux (Fig. 5 A), low sugar uptake and a consequent impairment in glycolysis might be responsible for the reduction in de novo fatty acid synthesis.

To substantiate the results of these expression studies, we first stained these cells with Boron-Dipyrromethane (BODIPY), a selective fluorescent stain for intracellular neutral lipid droplets. While low staining levels were observed in the PCs of flies reared on ND (Fig. 5, C–E), a robust increase in neutral lipids was detected in the PCs of HSD-fed flies (Fig. 5, D and E). The lipid enrichment in the PCs of HSD-fed flies was further evident upon LipidTOX (a validated marker for neutral lipids) staining (Fig. S3, A–C). A robust increase in the triacylglycerol level was also evidenced in the tissue that included the cardiac tube and the PCs (Fig. 5 F). Next, we checked for the levels of free fatty acids by staining the tissues with Nile Blue A (Boumelhem et al., 2022). As evident from Fig. 5, G and I, a significant increase in Nile Blue A staining was observed in the PCs of HSD-fed flies as compared with that observed in ND-fed flies. Therefore, these results, in conjunction with the upregulation in the expression of genes associated with lipogenesis and lipolysis, demonstrate that lipid metabolism gets altered in the PCs of the HSD-fed flies.

Cells, when experiencing low glycolytic flux, generally draw upon lipids as an alternate energy source. Lipid metabolism entails the conversion of stored TAG into fatty acids, followed by FAO in the mitochondria. Acetyl-CoA, one of the end products of FAO, feeds into the Krebs cycle to support energy generation by oxidative phosphorylation. To ascertain whether the PCs of HSD-fed flies rely on FAO, we checked for the reporter GFP expression in the PCs of Hnf4-Gal4, UAS-mCD8GFP flies reared on HSD. Compared with the flies reared on ND, elevated levels of reporter GFP expression were observed in the PCs of HSD-fed flies (Fig. 5, J–L). The gene Hepatocyte nuclear factor 4 (Hnf4) codes for a highly conserved nuclear receptor that regulates lipid mobilization and FAO (Palanker et al., 2009). In flies, the gene scully codes for the homolog of mammalian Hydroxy acyl-CoA dehydrogenase, an essential enzyme involved in FAO (Fig. 5 M). Feeding on HSD led to an almost twofold increase in the scully-YFP protein trap reporter expression in the PCs as compared with that observed in the PCs of flies reared on ND (Fig. 5, N and P). Similar enrichment in the levels of reporter YFP expression for acyl-CoA dehydrogenase (CG3902) was also observed in the PCs of HSD-fed flies (Fig. S3, D–F). Furthermore, compared with that observed for ND-fed flies, elevated levels of expression of the critical genes involved in FAO (Fig. 5 M), which include withered (whd), medium-chain acyl-CoA dehydrogenase, mitochondrial trifunctional protein α subunit (Mtpα), mitochondrial trifunctional protein β subunit (Mtpβ), scully, and yippee interacting protein 2, were detected in the RNA isolated from cardiac tissue and the PCs of HSD-fed flies (Fig. 5 Q). Based on these results, we conclude that the PCs of HSD-fed flies exhibit altered lipid metabolism and rely on elevated levels of FAO for their energy demands.

In Drosophila, lipids are transported in hemolymph as lipoprotein particles, the most abundant being lipophorin. About 95% of all hemolymph lipids are carried by the lipophorin particles predominantly in the form of DAG between organs depending on the changing physiological requirements (Palm et al., 2012; Rodríguez-Vázquez et al., 2015). Uptake of neutral lipids from the circulating lypophorins is mediated by the Lipophorin receptors, LpR1 and LpR2 (members of the low-density lipoprotein receptor family) present on cell surfaces (Fig. 5 A) by an endocytosis-independent mechanism yet to be fully characterized (Parra-Peralbo and Culi, 2011). In an attempt to modulate lipid metabolism in the PCs of HSD-fed flies, we independently knocked down the expressions of LpR1 and LpR2. Downregulation of either of these two genes resulted in a detectable drop in the levels of neutral lipids in the PCs of HSD-fed flies as evidenced by staining with BODIPY (Fig. 6, A–E). In tune with these results, a reduction in the levels of free fatty acids was detected in the PCs of HSD-fed flies upon independent downregulation of LpR1 and LpR2 (Fig. 6, F–J). In this context, it is important to note that downregulation of either of these genes failed to completely restore the levels of neutral lipids and free fatty acids to that observed in the PCs of flies reared on ND as these two genes are partially redundant (Parra-Peralbo and Culi, 2011). Next, we wanted to determine whether impairing lipid import into the PCs has any impact on the expression of upd3. Indeed, the increased levels of upd3 transcripts observed in the PCs of HSD-fed flies were partially suppressed upon downregulation of either LpR1 or LpR 2 (Fig. 6 K and Fig. S3 G). From these results, we conclude that impairing lipid uptake in the PCs of HSD-fed flies impedes the formation of lipid droplets and free fatty acids, eventually leading to partial restoration of the increased levels of upd3 expression.

In a converse set of experiments, we overexpressed bmm in the PCs of HSD-fed flies in an attempt to increase the levels of free fatty acids by enhanced lipolysis. As evident from Fig. 6, L–N, overexpression of bmm led to a remarkable reduction in the levels of neutral lipids in the PCs of HSD-fed flies. The amount of neutral lipids present in the PCs of these flies was even lower than that found in the PCs of flies reared on ND (Fig. 6 O). Concurrent with this result, a significant increase in the levels of free fatty acids was observed in the PCs of the HSD-fed flies (Fig. 6, P–R). More importantly, the elevated levels of upd3 expression associated with the PCs of HSD-fed flies got further augmented due to overexpression of bmm (Fig. 6 S and Fig. S3 H).

In sum, the results of these experiments provide a better understanding of the altered metabolic status in the PCs of HSD-fed flies and how it influences upd3 expression in these cells. The development of insulin resistivity due to high dietary sugar impedes sugar uptake by the PCs, resulting in a consequent drop in glycolytic flux. Under this situation, the PCs demonstrate increased uptake of neutral lipids from circulating lipophorins and convert them into TAGS by lipogenesis. Since the lipophorin particles carry the lipids predominantly in the form of DAG, compared with other genes involved in lipogenesis, we see a robust increase in the expression of Dgat2 responsible for converting DAG to TAG. BODIPY and LipidTOX stainings further endorse the presence of a large number of lipid droplets in these cells. Importantly, these PCs also demonstrate an enhanced rate of TAG mobilization into free fatty acids (FFA). Apart from an increase in the transcript levels of the lipolytic genes (bmm, Hsl), high levels of FFA are detected in these cells. Eventually, the increased levels of FFA fuel FAO to generate energy. Most intriguingly, the elevated levels of FFA upregulate upd3 expression in the PCs of the HSD-fed flies. Therefore, these results demonstrate that high dietary sugar induced increased lipid uptake and its mobilization to FFA in the PCs contribute to upregulating upd3 expression.

Next, we investigated whether increased FAO, in any way, is responsible for the upregulated upd3 expression in the PCs of HSD-fed flies. Knocking down the expression of whd, which codes for a rate-limiting enzyme for FAO (Strub et al., 2008), specifically in the PCs of HSD-fed flies, resulted in a robust drop in the levels of upd3 transcripts in the RNA isolated from PCs and cardiac tube (Fig. 7 A and Fig. S4 A). However, knocking down whd in the PCs of flies reared on ND had no impact on the expression of upd3 (Fig. 7 A). Similarly, even though no change in upd3 expression was detected upon CRISPR/Cas9-mediated knockout (KO) of whd in the PCs of ND-fed flies, a substantial reduction in upd3 expression was detected when whd was knocked out in the PCs of HSD-fed flies (Fig. 7 A and Fig. S4 A). Additionally, upd3 expression was also analyzed in the RNA isolated from PCs and cardiac tube of whd1 homozygous null mutant flies. While the expression of upd3 in whd1 mutants reared on ND was comparable with that of wild-type control, loss of whd caused a significant reduction in the levels of upd3 expression in the PCs of HSD-fed flies (Fig. 7 B). To further substantiate these observations, we analyzed upd3 expression in the mutants of two essential enzymes of FAO: Mtpα and Mtpβ (Kishita et al., 2012). The levels of upd3 transcripts in the RNA isolated from PCs and cardiac tube of Mtpα[KO] or Mtpβ[KO] mutants were comparable with that observed in flies reared on ND (Fig. 7 B). In contrast, loss of MTPα or MTPβ resulted in a remarkable drop in upd3 expression when the flies were reared on HSD (Fig. 7 B). Given that upd3 expression is restricted to the PCs (please refer to Fig. 2 A), these results connect increased FAO to upregulated upd3 expression in the PCs of HSD-fed flies.

To ascertain whether the drop in upd3 expression, as observed in the PCs of HSD-fed flies upon attenuating FAO, impacted prc expression in the fat cells, we analyzed the expression of prc in the fat cells of these flies. Knocking down the expression of whd in the PCs resulted in a drastic reduction in the levels of prc expression in the fat cells of HSD-fed flies, whereas prc expression remained unaltered in the flies reared on ND (Fig. 7 C and Fig. S4 B). A similar drop in fat-specific prc expression was detected when whd was knocked out in the PCs of HSD-fed flies (Fig. 7 C and Fig. S4 B). In agreement with these observations, a notable reduction in the amount of Prc accumulation around the heart of HSD-fed flies was observed upon downregulating whd in the PCs (Fig. 7, D–H). Finally, we analyzed the effect of downregulating whd in the PCs on the cardiac function of HSD-fed flies. As evident from Fig. 7 I and Videos 6, 7, 8, and 9, the irregularity in heart beating pattern, as observed in HSD-fed flies, was significantly restored upon downregulating whd in the PCs. Accordingly, an appreciable rescue in the arrhythmia index was detected (Fig. 7 J). The heart period was partially restored when whd was either knocked down or knocked out (Fig. S4 C). However, restorations in diastolic interval (Fig. S4 D) and systolic interval (Fig. S4 E) were more apparent in cardiac tubes when whd was knocked out.

Given that either attenuating JNK signaling or downregulating FAO in the PCs of HSD-fed flies led to an appreciable suppression of the increased upd3 expression, we wanted to determine the relation between these two pathways. To start with, the expression of TRE-dsRed in the PCs upon attenuating FAO was assayed. As evident from Fig. 8, A–E, the expression of TRE-dsRed remained unaltered in the PCs of HSD-fed flies upon downregulating whd. In a converse experiment, we analyzed the expressions of FAO-related genes when JNK signaling was downregulated by driving a dominant negative form of bsk in the PCs of HSD-fed flies. While a remarkable drop in whd transcripts was detected in the RNA isolated from PCs and cardiac tubes (Fig. 8 F), the expression of the rest of the genes coding for other FAO enzymes either exhibited a mild drop or no significant alteration upon loss of bsk from the PCs of HSD-fed flies (Fig. S4 F), indicating that the expression of whd is dependent on JNK signaling.

Next, we wanted to check the relation between JNK signaling and FAO in controlling upd3 expression. Downregulating JNK signaling by expressing a dominant negative form of bsk, specifically in the PCs, resulted in a remarkable drop in upd3 transcripts in the RNA isolated from PCs and cardiac tubes of flies reared on either normal or HSD (Fig. 8 G and Fig. S4 G). On the other hand, overexpression of whd in the PCs of flies reared on ND led to an increase in upd3 transcripts as compared with control (Fig. 8 G and Fig. S4 G). Though not significant, a modest increase in upd3 transcripts was detected in the RNA isolated from PCs and cardiac tube of HSD-fed flies upon upregulating whd expression (Fig. 8 G). Subsequently, we checked for the levels of upd3 transcripts by simultaneously upregulating whd expression and downregulating JNK signaling in the PCs of HSD-fed flies. As shown in Fig. 8 G, the expected drop in the level of upd3 transcripts because of attenuating JNK signaling got rescued due to overexpression of whd. Considering that the expression of upd3 is detected only in the PCs, but not in the cardiac tube, these results establish that JNK signaling, apart from its role as a transcriptional activator for upd3, acts upstream of FAO in regulating upd3 expression in the PCs of HSD-fed flies (Fig. 8 H).

Finally, we ventured to identify the mechanism underlying FAO-mediated upregulation of upd3 expression in the PCs of HSD-fed flies. Besides serving as a substrate for the Krebs cycle, acetyl-CoA generated from FAO also plays an important role in the acetylation of various proteins, including histones, to modulate gene expression (Fig. 9 A) (McCarthy et al., 2018; Tiwari et al., 2020). Transfer of acetyl group from acetyl-CoA to conserved lysine residues on histones is achieved by histone acetyl transferases (HATs). Acetylation of lysine residues, in turn, neutralizes their positive charge, thereby causing histones to drift away from DNA. This released structure facilitates access to the transcriptional machinery leading to enhanced gene expression (Shvedunova and Akhtar, 2022). To ascertain whether increased histone acetylation contributes to upregulation of upd3 expression in the PCs of HSD-fed flies, we first checked for the expression of the genes coding for HATs. Compared with the flies reared on ND, a remarkable enrichment in the transcript levels of the two HAT genes, Gcn5 (Carré et al., 2005) and chameau (chm) (Grienenberger et al., 2002; Miotto et al., 2006), was observed in the RNA isolated from the PCs and cardiac tube of HSD-fed flies (Fig. 9 B). To document the status of histone acetylation in the PCs, we performed immunostainings with a battery of antibodies against acetylated H3K9, acetylated H3K18, and acetylated H3K27. As shown in Fig. 9, C, D″, and G, elevated levels of H3K9 acetylation were observed in the PCs of HSD-fed flies as compared with that observed in ND-fed flies. However, the levels of acetylated H3K18 (Fig. S5, A–C) and acetylated H3K27 (Fig. S5, D–F) remained unaltered in the PCs of HSD-fed flies when compared with that observed in the PCs of ND-fed flies. Increased levels of H3K9ac, but not H3K18ac or H3K27ac, suggest that high dietary sugar does not lead to increased acetylation of all lysine residues of histones. Instead, by site-specific histone acetylation, it generates a unique histone acetylation signature responsible for modulating gene expression in the PCs.

Next, we wanted to investigate whether the elevated levels of H3K9ac observed in the PCs of HSD-fed flies are an outcome of increased FAO. For that purpose, we downregulated the expression of whd in the PCs of HSD-fed flies and checked for the status of H3K9ac. PC-specific loss of whd resulted in significant suppression of the high levels of H3K9ac detected in the PCs of HSD-fed flies (Fig. 9, E–G). A similar drop in the levels of H3K9ac was detected upon attenuating JNK signaling by expressing a dominant negative form of bsk in the PCs of HSD-fed flies (Fig. 9, H–K). Together, these results connect high dietary sugar–induced JNK-FAO axis with high levels of H3K9 acetylation in the PCs.

Finally, we wanted to determine whether the elevated levels of HATs expressed in the PCs of HSD-fed flies contribute to upregulating upd3 expression. Independent downregulation of the HAT genes, Gcn5 and chm, in the PCs of HSD-fed flies resulted in a remarkable restoration in the elevated levels of upd3 expression (Fig. 9 L and Fig. S5 G) associated with the PCs of flies reared on high dietary sugar. In an alternate approach to decipher the link between histone acetylation and elevated levels of upd3 expression, we specifically knocked down the expression of acetyl-CoA synthase/AcCoAS (the Drosophila ortholog of ACSS2) (Mews et al., 2017) in the PCs of HSD-fed flies and checked for upd3 expression. AcCoAS is a crucial enzyme that catalyzes the conversion of free acetate into acetyl-CoA and in doing so, influences gene expression by regulating the acetylation of various proteins including histones (Ling et al., 2022; Mews et al., 2017). Downregulation of AcCoAS in the PCs of HSD-fed flies resulted in a remarkable drop in the elevated levels of upd3 expression (Fig. 9 L and Fig. S5 G) comparable with that observed upon downregulation of whd and HAT genes. Put together, these molecular and genetic analyses confirm the essential role of FAO-dependent acetylation of histones in upd3 upregulation in the PCs of HSD-fed flies.

A schematic of the inter-organ signaling cascade linking increased FAO in the PCs of HSD-fed flies to upregulation of prc expression in the fat cells leading to cardiac dysfunction resulting from excess Prc accumulation in the cardiac ECM, unraveled primarily based on in vivo loss-of-function genetic analysis, is shown in Fig. 10. We conclude that consumption of high dietary sugar leads to altered lipid metabolism and JNK-dependent upregulation of FAO in the PCs. Elevated levels of FAO, in turn, potentiate upd3 overexpression by modulating the acetylation state of H3K9. Eventually, overexpression of pericardial upd3 induces excessive Prc synthesis from the fat cells, leading to progressive cardiac fibrosis. Genetic manipulations of members of this newly identified pathway in flies reared on high dietary sugar prevent upregulation of upd3 in the PCs and help partially rescue the cardiac defects associated with excessive Prc accumulation in the cardiac ECM. Though the reason behind increased JNK activation in the PCs of HSD-fed flies is unclear, the results establish that the ROS-Ask1-JNK/p38 signaling cascade, implicated in upd3 regulation during normal physiological conditions, does not play any role in HSD conditions. Instead, the outcome of this study unravels a two-tier regulatory mechanism by which JNK controls upd3 expression in the PCs of HSD-fed flies. Besides its instructive role as a transcriptional activator of upd3, JNK elicits a permissive role via FAO-dependent modification of the epigenetic landscape to facilitate upd3 expression. Intriguingly, the PCs of HSD-fed flies exhibit a unique histone acetylation signature with increased H3K9ac and unaltered H3K18ac/H3K27ac compared with the flies reared on ND. Of note, several studies in the recent past have documented the dietary influence on site-specific histone acetylation in different organs. For instance, high-fat feeding in mice has been associated with reduced H3K18ac and H3K23ac levels in white adipose tissue and the pancreas (Carrer et al., 2017) while hepatic gene transcription regulation is associated with changes in H3K27ac levels (Siersbæk et al., 2017). Furthermore, in type 2 diabetic mice, progressive glomerulosclerosis is associated with renal histone H3K9 and H3K23 acetylation (Sayyed et al., 2010).

Recent studies involving both fly and mouse models have attributed flux through the hexosamine biosynthetic pathway to diet-induced cardiac dysfunction. In flies, increasing hexosamine flux phenocopies HSD-mediated heart dysfunction and genetic attenuation of the hexosamine biosynthetic pathway, specifically in the adult heart, leads to improved heart function of HSD-fed flies. Though the underlying mechanism is not evident, at least it is clear that the restoration in cardiac function is not associated with any detectable alteration in the elevated levels of Prc accumulation observed in HSD-fed flies (Na et al., 2013). In contrast, it has been demonstrated that elevated glucose concentration induces arrhythmias in murine cardiac myocytes by autonomous O-linked glycosylation-dependent activation of CaMKII (Erickson et al., 2013). Intriguingly, a high dietary sugar–induced hexosamine pathway has also been implicated in the etiology of PC dysfunction in adult flies for its role in epigenetic modification–induced altered gene expression. Activation of the Polycomb repressor complex due to O-GlycNAcylation of the PRC1 subunit Polyhomeotic (Ph) impacts nuclear gene expression and thereby links increased dietary sugar to loss of the Nephrin ortholog Sns in the PCs, leading to reduced nephrocytic function (Na et al., 2015). By demonstrating the role of JNK-FAO-H3K9Ac axis in upregulating upd3 expression in the PCs of HSD-fed flies, the outcome of our study not only provides evidence for an alternate epigenetic mechanism by which high dietary sugar modulates gene expression in the PCs but also braids with the previous observations in providing a better understanding of high dietary sugar–induced diabetic nephropathy and cardiac dysfunction. An increased rate of FAO performs two distinct roles in regulating gene expression in the PCs of HSD-fed flies. On the one hand, it promotes hexosamine pathway-dependent O-GlycNAcylation of Ph by supplying acetyl-CoA, essential for the hexosamine biosynthetic pathway; on the other hand, acetyl-CoA is utilized for histone acetylation to modulate upd3 expression. While the first process exacerbates the nephrocytic function of the PCs, the second one aggravates cardiac dysfunction by augmenting excessive accumulation of Prc around the cardiac tube. Furthermore, our results, in conjunction with the previous findings, elucidate that activation of the JNK-FAO-H3K9Ac axis in the PCs of HSD-fed flies enhances cardiac dysfunction by increased Prc accumulation around the heart, whereas escalated hexosamine flux acts on other specific, yet unknown, components of the heart to induce cardiac dysfunction of HSD-fed flies. This might be why the increase in Prc accumulation around the cardiac tube of HSD-fed flies remained unaltered even after heart-specific attenuation of the hexosamine pathway.

Cytokines are pleiotropic molecules that modulate a broad spectrum of biological processes in a highly context-dependent and cell-type-specific manner. Although cytokines are traditionally studied concerning immune responses, many findings have recognized their role in modulating lipid metabolism in cancer and metabolic disorders. Numerous inflammatory cytokines including IL-17A, IL-6, IL-4, TNF-a, IL-15, IL-1, IL-32, and IL-33 have been implicated in modulating the levels of lipolysis (Fuster et al., 2011), lipogenesis (Tsao et al., 2014; Zúñiga et al., 2010), FAO (Carey et al., 2006; Stienstra et al., 2010), cholesterol accumulation (Isoda et al., 2005; Xu et al., 2017), and triglyceride storage (Kralisch et al., 2005) to either promote or inhibit the development of several metabolic diseases. Likewise, specific cytokines such as IL-6, TNF-a, IL-15, and IL-1 have been postulated to promote or inhibit cancer cachexia and prostate cancer by affecting several aspects of lipid metabolism (Batista et al., 2012; Han et al., 2018; Rohena-Rivera et al., 2017). However, to our knowledge, any evidence supporting FAO-mediated activation of cytokine expression is yet to be reported. By providing in vivo genetic evidence for activation of the cytokine Upd3 by FAO, our results shed light on a new facet that integrates cytokine biology and lipid metabolism. Given that the underlying mechanism involves the highly conserved process of histone acetylation, this regulation can modulate cytokine expression in different physiological contexts.

Excessive lipid accumulation and altered fatty acid metabolism are increasingly recognized as critical pathogenic processes contributing to the severity of diabetic cardiomyopathy. The increased reliance of the diabetic heart on fatty acids negatively impacts cardiac efficiency (Lopaschuk et al., 2021; Mazumder et al., 2004; Randle et al., 1963), impairs insulin metabolic signaling (Atkinson et al., 2003; Samuel and Shulman, 2016), elevates the production of ROS (Tsushima et al., 2018), increases mitochondrial membrane uncoupling, promotes lipotoxicity (Finck et al., 2002; Zhou et al., 2000), and exacerbates diabetic cardiomyopathy. As a result, the attenuation of cardiac FAO represents a potential therapeutic target for treating diabetic cardiomyopathy (Fragasso et al., 2006; Saeedi et al., 2005; Yue et al., 2005). However, even though the development of myocardial fibrosis, characterized by the uncontrolled accumulation of collagen and fibronectin, promotes cardiac stiffness and is a cardinal component of diabetic cardiomyopathy, the contribution of elevated FAO in diabetic cardiac fibrosis, if any, has not been evidenced. On a similar note, although epigenetic histone modifications profoundly influence fibroblast gene transcription in injured tissues (Schuetze et al., 2014; Zhang et al., 2017), the mechanistic underpinnings that link metabolic perturbations and epigenetic modifications to trigger fibrogenic activation are obscure. From these perspectives, our findings provide in vivo evidence for a mechanism that connects high dietary sugar-induced accelerated FAO to histone acetylation–dependent upregulation in cytokine expression that potentiates progressive cardiac fibrosis by increased expression of the collagen-like molecule Prc. Investigating whether the pathomechanism identified here also applies to fibrotic disorders in higher vertebrates would be imperative.

All the fly stocks were routinely maintained at 25°C (if not otherwise mentioned) on a standard cornmeal-yeast-agar food containing 0.15 M sucrose. For experimental purposes, freshly eclosed flies were reared on ND (standard cornmeal-yeast-agar food with 0.15 M sucrose) for 2 days. Thereafter, while the control flies were reared on ND for the rest of their life, the experimental batches of flies were reared on HSD (standard cornmeal-yeast-agar food with a final concentration of 1.0 M sucrose). 100 ml food contained maize powder (5.1 g), agar (0.9 g), yeast (1.8 g), methyl paraben (0.3 g), propionic acid (300 μl), and sucrose (5.13 g = 0.15 M for ND and 34.2 g = 1 M for HSD). 20-day-old adult female flies were used for all the analyses. To induce transgene expression using Gal4 drivers, flies of different genetic backgrounds were maintained at 18°C till eclosion. 2-day-old freshly eclosed flies were then reared at 29°C for 14 days (equivalent to 18 days at 25°C) for analyses. The different fly strains used in this study are provided in Table S1.

All immunostaining experiments were carried out at least three times in accordance with the standard protocol. Adult female abdomens were dissected on ice in 1× phosphate-buffered saline (PBS) and fixed overnight at 4°C in 10% paraformaldehyde. The samples were washed the next day with PBS and all the adnominal fat bodies were removed so that the cardiac tissue and the PCs could be properly exposed to all the chemical treatments. The samples were then fixed in 5% paraformaldehyde for about 2 h at room temperature before being washed with 0.3% PBT (0.3% Triton X-100 in 1X PBS) for 90 min. Tissues were then incubated in primary antibody for 24 h diluted in PAXDG (0.3% PBT with 0.3% sodium deoxycholate, 5% normal goat serum, and 1% bovine serum albumin) at 4°C. Following washes in 0.3% PBT, the samples were incubated for 24 h in secondary antibody in PAXDG at 4°C. The tissues were washed in PBT and stained with phalloidin (1:500 in PBS) at 4°C for 48 h followed by staining with 4′,6-diamidino-2-phenylindole at room temperature for 30 min.

For immunostaining with antibodies against acetylated histones, a slight modification was done in the above-mentioned protocol. The adult tissues were dissected in 1X PBS supplemented with deacetylase inhibitors: 10 mM sodium butyrate (EMD 19-137; Millipore) and 10 mM nicotinamide (#72345; Sigma-Aldrich). The rest of the protocol remained the same as already mentioned.

The samples were mounted in vectashield and immunofluorescence images were captured in a Leica SP8 confocal microscope under identical conditions (laser power, gain, and scan settings). The images were processed in parallel and analyzed under the same conditions using Fiji software. The following primary antibodies were used: mouse anti-Prc (1:10; Developmental Studies Hybridoma Bank EC11), mouse anti-Prospero (1:20; Developmental Studies Hybridoma Bank MR1A), rabbit anti-phospho p38 (1:100; Thr¹⁸⁰/Tyr¹⁸², clone 3D7; Cell Signaling Technology), mouse anti-β galactosidase (1:100; no. Z3781; Promega), rabbit anti-H3K9ac (Cat#9927; Cell Signaling Technology), rabbit anti-H3K18ac (Cat#9927; Cell Signaling Technology), and rabbit anti-H3K27ac (Cat#9927; Cell Signaling Technology).

ROS detection with DHE dye (Molecular Probes no. D11347; Invitrogen) was carried out as previously described in (Owusu-Ansah and Banerjee, 2009). Adult female abdomens were dissected in 1X PBS at room temperature followed by incubation in 30 µM DHE solution in 1X PBS for 15 min at room temperature under dark conditions. After DHE incubation, tissues were washed with 1X PBS for 5 min before being fixed with 10% paraformaldehyde for 10 min and again washed with 1X PBS for 5 min at room temperature in dark. Finally, the samples were immediately mounted and imaged using a Leica SP8 confocal microscope (Leica Application Suite X acquisition software) under identical conditions (laser power, gain, and frame size). The images were analyzed and processed using Fiji software under the same conditions.

Adult female abdomen were dissected on ice in 1X PBS and fixed with 10% paraformaldehyde in 1X PBS overnight at 4°C. The following day, the tissues were washed with 1X PBS and the abdominal fat body cells were removed, especially around the PCs, for proper penetration of the stain. Samples were again fixed with 10% paraformaldehyde for 1 h followed by washes in 0.3% PBT (0.3% Triton-X in 1X PBS) for 2 h (8 washes of 15 min each) at room temperature. Then, the tissues were washed twice with 1X PBS before incubating in Bodipy (1:500 in 1X PBS) or LipidTOX solution (1:500 in 1X PBS) for 30 min, followed by washing with 1X PBS. The processed samples were immediately mounted in vectashield, imaged using Laser Scanning Confocal Microscope (Zeiss LSM 900; ZEN 3.1 acquisition software), and processed using Fiji software.

Adult female abdomens were dissected on ice in 1X PBS and fixed with 10% paraformaldehyde in 1X PBS overnight at 4°C. The following day, the tissues were washed with 1X PBS and the abdominal fat body cells were removed, especially around the PCs, for proper penetration of the stain. Samples were again fixed with 10% paraformaldehyde for 1 h followed by washes in 1X PBS for 45 min (three washes of 15 min each) at room temperature. Then, the tissues were incubated in Nile Blue (5 µM in 1X PBS) (Cat. No. #N0766; Sigma-Aldrich) solution for 15 min, followed by three washes with 1X PBS. The processed samples were immediately mounted in vectashield, imaged using Laser Scanning Confocal Microscope (Zeiss LSM 900; ZEN 3.1 acquisition software), and processed using Fiji software.

Adult female abdomens were dissected in Schneider’s medium at room temperature and immediately incubated in 0.25 mM 2-NBDG solution (Cat. No. # N13195; Invitrogen) for 45 min at room temperature under dark conditions as prescribed in de la Cova et al. (2014). After incubation, followed by a quick PBS wash, the tissues were fixed in precooled 10% paraformaldehyde prepared in 1X PBS on ice for 1.5 h. Then the tissues were washed with chilled 1X PBS for 5 min, immediately mounted in vectashield, imaged using Laser Scanning Confocal Microscope (Zeiss LSM 780; ZEN, 2010 B SP1 acquisition software), and processed using Fiji software.

To study the expression levels of genes in the PCs, the PCs along with the adult cardiac tubes were carefully dissected out from around 40 to 50 female flies in 1X PBS. Care was taken to have the cardiac tube and the associated PCs only. 1X PBS was replaced with TRIzol (Ambion no. 15596018) and RNA was isolated according to the manufacturer’s instructions.

To study the expression of prc in the fat cells, abdominal cuticles (with fats, cuticular muscles, and cuticle) from 25 to 30 adult flies were dissected in Schneider’s insect medium on ice. Only the fat bodies were carefully dissected from these abdominal tissues and collected in a 1.5-ml microcentrifuge tube with Schneider’s insect medium on ice. The samples were then centrifuged for 5 min at 5,000 RPM (4°C) to settle down the fat cells at the bottom of the tube. Then the samples were homogenized in Schneider’s insect medium and the lysate was passed through a 22G syringe (BD No. 309631) two to three times. The collected lysate was centrifuged for 5 min at 5,000 RPM (4°C) to pellet down the cells. The Schneider’s insect medium was replaced with TRIzol (Ambion no. 15596018) and RNA was isolated according to the manufacturer’s instructions.

cDNA was generated using the Verso cDNA synthesis kit (Molecular Probes no. AB1453A) according to the manufacturer’s instructions. The Bio-Rad CFX96 Touch Real-Time PCR Detection System was used to perform real-time quantitative polymerase chain reaction (RT-qPCR) using the TB Green Premix Ex Taq (Tli RNaseH Plus) (no. RR440A; Takara). Graded RT-qPCR was used to determine the optimal primer melting temperature for each primer set. The 2-δCt method was used to determine the relative levels of mRNAs. The transcript levels were normalized to those of the constitutive ribosomal gene coding for ribosomal protein 49 (rp49). Three to four independent cDNA samples were used in each RT-qPCR experiment. The figures depict the total data from all samples. Primers were created in such a way that they span the exon-exon junction. Table S2 provides the list of primers used in our RT-qPCR experiments.

For insulin sensitivity assay, adult female flies were dissected open in Schneider’s medium supplemented with 10% FBS to completely expose the body cavity to the following chemical treatments. 5 μM recombinant human insulin (Cat # I1376497001; Roche) was added to the Schneider’s medium and samples were incubated at room temperature for 20 min. For the control samples, dilution buffer (10 mM HEPES) was added to the medium. The rest of the treatments were the same for both the sample groups.

Once the incubation was done, the heads were removed and the bodies were homogenized on ice using an autoclave pestle in 1% Triton X-100 in 1X PBS buffer (pH 7.2) supplemented with 1% mammalian Protease Inhibitor Cocktail (P8340; Sigma-Aldrich) along with phosphatase inhibitor. The homogenate was centrifuged at 20,000 g for 15 min at 4°C and the supernatant was mixed with an equal volume of 2× Laemmli Buffer (100 mM tris-HCl [pH 6.8]; 4% SDS: Sigma-Aldrich, catalog no. L3771; 0.2% Bromophenol Blue, HIMEDIA, catalog no. MB123; 20% glycerol and 200 mM β-mercaptoethanol, catalog no. M3148; Sigma-Aldrich). Finally, the protein solution was boiled at 95°C for 5 min before it was ready to use for further analysis. The Bradford Assay (catalog no. 5000006; Bio-Rad) was used to quantify the protein concentration. 1 μl of protein sample was mixed with 1 ml of five-times-diluted Bradford reagent, incubated at room temperature for 5 min, and the OD of the solution was measured at 595 nm in a spectrophotometer. The concentration of protein was calculated using a freshly prepared standard curve. The primary antibodies used for western blot analysis are as follows: mouse anti-Tubulin (1:1,000; T6199; Sigma-Aldrich) and rabbit anti-active pAkt (1:1,000; Cat # 4054; Cell signaling). For secondary antibodies, horseradish peroxidase (HRP)–conjugated anti-mouse (no. A00160; GenScript) and HRP-conjugated anti-rabbit (no. A00098; GenScript) were used at 1:5,000 dilution. The western blots were developed using Luminata Crescendo Western HRP substrate (Millipore). The band intensities were analyzed using Fiji software and normalized with tubulin bands that served as loading controls. Western blot analyses were conducted on three to four independent protein samples per condition. The graphs represent the combined data from all samples.

For all the cardiac functional analyses, semi-intact female Drosophila were used and the cardiac contractility was measured as previously described in Fink et al. (2009). High-speed digital movies of beating hearts were captured under bright-field using QImaging optiMOS scMOS camera R696-M-16-C with a 20× water immersion lens (Carl Zeiss). The neighboring PCs and fat tissue were removed during sample preparation to clearly access the cardiac tube for video recording. The movies were captured at 100 fps for a duration of 10 s. M-modes and quantitative data were generated using SOHA, a MATLAB-based image analysis software (Fink et al., 2009). M-mode of a particular sample is the snapshot of the movement of heart edges over time. Heart periods were defined as the time between the ends of two consecutive diastolic intervals, and arrhythmia index was quantified as the standard deviation (SD) of all heart periods in each record normalized to the median heart period for each fly.

The life-span analysis experiments were carried out with three to four replicates of 60 flies (40 females and 20 males) per genotype. Every day, the flies were transferred to new food bottles. When the flies were flipped, dead female flies in each bottle were counted and removed. Female survival curves for each genotype (n = 120–160 flies, three to four independent replicates) were generated in GraphPad Prism version 6.00 using a log-rank Mantel–Cox test.

For intensity analyses of DHE, phospho-p38, HexA-GFP, pgi-GFP, Glut1-GFP, scully-YFP, CG3902-YFP, 2-NBDG, LipidTOX, Bodipy, Hnf4-Gal4UASmcd8GFP, TRE-dsRed, and Nile Blue A, the PCs of the control and experimental samples were marked, and the mean fluorescence intensity (mean gray pixel intensity of the desired area of maximum intensity projections of the Z stacks) was measured using Fiji software. In case of immunostainings with antibodies against acetylated histones, the nuclei of the PCs were marked and mean fluorescence intensity was measured using Fiji software. A total of 50–70 PCs from 12 to 15 individual flies were analyzed for each genotype/dietary alteration. In particular, the PCs around the second chamber of the cardiac tube were analyzed for the sake of uniformity. To determine the intensity of Prc accumulation around the second chamber of the cardiac tube, the entire second chamber was marked and the mean fluorescence intensity (mean gray pixel intensity of the desired area of maximum intensity projections of the Z stacks) was measured using Fiji software. Data were obtained from 20 individual flies for analyzing the intensity of Prc accumulation around the second chamber of the cardiac tube. For all the genotypes, the samples were processed in parallel and all the analyses were done under identical conditions. From these readings, the mean gray values (for the same size of region of interest) of the background were deducted to obtain the mean gray value minus blank. The final intensity values were plotted using GraphPad Prism version 6.00, where the dots represent the number of PCs/cardiac chambers analyzed and the error bars represent SD.

For all the statistical analyses, GraphPad Prism 6 software was used. The analyses of the data were performed by using either unpaired Student’s t test (two-tailed) with Welch’s correction or one-way/two-way analysis of variance (ANOVA) followed by Tukey’s multiple comparison test. For the life-span data, Kaplan–Meier survival curves were generated with GraphPad Prism 6 software and curve comparison was performed using Log-Rank (Mantel–Cox) test. All of the survival graphs shown are the result of three to four independent repeats of 40 flies per genotype. Statistical significance was accepted with P values of <0.05; <0.01; and <0.001, mentioned as *, **, and ***, respectively.

Fig. S1 shows the hallmark features of T2D exhibited by the flies reared on HSD including an increase in prc expression. Fig S2 shows the changes in the levels of upd3 and prc expressions in the control lines when reared on HSD, changes in heart period, diastolic and systolic intervals upon knocking down upd3 in the PCs of HSD-fed flies, and drop in elevated levels of physiological ROS in the PCs of HSD-fed flies. Fig. S3 shows altered lipid metabolism in the PCs of HSD-fed flies and changes in the levels of upd3 expressions in the control lines when reared on HSD. Fig. S4 shows the changes in the levels of upd3 and prc expressions in the control lines when reared on HSD, changes in heart period, diastolic and systolic intervals upon either knocking down or knocking out of whd in the PCs of HSD-fed flies, and expression levels of genes involved in FAO upon attenuating JNK signaling in the PCs of HSD fed flies. Fig. S5 shows the levels of H3K18ac and H3K27ac in the PCs of HSD-fed flies and the expression levels of upd3 in the control lines when reared on HSD. Table S1 shows the details (genotypes and sources) of the Drosophila stocks used in this study. Table S2 has the list of primers and their sequences used for this study. Videos 1 and 2 show the beating second chamber of the cardiac tube (heart) of adult Drosophila (Ore^R) reared on ND and HSD, respectively. These movies are related to Fig. 1 H. Videos 3 and 4 show the beating second chamber of the heart of adult flies (dot-Gal4; tub-Gal80^ts) reared on ND and HSD, respectively. Video 5 shows the beating second chamber of the heart of adult flies (dot-Gal4; tub-Gal80^ts/UAS-upd3dsRNA) reared on HSD. Videos 3, 4, and 5 are related to Fig. 2 J. Videos 6 and 7 show the beating second chamber of the heart of adult flies (dot-Gal4; tub-Gal80^ts) reared on ND and HSD, respectively. While Video 8 shows the beating second chamber of the heart of adult flies (dot-Gal4; UAS-whddsRNA/tub-Gal80^ts) reared on HSD, Video 9 represents the beating second chamber of the heart of adult flies (dot-Gal4/whd^TKO.GS00854; UAS-Cas9/tub-Gal80^ts) reared on HSD. Videos 6, 7, 8, and 9 are related to Fig. 7 I.

All data underlying the research present in this manuscript are available from the corresponding author upon request ([email protected]).

We thank E. Bach (New York University Grossman School of Medicine, New York, NY, USA), J.A. Hoffmann (University of Strasbourg, Strasbourg, France), S. Newfeld (Arizona State University, Tempe, AZ, USA), U. Banerjee (University of California, Los Angeles, Los Angeles, CA, USA), Irene Miguel-Aliaga (The Francis Crick Institute, London, UK), and D. Bohmann (University of Rochester Medical Center, Rochester, NY, USA) for reagents; Bloomington Drosophila Stock Center and Kyoto Stock Center Drosophila Genetic Resource Center for fly stocks; Developmental Studies Hybridoma Bank for antibodies; other members of the laboratory for critical input; and Indian Institute of Science Education and Research Mohali (IISER Mohali) confocal (Leica SP8; Zeiss LSM900 and Zeiss LSM780 systems) microscopy facility for imaging. The models were created using BIORENDER.

This study was funded by IISER Mohali (J. Gera, G. Chauhan, A. Choudhary, L. Rani), Indian Council of Medical Research, India (3/1/3/JRF-2020/HRD(LS)/132109/46) (D. Kumar), Science and Engineering Research Board, Department of Science and Technology, India (CRG/2020/000511) (S. Mandal), and Wellcome Trust DBT India Alliance, India (IA/S/17/1/503100) (L. Mandal). Open Access funding provided by IISER Mohali.

Author contributions: J. Gera: Formal analysis, Investigation, Visualization, Writing—review and editing, D. Kumar: Investigation, Visualization, G. Chauhan: Investigation, Visualization, A. Choudhary: Investigation, Visualization, L. Rani: Investigation, L. Mandal: Funding acquisition, Visualization, Writing—original draft, Writing—review and editing, S. Mandal: Conceptualization, Formal analysis, Funding acquisition, Methodology, Project administration, Supervision, Visualization, Writing—original draft, Writing—review and editing.