Discovering signaling mechanisms governing metabolism and metabolic diseases with Drosophila

Summary

There has been rapid growth in the use of Drosophila and other invertebrate systems to dissect mechanisms governing metabolism. New assays and approaches to physiology have aligned with superlative genetic tools in fruit flies to provide a powerful platform for posing new questions, or dissecting classical problems in metabolism and disease genetics. In multiple examples, these discoveries exploit experimental advantages as yet unavailable in mammalian systems. Here we illustrate how fly studies have addressed long-standing questions in three broad areas – inter-organ signalling through hormonal or neural mechanisms governing metabolism, intestinal interoception and feeding, and the cellular and signalling basis of sexually dimorphic metabolism and physiology – and how these findings relate to human (patho)physiology. The imaginative application of integrative physiology and related approaches in flies to questions in metabolism is expanding, and will be an engine of discovery, revealing paradigmatic features of metabolism underlying human diseases, and physiological equipoise in health.

“Progress, far from consisting in change, depends on retentiveness.

[W]hen experience is not retained,…infancy is perpetual.” – Santayana

Introduction

Studies with Drosophila melanogaster have been a wellspring of biological discoveries for over a century, including pioneering, widely heralded studies revealing general principles of genetics, development, immunity, circadian physiology, neurobiology and behavior (Bilder and Irvine, 2017). Foundational studies in fruit flies have repeatedly presaged important findings in humans and other vertebrates, making Drosophila an indispensable organism for biology. Common elements of animal metabolism include the primacy of glucose for energy generation, the need for mobilizing energy stores like glycogen and lipids in periods of caloric restriction or reproduction, and the mandate to adapt feeding behavior to match nutrient needs. In both Drosophila larvae and adults, a high-sugar diet causes hyperglycemia, hyperinsulinemia and insulin resistance, with adults also demonstrating obesity (Mattila and Hietakangas, 2017; Morris et al., 2012; Musselman et al., 2011; Pasco and Léopold, 2012; van Dam et al., 2020). Similarly, flies fed a high-fat diet also demonstrate insulin resistance, elevated triglycerides and cardiac dysfunction (Birse et al., 2010). Akin to their effects in humans, adipogenic diets can also promote tumor formation (Hirabayashi, 2016; Hirabayashi et al., 2013; Newton et al., 2020). Based on these and other unifying features, there is growing evidence that integrative studies in flies can reveal important principles of animal metabolism, including the genetic and signaling mechanisms that maintain health and underlie metabolic diseases like obesity and diabetes mellitus. Here, we illustrate findings from recent research in adult flies to expand awareness of this view. An abbreviated selection of studies of larval metabolism and growth is also included, but influential reports or reviews focused on this distinct developmental stage are found elsewhere (Böhni et al., 1999; Brankatschk et al., 2014; Gillette et al., 2021; Grenier and Leulier, 2020; Pasco and Leopold, 2012; Rajan and Perrimon, 2012). Likewise, prior more focused reports and reviews provide excellent summaries of research on fly lipid metabolism (Heier and Kühnlein, 2018; Kühnlein, 2011; Musselman and Kühnlein, 2018; Palm et al., 2012; Storelli et al., 2019), Warburg effect and tumour growth (Drummond-Barbosa and Tennessen, 2020; Tennessen and Thummel, 2011), hormone biology (Ahmad et al., 2020), circadian regulation of metabolism (King and Sehgal, 2020; Patke et al., 2020) and diabetes modeling (Alfa and Kim, 2016).

The success of metazoans in navigating responses to physiologic and pathologic challenges to metabolism is determined by a combination of intrinsic cellular responses, and adaptations in multiple tissues coordinated by intercellular signaling. The experimental toolkit for Drosophila is arguably most powerful when applied to the study of inter-organ communication (Droujinine and Perrimon, 2016, 2019). These communication axes are critical for the regulation of hormone or neuropeptide secretion and their signaling. Beyond endocrine signaling of energy status (reviewed below), lipoproteins also communicate information about dietary lipid intake to the brain to regulate insulin secretion (Brankatschk et al., 2014; Palm and Rodenfels, 2020). The transcriptome of each organ is now available, including organ-specific single-cell transcriptomes for the adult brain and intestine (Allen et al., 2020; Croset et al., 2018; Davie et al., 2018; Guo et al., 2019; Hung et al., 2020; Leader et al., 2018; Robinson et al., 2013). These descriptive datasets allow identification of candidate systemic signals and their remote targets. Predictions can then be probed functionally thanks to an increasing repertoire of binary systems (Kockel et al., 2019; Lin and Potter, 2016) for gene inhibition, overexpression and mutation, and an expanding nanobody-based genetic toolkit for protein degradation or re-localization (Aguilar et al., 2019). These tools include the use of the Gal4/UAS system of binary transcriptional activation, and the CRISPR/Cas9 system of genome editing (Bassett and Liu, 2014; Caygill and Brand, 2016; McGuire et al., 2004b; Xu et al., 2019). These systems combine the use of tissue-specific promoters to drive the expression of a transcriptional activator (e.g. Gal4), which binds the upstream activating sequence (UAS) to express virtually any DNA sequence of the experimenter’s choice. Today, multiple binary systems (e.g. Gal4/UAS, LexA/LexAop, Q systems) can be combined in a single fly to conduct genetic perturbations of multiple tissues simultaneously, and measure the effects of those perturbations (Figure 1). Together, these tools allow exquisite spatial and temporal control of gene expression and protein function, including the ability to genetically target different tissues or organs independently, manipulate the function of a gene in a given tissue or their activity, and determine the interactions amongst different tissues and organs (Kockel et al., 2019; Lin and Potter, 2016; Wendler et al., 2020) (and references therein). Use of genetic and physiological tools that can unequivocally establish the directionality and significance of signaling across organs is highlighted below. This includes signaling between brain neurons, neuroendocrine and gastrointestinal cells, including enteroendocrine cells, gonads and the fat body, an organ combining features of mammalian liver and adipose cells (Figure 2A–B).

Figure 1. Binary expression systems to study inter-organ signaling in Drosophila.

The expression of specific genes, reporter genes, guide/siRNAs or cell activators/silencers can be confined temporally and/or to specific cell types using publicly available binary expression systems. Gal4-driven expression of UAS-fused transgenes (Brand and Perrimon, 1993) can be further refined with a third transgene allowing for tissue-specific and/or temporally controlled expression of a Gal80 protein, which prevents Gal4 function (McGuire et al., 2003; McGuire et al., 2004a). The LexA-LexAaop (Lai and Lee, 2006) and QF-QUAS (Potter et al., 2010) systems provide alternative binary systems which, like the Gal4-UAS system, can be further refined with Gal80 or QS/quinic acid, respectively. The cell type-specificity of “driver” (Gal4/LexA/QF) lines can be further increased through intersectional “split” approaches, which use two enhancers with activity in overlapping cell groups to confine Gal4/LexA/QF expression to the intersection of the two groups (Luan et al., 2006; Riabinina et al., 2019; Ting et al., 2011). Finally, the Flp-FRT system (Golic and Lindquist, 1989; Lee and Luo, 2001) can be used to further restrict these binary expression systems to lineage-related subsets of cells, resulting in mosaic expression within a tissue.

Figure 2. Overview of anatomy and cell interactions in adult flies.

(A) Schematic of an adult female Drosophila highlighting organs involved in energy homeostasis and metabolism. Insulin-producing cells and other cells discussed here (not shown) are located in the pars intercerebralis (PI) of the brain, a distinct structure from the abdominal ganglion or ventral nerve cord (VNC). Cells in the corpora cardiaca (CC) secrete AKH, a glucagon-like hormone. Processes emanating from both IPCs and CC cells have direct contact with the heart tube. The corpora allata (CA) produces juvenile hormone to sustain intestinal stem cell proliferation and remodel enterocytes in mated female flies. The fat body, an insulin target tissue analogous to mammalian liver and adipose tissue, lines the body cavity of the abdomen and thorax. Neurons in the PI and hypocerebral ganglion (HCG) innervate the fly crop and gut. Food is stored in the crop where digestion begins, then transits through the foregut, midgut and hindgut and rectal ampulla. The midgut, analogous to the small intestine, is involved in nutrient sensing and interoception. The Drosophila ovary is posterior and adjacent to the gastrointestinal tract. (B) Neurons including insulin-producing cells (IPCs), DH44, Ms and CN neurons are depicted and discussed in the text. Plus (+) and minus (−) signs indicate activating or inhibitory interactions between CN neurons and IPCs or CC cells, which produce AKH and are adjacent to the HCG. Projections of IPC and Ms neurons to the crop and other GI organs, or the IPCs and CC cells to the heart are shown.

I. Endocrine regulation of metabolism: focus on Drosophila insulin, glucagon and leptins

To develop, grow, and generate their progeny, all organisms acquire nutrients in order to survive periodic or prolonged nutrient scarcity. This challenge of feeding and fasting embodies one of the ne plus ultra selective forces in evolution. The conservation of insulin signaling across metazoa, from insects to mammals (Srivastava et al., 2010), indicates the selective advantage of endocrine systems in the coordination of metabolic responses to feeding and fasting states. Here we review recent progress in understanding the roles and regulation of fly insulin, glucagon and other hormones in governing metabolism.

Conserved endocrine mechanisms govern metabolism in Drosophila and humans

Drosophila research over the past two decades has demonstrated that glucose, amino acid and lipid metabolism is regulated by fly orthologues of insulin, glucagon, leptin, and other hormones (Table 1). In Drosophila and other insects, insulin-producing cells (IPCs) are specialized neurons that synthesize and secrete insulin to maintain glucose and lipid homeostasis (Gronke et al., 2010; Haselton and Fridell, 2010; Ikeya et al., 2002; Rulifson et al., 2002), through activation of insulin receptor (InR) and InR substrates (IRS1/2) in targets like brain, muscle, and the fat body (Figure 2A). Corpora cardiaca (CC) cells produce and secrete adipokinetic hormone (AKH), the insect orthologue of glucagon (Alfa and Kim, 2016; Isabel et al., 2005; Kim and Rulifson, 2004; Lee and Park, 2004). Elements of IPC and CC developmental genetics and specification resemble those of pancreatic islet β and α cells (Clements et al., 2008; Kim and Rulifson, 2004; Miguel-Aliaga et al., 2008; Park et al., 2011), like transcription factors governing IPC and β cell development and functional maturation (Barry and Thummel, 2016). Moreover, direct contact between fly IPCs and CC cells with the fly heart and other cells described below appear to be homologous to those between β cells, α cells and vessels in islets (Figure 2B). Fly leptin-like adipokines and other hormones have also been shown to regulate metabolism (Ahmad et al., 2020; Alfa et al., 2015; Beshel et al., 2017; Hentze et al., 2015; Mattila and Hietakangas, 2017; Rajan and Perrimon, 2012). Below, we focus on recent findings involving fly insulin-like peptides, the glucagon orthologue, AKH, and leptins.

Table 1: Advantages of Drosophila for studies of metabolism.

See text for details.

Advantages of Drosophila for studies of metabolism
Conserved physiology
• Systemic Insulin from Insulin-Producing Cells (IPCs) regulates metabolism
• Systemic glucagon-like hormone (AKH from CC cells) regulates metabolism
• Functional orthologues of leptin (Upd1, Upd2) identified in Drosophila
• Insulin, AKH and other hormone secretion is responsive to fasting and re-feeding
• Fly IPCs and human β cells have similar transcriptomes and stimulus-secretion coupling and nutrient regulation
• In multiple cases (14/14), loss of gene function in IPCs and islet β cells had a similar effect on insulin output: Insulin, InR, IRS1/2, AKT1, GLUT1, GLIS3, ZNT8, ABCC8, DGKB, SUR1, ADRA2, BCL11A, SIX2, PRC1
• Fly AKH-secreting CC cells and human a cells have similar stimulus-secretion coupling and nutrient regulation
Conserved pathophysiology
• Can challenge flies by fasting, re-feeding, diet and other metabolic stress
• Insulin insufficiency leads to hyperglycemia
• Insulin excess leads to excess adiposity, growth, hypoglycemia
• Glucagon-like hormone (AKH) deficiency leads to hypoglycemia
• Insulin resistance develops from dietary challenge or mutation, and stimulates adaptive hyperinsulinemia
• Striking concordance of insulin output phenotypes after loss- or gain-of-function studies in IPCs and pancreatic islet β cells
Experimental advantages
• Unbiased screens to identify novel mechanisms governing metabolism
• Genetic toolkit permits targeted loss- or gain-of-function studies in specific cells, including simultaneous targeting in two or more distinct cell types
• Temporal control of gene expression permitting uncoupling of developmental from homeostatic/adult effects
• Quantitative assays to measure glucose, adiposity, weight, insulin and AKH levels

Secretion of insulin by pancreatic β cells and glucagon by α cells is governed by nutrient sensing coupled to well-delineated electrophysiological signaling mechanisms (Macdonald, 2016; Rorsman and Braun, 2013). Genetic studies, measures of insulin or AKH secretion, and electrophysiology show that principal features of α cell and β cell ‘stimulus-secretion coupling’ are remarkably conserved in adult fly IPCs and CC cells. For example, AKH secretion by CC cells appears to be suppressed by feeding or glucose, reminiscent of glucagon regulation in islet α cells (Alfa et al., 2015; Kim and Rulifson, 2004; Oh et al., 2019). In IPCs, secretion of Insulin-like peptides is regulated by glucose and lipids (Ahmad et al., 2020; Alfa and Kim, 2016). Also like in islet β cells, IPC secretion is governed by mitochondrial metabolism (Barry and Thummel, 2016; Fridell et al., 2009; Storelli et al., 2019), inward rectifying potassium channel-dependent mechanisms and depolarization, resulting in calcium transients (Kreneisz et al., 2010), and release of a minor fraction of pre-synthesized, processed and stored insulin (Park et al., 2014). After fasting and re-feeding, the degree and tempo of Insulin-like peptide 2 (Ilp2) increase and clearance in adult fly hemolymph (Figure 3A) is strikingly similar to serum insulin excursions observed in mice or humans after glucose challenge (Figure 3B).

Figure 3. Using Drosophila to discover novel regulators of human islet function.

(A,B) Glucose-stimulated insulin secretion and clearance in Drosophila (A) and mouse (B), measured by insulin ELISA. (C) Experimental strategy connecting human studies like Genome-wide association studies (GWAS) of disease risk to in vivo testing in Drosophila, thereby prioritizing secondary and tertiary studies in mammalian systems (mice shown here) and human cells or tissues. Data in panel A adapted from Park et al 2014.

The multiple homologies of IPCs and islet β cells, coupled with the ability to measure total and circulating insulins (like Ilp2) using ELISA assays have motivated genetic studies to identify intrinsic regulators of insulin production and secretion (Barry and Thummel, 2016; Park et al., 2014; Peiris et al., 2018). For example, using RNAi-based suppression of genes encoding orthologues known to regulate islet β cell insulin production or secretion, it was shown – in 14/14 cases – that targeted loss-of-function studies in fly IPCs led to changes in insulin output similar to those observed after homologous loss-of-function studies in pancreatic islets. This included genes encoding Ilp2, and orthologues of Insulin receptor, Insulin receptor substrates 1/2 (IRS1/2), GLUT1, GLIS3, ZNT8, ABCC8, DGKB and ADRA2. Changes of insulin production and output were distinct or not detected after shRNA-mediated gene suppression in the fly fat body, demonstrating specific requirements for these factors in fly insulin-producing cells (Park et al., 2014; Peiris et al., 2018)

Discovering regulators of pancreatic islet function with flies

Genetic and physiological homologies between fly IPCs and islet β cells predicted that discovery of IPC regulators could unveil conserved mechanisms governing insulin secretion. Multiple recent studies have supported this heuristic (Bevacqua et al., 2020; Peiris et al., 2018). Peiris et al (2018) investigated the in vivo function of fly genes orthologous to imputed human diabetes risk genes without known roles in β cells (Mahajan et al., 2018). Measures of insulin output after RNAi targeting in IPCs led to identification of three novel IPC regulators, CG9650, fascetto, and optix, the respective orthologues of mammalian genes BCL11A, PRC1 and SIX2. In fly IPCs, RNAi-mediated suppression of CG9650 (BCL11A) or fascetto (PRC1) led to increased circulating levels of insulin; remarkably, loss of BCL11A in primary human β cells or mouse β cells also led to increased insulin output (Park et al., 2014; Peiris et al., 2018), while induction of CG9650 in IPCs or BCL11A in β cells led to reduced insulin output. Targeted suppression of optix in IPCs led to reduced insulin secretion, and recent studies show that SIX2 loss in human β cells also leads to reduced glucose-dependent insulin output (Bevacqua et al., 2020). In these examples, fly studies correctly predicted the direction of islet β cell phenotypes arising from genetic loss-of-function. By contrast, knockdown of CG9650, fascetto or optix in the adult fat body did not detectably alter circulating Ilp2HF levels (Peiris et al., 2018). Thus, integrated genetic, molecular and physiological approaches using fruit flies, mice and human tissues provide a powerful new strategy for discovering tissue-specific functions of imputed diabetes risk regulators (Figure 3C; Table 1).

Discovering systemic regulators of insulin and AKH output

Communication between different tissues and cells reflecting fluctuating nutrient availability and energy status enables whole-organism metabolic homeostasis. For example, human β cells are regulated by circulating adipokines and hepatokines (Cantley, 2014; Wente et al., 2006), as well as by intra-islet paracrine signals like somatostatin, glucagon and the incretin glucagon-like peptide 1 (GLP-1). Fly IPCs and CC cells also receive and integrate long- and short-range signals to regulate Insulin or AKH expression and secretion (Ahmad et al., 2020), and exciting work has unveiled new fat body- and brain-based mechanisms for controlling the output of these hormones in adult flies.

Recent studies have identified Drosophila neurons that coordinate CC cell and IPC output (Oh et al., 2019). Genetic screens in the brain identified a pair of glucose-sensing neurons, termed CN, that project bifurcated axons – one toward CC cells and the other toward IPCs (Figure 2B). CN neuronal excitation by feeding and systemic glucose flux led to simultaneous inhibition of CC cells and AKH secretion, and stimulation of IPCs and insulin secretion. Fasted flies had a reduction in CN neuronal activity, accompanied by a reduction in insulin secretion and an increase in AKH secretion. Electrical silencing of CN neurons resulted in elevated glucose levels in circulating hemolymph (analogous to hyperglycemia). Thus, in addition to their intrinsic mechanisms of glucose sensing (Kim and Rulifson, 2004; Park et al., 2014), these studies reveal that IPC and CC cell activity are coordinated by glucose-sensing neurons, whose functions are required to maintain systemic glucose homeostasis.

The fat body also releases factors that regulate Insulin-like peptide expression, secretion, and signaling (Colombani et al., 2005; Geminard et al., 2009; Ghosh and O’Connor, 2014; Koyama and Mirth, 2016; Sano et al., 2015). These include Imp-L2 and dALS, which bind to and inhibit Ilp2 signaling (Arquier et al., 2008; Honegger et al., 2008) and Stunted, which stimulates insulin-like peptide secretion following amino acid ingestion in larvae (Delanoue et al., 2016). Imp-L2 has also been shown to interrupt insulin signaling and mediate cachexia-like wasting in adult flies transplanted with malignant tumours (Figueroa-Clarevega and Bilder, 2015; Kwon et al., 2015). In addition to these secreted factors, the Drosophila leptin homolog Unpaired-2 (Upd2) is secreted from fat body following prolonged high-sugar or high-fat feeding. Upd2 elaborates fat body signals that remotely relieve central GABA-ergic neuronal inhibition of IPCs, leading to increased insulin output (Rajan and Perrimon, 2012). Subsequent work has also revealed how glucagon-leptin-insulin axes are regulated by adipose tissue, demonstrating that AKH signaling in the fat body reduces Upd2 secretion, thereby inhibiting insulin release (Rajan et al., 2017). Like in mammals, recent work demonstrates the essential role of leptin-like molecules called Unpaired-1 (Upd1) for regulating obesity-related traits in adult flies. Disrupting brain-derived Upd1 production leads to phenotypes observed in mammalian obesity, including increased attraction to food cues, hyperphagia, increased weight and disruption in insulin secretion (Beshel et al., 2017). Thus, Upd1 and Upd2 may regulate distinct central neural circuits governing growth and weight regulation in adult flies.

In adult flies with chronic nutrient excess, Brent and Rajan (2020) recently reported that Upd2-regulated fat body signaling led to synapse reorganization in central GABAergic inhibitory neurons, reducing bouton number and promoting insulin release (Brent and Rajan, 2020). They also found that insulin feeds back on GABAergic neurons to increase their bouton number, and re-enforce a negative neural tone for insulin release. Thus, two nutrient surplus-sensing hormonal systems, Upd2 and insulin, signal through a structurally dynamic cellular circuit to regulate insulin output. Intriguingly, pancreatic islet δ-cells – which secrete somatostatin to inhibit β cell insulin secretion – were recently demonstrated to have dynamic contacts that regulate β cell secretion, and are regulated by hyperglycemia (Arrojo et al., 2019). Moreover, prior studies have shown that mammalian neuronal somatostatin secretion may be inhibited by leptin (Quintela et al., 1997). Further studies are needed to test the possibility that adipokines like leptin could remodel δ-cell contacts within pancreatic islets to regulate insulin (or glucagon) secretion.

Investigating the polygenic and multi-organ basis of diabetes in Drosophila

Diabetes mellitus is the most common metabolic disease worldwide, and the preponderance of evidence shows that there are both acquired (environmental) and intrinsic (genetic) risks in diabetes, whose sine qua non is hyperglycemia. Insulin resistance in specific tissues like fat, liver, and muscle is thought to ‘drive’ the pathogenesis of type 2 diabetes, the most common form in humans (Brown and Goldstein, 2008; Unger and Orci, 2010). Drosophila research has made important contributions to elucidating molecular and genetic regulation of insulin signaling in larval and adult organs (Alfa and Kim, 2016), but progress in this area has also been hampered by an overreliance on semi-quantitative assays (Britton and Edgar, 1998; Kockel et al., 2010; Puig et al., 2003) to assess insulin signaling. Future advances should be accelerated by the adoption of tissue-specific assays to quantify readouts of insulin receptor activation (like Akt/PKB phosphorylation), as described in a recent study (Park et al., in preparation). Likewise, investigations of glucagon resistance, another pathophysiological driver of human diabetes (Unger and Cherrington, 2012), should be advanced by quantitative assays of AKH output (Oh et al., 2019) and AKH-resistance.

Like in mammals, deficiency for insulin in adult flies elicits an elevation of circulating glucose (Park et al., 2014) as well as impaired regulation of trehalose, a glucose dimer (Broughton et al., 2008; Gronke et al., 2010). While loss of insulin signaling orthologues of IRS1/2, AKT and Insulin Receptor affects fertility, size, and lifespan, deficiency for these factors does not reliably produce adult hyperglycemia (Böhni et al., 1999; Park et al., 2014; Ugrankar et al., 2015). Likewise, loss of Upd2 does not cause hyperglycemia in adult flies (Rajan and Perrimon, 2012). This likely reflects compensatory increases of insulin output from IPCs (Park et al., 2014), analogous to human responses to insulin resistance (Hollenbeck and Reaven, 1987). Thus, like multi-organ pathogenesis of diabetes in mammals, hyperglycemia in flies may only manifest with peripheral insulin resistance combined with insulin secretion defects in the IPCs. Reconstituting polygenic and multi-organ mechanisms thought to underlie hyperglycemia in human diabetes is an important unmet goal, but well-matched to the experimental paradigms available in flies.

II. A gut feeling: intestinal interoception

The gastrointestinal tract is recognized as a central regulator of food intake and energy balance (Clemmensen et al., 2017; Soty et al., 2017). It is also a remarkably flexible organ system that can undergo marked adaptations in response to diet and internal state (Beumer and Clevers, 2021). Like its mammalian counterpart, the digestive tract of Drosophila is functionally regionalized (Figure 2A, Figure 4A) (Buchon et al., 2013; Marianes and Spradling, 2013; Miguel-Aliaga et al., 2018; O’Brien et al., 2011). It harbours a resident microbiota, and consists of different cell types similar to those found in the human GI tract, including digestive/absorptive enterocytes, hormone-secreting enteroendocrine cells and intestinal stem cells (Micchelli and Perrimon, 2006; Ohlstein and Spradling, 2006). Over the past decade, Drosophila has been leveraged to identify molecular and cellular mediators of intestinal plasticity, revealing a central role for adult intestinal stem cells; these have been recently reviewed elsewhere (Funk et al., 2020; Miguel-Aliaga et al., 2018). Here we review recent studies revealing other modes of intestinal sensing and adaptation.

Figure 4. Signaling and cell interactions coordinating fly metabolism.

(A) Signaling interactions regulating AKH output by CC cells by enteroendocrine cell-derived Bursicon α signaling. During starvation (top half), glucose entry into EE cells is diminished, Bursicon α is retained in EE cells, and AKH is secreted from CC cells, leading to subsequent catabolism of peripheral fat stores. CC cells also secrete Limostatin (Lst), a decretin-like hormone that suppresses insulin-producing cell function. In the fed state (bottom half), Bursicon α is secreted from enteroendocrine cells and suppresses Akh release from the CC cells. The magnified insert depicts enterocytes (EC, brown), enteroendocrine cells (EE, blue), and intestinal progenitors (enteroblasts (EB) or intestinal stem cells (ISC), yellow) (B) Gut responses to micronutrient consumption. The midgut has a central region with high luminal acidity harboring two types of specialised enterocytes: acid-producing copper cells (CC, dark orange) interspersed between interstitial cells (IC, light orange). In response to zinc ingestion by larval flies, Hodor in the interstitial cells sustains lysosomal acidification and activation of Tor signaling. This signaling increased food intake, dietary zinc preference, and increased Ilp release from IPCs (dark blue). (C) Gut responses to amino acid deprivation. In response to essential amino acid (EAA) deprivation, gut enterocytes secrete the neuropeptide CNMamide (CNMa), which may mediate the EAA feeding preference in flies deprived of dietary protein. Importantly, these enterocytes do not secrete CNMa in response to non-essential amino acid (NEAA) deprivation.

Interoception coordinates switching to catabolism

Drosophila is beginning to reveal both conserved and novel mechanisms of intestinal interoception that modulate food intake and choice. For example, reminiscent of the roles of the glucose transporter Glut2 in regulating the post-prandial secretion of the incretin hormone glucagon-like peptide-1 (GLP-1) from enteroendocrine cells (Cani et al., 2007), a Drosophila Glut1 homologue gates the release of enteroendocrine hormone Bursα (Scopelliti et al., 2019). Circulating Bursα normally signifies the ‘fed’ state, preventing secretion of AKH from CC cells, which are adjacent to the gastrointestinal tract. During starvation (or following Glut1 downregulation), however, Bursα is retained in EE cells, leading to AKH release and the consequent mobilization of peripheral fat stores (Scopelliti et al., 2019). In parallel, starvation also triggers the release of a decretin-like hormone (Limostatin, Lst; (Alfa et al., 2015). Lst suppresses insulin output through its action on a GPCR (LstR) expressed in IPCs (Alfa et al., 2015). Further studies are needed, however, to identify the mammalian islet orthologues of fly IPC Lst/LstR signaling (Alfa et al., 2015; Kuhre et al., 2019). In sum, studies of Bursα and Lst have revealed intestine-associated mechanisms that regulate AKH secretion, mediating the switch to a catabolic state that allows flies to withstand starvation.

Discovery of mechanisms regulating nutrient uptake and food preferences

Sufficient dietary amino acids, and micronutrients like the trace metals, can profoundly impact health and disease in humans. Drosophila studies have revealed unexpected mechanisms of intestinal micronutrient and amino acid sensing. The novel roles of an intestinal metal sensor in food intake regulation and growth control by enterocytes are a case in point. Redhai et al. identified a novel population of zinc-sensing enterocytes that sustain the voracious feeding of Drosophila larvae (Redhai et al., 2020). Within these enterocytes (known as ‘interstitial cells’), a zinc-gated chloride channel (Hodor) responds to zinc ingestion by sustaining lysosomal acidification and activating Tor signaling (Figure 4B). Hodor-mediated Tor signaling activity within these enterocytes leads to increased food intake and Insulin-like peptide release via an as-yet unidentified systemic signal. Intestinal Hodor also mediates a larval preference for dietary zinc (Redhai et al., 2020). Similar regulation of insulin secretion in response to micronutrient availability is conserved in mammals. In mice, for example, oral zinc administration enhances insulin secretion, likely through gastric inhibitory peptide (GIP) secretion triggered by the zinc-sensing GPR39 receptor on L- and K-cells in the gut (Moran et al., 2019). While Hodor signaling appears specific to insects, an orthologous system or a zinc-gated channel may exist in mammals with roles that might extend beyond regulation of food intake (Fernández-Gallego et al., 2020).

When deprived of dietary protein, Drosophila and other animals select a food source that contains a greater amount of dietary protein or essential amino acids (EAAs) (Raubenheimer and Jones, 2006; Theall et al., 1984). This suggests that food selection is geared toward acquiring specific macronutrient targets. How this choice is driven has remained a mystery, since known sensors of amino acids (AAs) including taste receptors like T1R1-T1R3 and intracellular factors like GCN2 and TOR do not discriminate between EAAs and non-essential AAs (NEAAs) (Efeyan et al., 2015; Nelson et al., 2002). Recent work shows that protein or EAA deprivation (but not NEAA deprivation) in flies induces production of the neuropeptide CNMamide (CNMa) in a specific population of enterocytes in the gut. Genetic silencing of the CNMa-CNMa receptor axis blocked the EAA-preference in these flies (Kim et al., 2021). This mechanism is reminiscent of how peripheral tissue induction of FGF21 in protein-deprived mammals can signal the brain to regulate feeding behavior (Hill et al., 2017; Hill et al., 2019; Solon-Biet et al., 2016). Moreover, gnotobiotic flies bearing an EAA-producing symbiotic microbiome exhibited reduced compensatory appetite for EAAs. By contrast, gnotobiotic flies carrying a mutant microbiome that failed to produce leucine or other EAAs displayed higher CNMa expression and greater compensatory EAA appetite (Kim et al., 2021). These findings reveal that different types of cells in the gut including enterocytes act as a frontline sensor to detect and respond to micro- and macro-nutrients. It also raises the possibility that these nutrient-sensing cells work together with the gut microbiome to establish nutrient homeostasis.

Finally, beyond canonical signals like peptide hormones, systemic metabolites may also play important roles in the modulation of food intake and choice. Two salient examples are the role of gut-derived citrate in promoting food intake in males, described below (Hudry et al., 2019) and the finding that pentose phosphate pathway activity in the female germline increased an appetite for sugar (Carvalho-Santos et al., 2020). Modulation in sugar appetite by the germline is achieved by regulating the expression of the fat-body-secreted satiety factor, Fit, a sexually dimorphic protein previously shown to suppress protein appetite and promote insulin-like peptide release (Sun et al., 2017). Investigating regulatory roles of metabolites on behavior in the context of inter-organ communication should emerge as an exciting and fruitful area of future research.

Mechanosensory mechanisms of feeding regulation

While a central role for intestinal nutrient sensing is also emerging from studies in mice (Clemmensen et al., 2017; de Araujo et al., 2020; Gribble and Reimann, 2019), two recent Drosophila studies remind us that other sensory modalities like mechanosensation may control acute feeding (Clemmensen et al., 2017; de Araujo et al., 2020; Gribble and Reimann, 2019; Min et al., 2020; Wang et al., 2020). The fly homologue of Piezo, a mechanotransduction channel, restrains feeding from at least two independent crop-innervating neuronal populations (IPCs in the brain and a rare population of enteric neurons); silencing or stimulating of either neuronal population results in an increase or a decrease of food consumption, respectively. Other work illustrates that Piezo also inhibits sugar intake through another layer of regulation. Six DH44-expressing neuronal cells, located adjacent to IPCs (Figure 2B), detect the nutritional value of sugar and consumption of sugar macronutrient specifically during food deprivation (Figure 2B)(Dus et al., 2015; Dus et al., 2011). When animals are satiated in the fed state, Piezo suppresses the function of DH44 neurons, thereby suppressing sugar intake (Oh et al., 2021). In future studies, it will be interesting to explore whether Piezo expression and/or activity are regulated by the postmating signals recently reported to increase maternal food intake through modulation of Ms neurons, a third population of crop-innervating neurons (Hadjieconomou et al., 2020). The recent finding of vagal mechanosensory neurons with a role in feeding regulation in mice (Bai et al., 2019; Kim et al., 2020) also raises the possibility that Piezo may play similar roles in mammals.

III. Sexually dimorphic control of gut function, metabolism and reproduction

There is a growing realization that many aspects of (patho)physiology differ between the sexes (Mauvais-Jarvis et al., 2017; Ober et al., 2008; Tannenbaum et al., 2019; Tramunt et al., 2020). Drosophila is no exception: important studies have uncovered sex differences in how insulin-producing cells and the fat body communicate to control larval growth (Millington et al., 2021; Rideout et al., 2015; Sawala and Gould, 2017). Unexpectedly, recent studies have revealed the importance of sex differences in intestinal cells, and uncovered novel gut-gonad axes in both males and females.

Sex matters: identifying new signaling axes between the gut and gonads

Most, if not all, organs of the adult fly display sex differences in gene expression (Leader et al., 2018) which impact features of adult physiology such as lipid metabolism (Sieber and Spradling, 2015; Wat et al., 2020). Recent studies have leveraged integrated genetic and physiological approaches to reveal the importance of sex differences in intestinal cells, and uncover novel gut-gonad axes in both males and females. An earlier study had indicated that the feces of adult Drosophila is unexpectedly predictive of both sex and reproductive status (Cognigni et al., 2011). The subsequent finding that, in the adult midgut (analogous to the mammalian small intestine), approximately 10% of genes are expressed and/or alternatively spliced in a sexually dimorphic manner (Hudry et al., 2016) further suggested sex-specific intestinal physiology. Since then, several studies have illuminated the nature and significance of these sex differences.

One prominent difference lies in the midgut: the adult intestinal stem cells (ISCs) that normally replenish the epithelia divide more often in females than males (Ahmed et al., 2020; Hudry et al., 2016). Increased ISC proliferation maintains the larger size of the midgut in virgin females compared to males, makes females more resistant to acute intestinal challenges such as infection, and allows increases of intestinal size during reproduction (Ahmed et al., 2020; Hudry et al., 2016; Regan et al., 2016; Reiff et al., 2015). This can be good for fly mothers – genetically preventing reproductive intestinal remodelling compromises their fecundity (Ahmed et al., 2020; Reiff et al., 2015). But this advantage comes at a significant cost: increased ISC proliferation renders female flies more susceptible to age-related dysplasia and tumorigenic insults (Ahmed et al., 2020; Hudry et al., 2016; Regan et al., 2016). Several mechanisms account for the sex differences in ISC proliferation. One is the intrinsic sexual identity of ISCs, which explains the basal higher proliferation rate of virgin female versus virgin male guts (Hudry et al., 2016). Second, after mating, a rise in circulating levels of juvenile hormone (JH) and ecdysone (an ovarian steroid hormone) further increases ISC proliferation in females, amplifying this sexual dimorphism (Ahmed et al., 2020; Reiff et al., 2015; Zipper et al., 2020).

A gut-gonad axis that sustains fertility and food intake

ISCs are not the only cells in the intestine that respond to – and subserve – reproduction. In female flies, a postmating rise in JH increases stem cell proliferation to yield a larger organ, but additionally remodels intestinal enterocytes to sustain fecundity (Reiff et al., 2015). Acting through intestinal bHLH-PAS domain proteins Methoprene-tolerant (Met) and Germ cell-expressed (Gce), JH signals directly to enterocytes to adjust their lipid metabolism by activating sterol regulatory element-binding protein (SREBP) and upregulating expression of genes involved in fatty acid synthesis and activation. Genetically preventing the reproductive, JH-driven metabolic remodeling of enterocytes reduces reproductive output. Mating also leads to an increased number of enteroendocrine cells and increased production of at least two of their peptide hormones. Neuropeptide F (the Drosophila homologue of Neuropeptide Y) signals back to the ovary to promote germline stem cell proliferation (Ameku et al., 2018). Bursicon α (Bursα, an insect-specific enteroendocrine hormone) signals, together with ecdysone, to a subset of enteric neurons that, through their release of Myosuppressin (Ms) peptide, control the expandability of the crop: a stomach-like organ (Hadjieconomou et al., 2020) (Figure 2B, Figure 5). The post-mating “awakening” of these enteric neurons is significant because, through their effects on the crop, they are responsible for the increased food intake apparent in female flies –like in many mammals– during reproduction. Indeed, preventing the reproductive remodelling of these enteric neurons reduces both reproductive hyperphagia and reproductive fitness (Hadjieconomou et al., 2020).

Figure 5. Sexually dimorphic control of gut function, metabolism and reproduction.

In males (upper half), testis-derived cytokines, including Upd1, upregulate Jak-Stat signaling within intestinal enterocytes (brown). This results in enhanced citrate secretion from the enterocyte, which in turn sustains spermatogenesis in the testis and promotes food intake through the action of an unknown neuron. In female flies (lower half), a postmating rise in juvenile hormone (JH) secreted from the corpora allata (CA) and 20-hydroxy ecdysone (20-HE) secreted from the ovary sustain increased intestinal stem cell (ISC, yellow) proliferation which maintains fecundity. JH also acts directly on enterocytes to adjust their lipid metabolism. After mating, enteroendocrine cells (blue) also increase their production of neuropeptide Y, which promotes germline stem cell proliferation in the ovary. The enteroendocrine cells and the ovary also secrete Bursicon α and 20-HE, respectively, to activate Ms neurons in the pars intercerebralis of the brain. The Ms neurons then increase the expandability of the crop, mediating the postmating rise in food intake seen in female flies.

It could be argued that the reproductive plasticity of all these different intestinal cell types is an insect peculiarity, arguably less relevant to humans and other mammals. By some measures, the nutritional demands of mammalian reproduction are less extreme (flies lay several times their weight in eggs every day), and mammalian adaptations like the placenta or post-partum nursing might have subsumed at least some of these nurturing roles. While descriptive, there is rather extensive data that argues otherwise: increased cell size and proliferation of intestinal epithelial cells has been reported in several mammals including mice and rats during pregnancy and/or lactation (Hammond, 1997; Nilaweera and Speakman, 2018). Similarly, while the reproductive plasticity of gastrointestinal innervation remains to be investigated, mammalian enteric neurons express sex- and reproductive-hormone receptors (Ameku et al., 2020), and enteroendocrine hormone levels change during reproduction (Johnson et al., 2019). These features suggest it will be productive to explore whether the human digestive system might be similarly modulated by reproductive cues to affect food intake.

Vive la différence

Like in females, singularities of the male gut sustain gametogenesis and fertility (Hudry et al., 2019). However, the mechanisms involved differ from those of females. For example, the male gonad “masculinizes” the enterocytes of a specific region of the intestine by upregulating Jak-Stat signaling, leading to rewiring of enterocyte carbohydrate metabolism and, ultimately, their secretion of citrate (Figure 5). Citrate derived from male enterocytes is then absorbed by the male gonad, and used to sustain spermatogenesis. Gut-derived citrate also acts on (as yet unidentified) neurons to promote food intake in males (Hudry et al., 2019). In mammals including humans, citrate is one of the highest circulating TCA cycle intermediates (Costello and Franklin, 2016; Hui et al., 2017; Mycielska et al., 2009), and recent work in pigs has revealed citrate fluxes across specific tissues (Jang et al., 2019).

Further studies exploring possibly conserved roles of citrate in sex-biased physiology seem warranted. The identification of a fly gonad-to-gut signaling axis also highlights that male fly gonads are adjacent to the gut region they communicate with, indicating previously unappreciated spatial stereotypy in the arrangement of internal organs (Hudry et al., 2019). This organ geometry could facilitate or restrict inter-organ communication, suggesting under-explored dimensions to the study of metabolic disorders and interventions such as obesity and/or gastric bypass.

Mechanistically, these studies have uncovered gut-gonad axes that are sex-specific and govern aspects of whole-body physiology including (but not confined to) reproductive output, and so may have broader implications for human health and diseases. They demonstrate that non-gonadal organs such as the intestine have a sexual identity that is (patho)physiologically significant. Different cell types within an organ acquire their sexual fate through different mechanisms – hormonal and cell-intrinsic. The mechanisms that specify sexual fate are actively maintained in the adult, and are therefore genetically reversible. This raises the possibility that they may be plastic in the context of (patho)physiology, motivating and warranting studies to identify internal or environmental cues that modulate the sexual fate of specific intestinal cells.

Conclusions and Prospects

Recent findings highlighted here illustrate the formidable experimental advantages of Drosophila for investigations of metabolism and inter-organ communication that exploits circulating hormones, short-acting neuropeptides and neural signaling. These advantages include: (1) the ability of investigators to perform high-throughput in vivo screens or assess cell interactions in ways less feasible or affordable in mammalian systems, (2) the use of powerful in vivo assays to quantify fly hormones, neuropeptides and metabolites often in a single fly, (3) the availability of behavioral or other physiological assays to discern and measure functional and signaling links between organs like the brain, endocrine cells, intestines, gonads, and fat body, (4) the flexibility of performing complementary gain- or loss-of-function genetics targeted to specific tissues and cell types, (5) the amenability of several binary systems that can be used to manipulate the activities of multiple tissues simultaneously and (6) the outpouring of new fly strains that enable superlative control of gene and cell function the gastrointestinal tract, endocrine cells, and other organ systems (Ariyapala et al., 2020; Kockel et al., 2019; Lim et al., 2020). These findings support the view that Drosophila studies will continue to unveil general principles about metabolism and metabolic diseases, serving at the vanguard of modern discoveries in these fields.