Insulin signaling and pharmacology in humans and in corals

Intended audience

This review aims to provide valuable knowledge to a diverse scientific audience, encompassing researchers and academics engaged in disciplines such as biochemistry, cellular biology, endocrinology, and marine biology. It seeks to furnish scientists specializing in human metabolism and endocrine disorders with novel insights gleaned from non-mammalian systems. Concurrently, it aspires to equip marine biologists and coral ecologists with a deeper understanding of potential INS signaling within coral biology. Moreover, this review will be beneficial for pharmacologists, particularly those seeking to apply their understanding of human INS signaling to burgeoning fields such as “coral pharmacology.” Educators may find this comparative study valuable as a teaching resource, and students will gain exposure to the application of established biological concepts to innovative contexts. Ultimately, by delineating connections between humans and corals, the review aims to foster interdisciplinary dialogues, encourage collaborative research initiatives, and contribute to both human and coral health.

Evolutionary conservation of INS

INS signaling is an evolutionarily ancient function, as evidenced by the fact that INS-like molecules have been identified in prokaryotes, microbial eukaryotes, insects, invertebrates (including Hydrozoa), and plants (Le Roith et al., 1980; LeRoith et al., 1981; Baig & Khaleeq, 2020). Antibodies raised against human INS recognize INS-like material from unicellular eukaryotes such as Tetrahymena pyriformis, a ciliated protozoan, and Neurospora crassa (Muthukumar & Lenard, 1991; Kole, Muthukumar & Lenard, 1991) and Aspergillus fumigatus, both fungi, and even prokaryotes (LeRoith et al., 1981). The fact that both prokaryotes and eukaryotes synthesize INS suggests that it may play a role in co-evolution.

Conservation at the sequence level is mirrored by conservation at the functional level (Abou-Sabe’ & Reilly, 1978; Le Roith et al., 1980). For example, effects of mammalian INS on E. coli have been described (Abou-Sabe’ & Reilly, 1978; LeRoith et al., 1981); similarly, INS shows metabolic effects on N. crassa cells such as enhanced glucose metabolism, enhanced growth, improved viability, and accumulation of intracellular sodium (Muthukumar & Lenard, 1991; Kole, Muthukumar & Lenard, 1991). INS-like preparations from more primitive organisms have effects on rat cells (Schmidt, Siegel & Creutzfeldt, 1985; Aguan et al., 1994; Cheng et al., 2007). These functional effects are likely achieved through a phosphorylation cascade as shown by the enhanced phosphorylation of specific proteins on serine/threonine and tyrosine residues (Kole, Muthukumar & Lenard, 1991).

Recently, an INS-INSR pair has been described in detail for Acanthamoeba castellanii, an early mitochondrial unicellular eukaryotic organism (Baig & Khaleeq, 2020). Not only did they show typical mammalian INS-induced effects on Acanthamoeba cells, but they also investigated the anti-diabetic drug metformin, and conducted homology modeling of the putative Acanthamoeba INS-INSR pair. This study strongly supports the notion of a high degree of conservation of the INS-INSR pair across billions of years of evolution and pioneers the use of a human antidiabetic drug (metformin) in the context of a primitive organism.

Most recently, remote homologues of INS and INSR have been identified in the stony coral, Pocillopora damicornis (Roger et al., 2022), using a new bioinformatics pipeline for identifying functions in non-model organisms (Kumar et al., 2023) based on remote homology detection suitable for comparison of sequences that show large divergence due to evolutionary separation called HHblits (Remmert et al., 2011). The conservation of INS signaling in corals is also seen in another cnidarian, Hydra, which belong to another phylum of cnidarian. Corals have evolved before the split into Deuterostomia such as humans and Protostomia like the model organisms C. elegans and Drosophila melanogaster 700 million years ago (mya). The Anthozoa-Hydrozoa separation, i.e., the coral-Hydra separation, occurred >500 mya (Khalturin et al., 2019). In Hydra, a receptor protein-tyrosine kinase responding to an INS-like molecule was found to be involved in regulating cell division and differentiation (Steele et al., 1996).

These evidences for evolutionary conservation of INS and its function open the door to comparison of model and non-model organisms. Most of our understanding of the function of INS comes from studies in human as a model organism. We will therefore first review what is known about INS in humans and then compare the findings in detail to one non-model organism, Pocillopora damicornis. Because corals are threatened by extinction, we hope that the extension of our understand from the well-studied human INS pharmacology may provide clues to how we could treat the growing problem of coral bleaching and stimulate a new era of “coral pharmacology”.

Introduction to glucose homeostasis and insulin function in humans

The primary source of energy for most cells in the body is glucose and it is also a substrate for many biochemical reactions (Nakrani, Wineland & Anjum, 2023). Blood glucose levels in the body are maintained and balanced by glucose homeostasis (Da Silva Xavier, 2018), as outlined in Fig. 1. Glucose, as a highly polar molecule, cannot diffuse into the lipid membranes of cells and its transport is therefore facilitated by glucose transporters (GLUTs), a family of 12 members (Navale & Paranjape, 2016). Glucose is phosphorylated upon entering the cell and is broken down through glycolysis, followed by oxidative phosphorylation of pyruvate in the TCA cycle, generating ATP (Fukunaga & Hunter, 1997; Watowich et al., 1999). Alternatively, it is polymerized to glycogen for storage of excess glucose (Nakrani, Wineland & Anjum, 2023). To maintain the balance between these opposing processes resulting in regulated blood glucose levels hormones are produced from a group of multicellular endocrine cells called Islets of Langerhans (Da Silva Xavier, 2018). On average, the human pancreas contains 3.2 million islets. Islets consist of four major types of cells: α-cells, β-cells, δ-cells, and pancreatic polypeptide (PP)-cells (Erlandsen et al., 1976; Da Silva Xavier, 2018). β-cells produce and store insulin (INS) which lowers blood glucose levels, while α-cells produce glucagon (GCG) raising blood glucose concentrations, and δ-cells produce somatostatin (SST) which inhibits the secretion of growth hormones and GCG, while PP cells secrete gastrointestinal and intestinal enzymes (Erlandsen et al., 1976), shown in Fig. 1.

INS release is triggered when the glucose concentration in blood rises above 90 mg/dL (5 mM) (Steiner et al., 1967; Fu, Gilbert & Liu, 2013). GLUT1 mediates intracellular glucose transport in the β-cells and triggers the immediate release of INS into the blood from β-cells. INS then binds to insulin receptors (INSRs) on target tissues (Steiner et al., 1967; Xu, Paxton & Fujita-Yamaguchi, 1990; Bremser et al., 1999). This is followed by a cascade of phosphorylation reactions initiated by INSR resulting in mitogenic and widespread metabolic effects of INS such as activation of the phosphatidylinositol-3-kinase (P13K) signaling pathway as described in detail below and references (Xu, Paxton & Fujita-Yamaguchi, 1990; Jones et al., 1991; Vainikka et al., 1994; Fukunaga & Hunter, 1997; Watowich et al., 1999; Hennige et al., 2000; Zhang et al., 2006; Kuo et al., 2007). INS signaling stimulates glucose translocation by INS-responsive GLUT4 and uptake of glucose by target tissues (Steiner et al., 1967; Kawanishi et al., 2000). GLUT4 is found in skeletal muscle, heart, brain and adipose tissues (Navale & Paranjape, 2016). Many putative intra-islet messengers have been implicated in regulating INS secretion, including ATP, Zn²⁺, γ-aminobutyric acid (GABA), and glucagon-like peptide-1 (GLP-1) (Reetz et al., 1991; Franklin & Wollheim, 2004; El et al., 2021). GABA released from β-cells binds and activates GABA_A receptors on α- and δ-cells, which in turn mediates glucose-dependent GCG release and increases SST secretion via activation of INSR on α- and δ-cells, respectively (Xu et al., 2006; Braun et al., 2009). Glucose-dependent insulinotropic polypeptide (GIP) and GLP-1 are secreted in response to ingestion of glucose and amino acids in the gut (El et al., 2021). In the pancreas, binding of GIP and GLP-1 to G protein coupled receptors (GPCRs) such as GLP-1R and glucagon receptor (GCGR) activates adenylyl cyclase/adenylate cyclase 8 (ADCY8) increasing intracellular cAMP signals and stimulates release of INS, GCG and SST (Moreau et al., 2006; Cheng et al., 2007; Fridlyand & Philipson, 2016). GCG from the pancreatic α-cells is the primary proglucagon derived peptide (PGDP) and GLP-1 and GLP-2 are related to GCG as they are co-encoded within the same proglucagon gene (Drucker, 2005). GLP-1 has been reported to improve insulin resistance and increase insulin sensitivity in obese and diabetic humans by modulating endoplasmic reticulum stress response via mTOR signaling pathway inhibition and activation of central GLP-1R (Sinclair & Drucker, 2005; Parlevliet et al., 2010; Jiang et al., 2018). GLP-2 is co-secreted with GLP-1 in the gut but very little is known about its action on glucose regulation in humans. So far, it is clear that GLP-2 is not involved in INS release (Schmidt, Siegel & Creutzfeldt, 1985; Sinclair & Drucker, 2005; Amato, Baldassano & Mulè, 2016). However, when GLP-2 is injected into healthy non-obese humans and diabetic patients, it increases glucagon secretion in plasma (Meier et al., 2006; Amato, Baldassano & Mulè, 2016). In humans afflicted by obesity and type 2 diabetes there is an association between GLP-2 and INS resistance with beneficial effects on glucose metabolism (Amato, Baldassano & Mulè, 2016). The main function of GLP-2 are energy uptake regulation and maintenance of intestinal mucosal integrity, function, and morphology (Amato, Baldassano & Mulè, 2016).

In summary, INS controls the glucose levels in the body and stimulates update of glucose which accumulates and is converted to glycogen and fat within muscles, liver, and adipose tissues (Quesada et al., 2008). GCG counterbalances INS action by activating glycogenolysis and gluconeogenesis in the liver. SST inhibits endocrine hormone secretions. The evidence so far further suggests that GLP-1 is involved in INS release, and less in INS signaling, while the role of GLP-2 in INS signaling has not yet been widely investigated.

Understanding these hormone actions and their downstream signaling pathways is crucial because of the clinical relevance for diabetes. Diabetes is a metabolic disorder in which the body produces less INS or has reduced sensitivity to INS. GLP-1R agonists are well-studied for the treatment of diabetes as they regulate blood glucose by increasing INS secretion and inhibit GCG secretion and appetite (Drucker, 2005). In the standard treatment for advanced diabetes, INS is supplied to compensate (Drucker, 2005). However, the INSR itself has also been considered a potential drug target to stimulate INS signaling by using INS ligands that directly bind to and activate the receptor (Kumar, Vizgaudis & Klein-Seetharaman, 2021). Other proteins such as GLUT-4 are also drug targets (Bouché et al., 2004), and one of the most successful drugs to treat diabetes, metformin (Bouché et al., 2004), likely has multiple targets, including the INSR (Bouché et al., 2004).

Systems biology of INS in humans: INS-related signal transduction cascades

The complexity of the INS-related signal transduction pathways is depicted in Fig. 2, and the proteins involved are listed in Table 1. Proteins are separated by pathway involvement based on whether the pathways are initiated by GCG, GLP-1, INS, or SST. Note that GLP-2 is omitted from Fig. 2 due to our gaps in knowledge of how it interfaces with the action of the other hormones and downstream signaling pathways. Full protein names, UniProt entry names, PDAM ID, E-value, P-value, sequence identity, similarity and references regarding biological function are provided and all protein isoforms of a related gene are listed in Tables 1 and 2 are discussed together. A total of 75 proteins (excluding isoform counting) are involved across the three pathways in humans (GCG and GLP-1 receptor binding events are treated as one pathway for simplicity). Of these, 17 proteins function exclusively within the GCG/GLP-1 signaling pathway, while 36 are exclusive to canonical INS signaling, and 10 are exclusive to the SST pathway. Eleven proteins are involved in two of the pathways, while only one protein is ubiquitous to all three pathways AKT (RAC-α/β/γ serine/threonine-protein kinase). Inhibitory signaling is found in GCG and INS pathways, while only stimulatory signaling is maintained in the SST pathway.

Overview of the INS signaling pathway. The INSR is a disulfide-linked tetramer that contains two α and two β subunits located on the cell surface (Xu, Paxton & Fujita-Yamaguchi, 1990; Bremser et al., 1999; Watowich et al., 1999; Kuo et al., 2007; Kumar, Vizgaudis & Klein-Seetharaman, 2021). The α subunits contain four INS binding sites in the extracellular domain (Xu, Paxton & Fujita-Yamaguchi, 1990; Bremser et al., 1999; Kumar, Vizgaudis & Klein-Seetharaman, 2021). The β subunits contain the intracellular tyrosine kinase domains (Xu, Paxton & Fujita-Yamaguchi, 1990; Kumar, Vizgaudis & Klein-Seetharaman, 2021). INS binding to the α subunits activates tyrosine kinase by changing conformations of the β subunits (Xu, Paxton & Fujita-Yamaguchi, 1990; Bremser et al., 1999; Kuo et al., 2007). This causes phosphorylation of insulin receptor substrates (IRS) and SHC which activates kinase cascades such as the PI3K/AKT and the MAPK signaling pathways (Xu, Paxton & Fujita-Yamaguchi, 1990; Jones et al., 1991; Vainikka et al., 1994; Fukunaga & Hunter, 1997; Watowich et al., 1999; Hennige et al., 2000; Zhang et al., 2006; Kuo et al., 2007). IRS activates PI3K to produce the secondary messenger PI-(3,4,5)trisphosphate (PIP3) by accelerating phosphatidylinositol (PI) phosphorylation across the membrane (Vainikka et al., 1994; Li & Sun, 1997; Fantin et al., 1998). IRS-1 integrates signals from different pathways and is considered the master regulator of INS sensitivity (Xu, Paxton & Fujita-Yamaguchi, 1990; Bremser et al., 1999; Kuo et al., 2007). SHIP2 acts as a PI3K antagonist and is involved in the downregulation of the AKT pathway by PIP3 degradation (Jones et al., 1991; Chardin et al., 1993; Habib et al., 1998; Zhang et al., 2006). PDK1/2 is recruited by PIP3 and phosphorylates AKT on serine/threonine residues (Li & Sun, 1997; Fukunaga & Hunter, 1997; Fantin et al., 1998). Subsequently, AKT phosphorylation activates several substrates including FOXO1 proteins that regulate TCF for gene transcription and GSK3 which regulates GLUT4 translocation to the cell membrane (Boyle et al., 1991; Holst et al., 1996; Shaodong et al., 1999; Nikoulina et al., 2000). Activation of IRS-1 also binds to GRB2 and SOS to activate RAS (Lowenstein et al., 1992; Chardin et al., 1993; Umikawa et al., 1999; Guil et al., 2003; Zhang et al., 2010). Activated SHC catalyzes GTP exchange by activating RAS on the plasma membrane (Warner et al., 2000; Hennige et al., 2000; Guil et al., 2003; Rönnstrand, 2004; Zhang et al., 2010; Yin et al., 2019); RAS activation leads to MAPK cascade activation via ERK phosphorylation (Sgouras et al., 1995; Gupta et al., 1996; Tee et al., 2002; Yang et al., 2012; Yin et al., 2019). The AKT pathway also mediates cell survival including growth, differentiation, and inhibiting apoptosis. INS synthesis at the translational level is mediated by mTOR which phosphorylates INS and promotes INS synthesis (Dalby et al., 1998; Kim et al., 2002; Tee et al., 2002). On the other hand, phosphorylation of serine residues on IRS-1 by its downstream effector protein kinase C-ζ (aPKC) which downregulates INS signaling and impairs P13K activity, constitutes negative feedback mechanism in response to INS (Vainikka et al., 1994; Kim et al., 2002; Kuo et al., 2007; Lee et al., 2008). In response to stimuli such as growth factors, IRS-1 phosphorylates multiple serine/threonine kinases such as p70 S6 kinase (P70S6K), mTOR and JNK (Sluss et al., 1994; Pullen et al., 1998; Deak et al., 1998; Watowich et al., 1999; Berns et al., 2004; Kuo et al., 2007; Nguyen et al., 2018).

Overview of GCG signaling pathway. As described above, glucose levels in the body are largely dependent on the coordinated release of GCG by α-cells and INS by β-cells of pancreatic islet, respectively, as shown in Fig. 1 (Orskov, Wettergren & Holst, 1993; MacNeil et al., 1994; Quesada et al., 2008). During hyperglycemia, INS secretion from β-cells is induced; however, during hypoglycemia α-cell secretion of GCG is induced. GCG exerts action via activation of GCG signaling and ADCY8 by coupling to GPCRs such as GCGR and GLP-1R (Orskov, Wettergren & Holst, 1993; MacNeil et al., 1994; Wang, Liang & Wang, 2013). The effects of INS are thus counterbalanced by GCG and the GCG signaling pathway promotes glucose production (Orskov, Wettergren & Holst, 1993; MacNeil et al., 1994; Quesada et al., 2008). GCG binding activates the receptors GNAS and GNAQ of G protein-mediated signaling (Orskov, Wettergren & Holst, 1993; Montrose-Rafizadeh et al., 1999; O’Donovan et al., 1999; Pak, Pham & Rotin, 2002; Alvarez-Curto et al., 2016). GNAS activates the PKA pathway and cAMP production (Orskov, Wettergren & Holst, 1993; MacNeil et al., 1994; Montrose-Rafizadeh et al., 1999). An increase in intracellular cAMP production phosphorylates CREB transcription factor to increase gluconeogenesis (Zhang & Bieker, 1998; O’Donovan et al., 1999). PKA activation also results in the inhibition of glycolysis by the inactivation of PFKFB which is one of the key enzymes in glucose metabolism (Algaier & Uyeda, 1988; Miki, Nagashima & Seino, 1999; Wang et al., 2000; Wu et al., 2002). GNAQ activates the IP3 calcium (Ca²⁺) signaling pathway and releases Ca²⁺ intracellularly which results in ERK1/2 phosphorylation causing CREB activation (Marklund et al., 1993; Sgouras et al., 1995; O’Donovan et al., 1999; Alvarez-Curto et al., 2016). Increased Ca²⁺ in GCG signaling is mediated by activated ADCY coupled to GLP-1R and GCGR. This activates PKA and phosphorylates Ca²⁺/calmodulin-dependent kinase II (CAMKII) in an IP3R- and Ca²⁺-dependent manner (Semizarov et al., 1998; Holz et al., 1999; Wang et al., 2000; Wu et al., 2002; Ding et al., 2004; Guan, Hou & Ricciardi, 2005; Glesne & Huberman, 2006; Diskar et al., 2010). Calmodulin (CALM) is a Ca²⁺ binding messenger activated upon binding of intracellular secondary Ca²⁺ and primarily transduces the Ca²⁺ signal in pancreatic cells (Chattopadhyaya et al., 1992; Tsang et al., 2006). Activation of CAMKII promotes FOXO1 nuclear translocation and plays a role in hepatic glucose production in response to fasting (Chattopadhyaya et al., 1992; Shaodong et al., 1999; Tsang et al., 2006). Hepatic expression of SMEK1/2 is also up-regulated during fasting, which is a key regulator of gluconeogenesis and elevates plasma glucose levels (Chowdhury et al., 2008). This causes dephosphorylation of CRTC2 which is responsible for transcriptional activation of gluconeogenic genes (O’Donovan et al., 1999; Iourgenko et al., 2003; Chowdhury et al., 2008).

Overview of SST signaling pathway. SST is a hormone that is involved in the inhibition of endocrine secretions such as INS, GCG, gastrin, and growth hormones as well as cell proliferation (Yamada et al., 1992; Pasquali et al., 2001; Luque & Kineman, 2018). SST binds to GPCRs called somatostatin receptors (SSTRs), of which there are five isoforms (SSTR1 to SSTR5) (Pasquali et al., 2001). Each receptor has distinct functions, and upon activation induce cascades of signaling pathways including protein tyrosine kinase activity (Yamada et al., 1992; Panetta et al., 1994). Upon tyrosine kinase stimulation, cytoplasmic protein-tyrosine phosphatase SHP1 is activated which triggers anti-proliferative and pro-apoptotic signals such as NF-kB, P53/Bax, and JNK (Sluss et al., 1994; Lisnock et al., 2000; Miao et al., 2000; Keilhack et al., 2001). Meanwhile, activation of SHP2 dephosphorylates the P13K/AKT and MEK pathways (Jones et al., 1991; Marklund et al., 1993; Keilhack et al., 2001; Zhang et al., 2006). This causes inhibition of cell proliferation via upregulation of P27, P21 cyclin kinase inhibitors, and the Zac1 tumor suppressor gene (Harper et al., 1993; Schneider, Montenarh & Wagner, 1998; Ishida et al., 2000). Pancreatic δ-cells secrete SST in response to elevated extracellular glucose concentrations (Hauge-Evans et al., 2009) and, within the islets, SST acts as a paracrine inhibitor of INS and GCG secretion (Hauge-Evans et al., 2009). SST is also a hypothalamic peptide known to inhibit somatic growth by inhibiting pituitary growth hormone (Hauge-Evans et al., 2009; Stengel, Rivier & Taché, 2013).

Introduction to corals

Corals are colonial marine invertebrates (cnidarians) that depend on a symbiotic relationship with dinoflagellate algae of the family Symbiodiniaceae (LaJeunesse et al., 2018). The algae harvest light and synthesize nutrients in exchange for shelter and nitrogen sources (Putnam et al., 2017). Coral reefs cover only 0.1% of the ocean floor but are home to the largest density of animals on earth, rivaling rain forest habitats in species diversity (LaJeunesse et al., 2018). The symbiosis, which was originally thought to be restricted to algae, is now known to extend to a much more complex community than anticipated with thousands of bacteria, bacteriophages, viruses, and fungi, in addition to endosymbiotic algae (Bourne et al., 2009). The entirety of the organism community in a coral is referred to as a holobiont, while the individual cnidarian host animals forming the colonies are called polyps.

The holobiont is characterized by balanced host-microbe molecular interactions. The complexity of these interactions in relation to stress and disease resistance, and recovery grow with every new study as questions arise regarding what molecules are responsible for symbiosis establishment and partner coexistence (Ainsworth & Gates, 2016; Kelly et al., 2021). These inter-partner exchanges are still poorly understood, and this is a particularly severe gap in our knowledge, since it is at the heart of the worldwide phenomenon of coral reef bleaching which refers to the breakdown of symbiosis (particularly the cnidarian host and endosymbiotic algae) due to thermal stress and high irradiance, including that brought about by global climate change. A recent study assessed 100 worldwide locations and found that the annual risk of coral bleaching has increased from an expected 8% of locations in the early 1980s to 31% in 2016 (Hughes et al., 2018; IPCC, 2022). Human impacts on coral reef ecosystems threaten fisheries and tourism, industries valued at hundreds of billions of dollars annually (Putnam et al., 2017). We are in urgent need of innovative solutions to increase corals’ resiliency to anthropogenic activities and facilitate their survival.

Climate change driven coral bleaching has now been recognized as the leading cause of the worldwide decline of coral reef cover and, overall, the biggest threat to reef-building coral survival (Hughes et al., 2017). Mass bleaching events have increased both in frequency and severity since the first recorded event in the 1980s (Oliver, Berkelmans & Eakin, 2018) and show no signs of reprieve as ocean warming gets compounded with traditional climate patterns such as the El Niño-Southern Oscillation (McPhaden, Zebiak & Glantz, 2006). Coral bleaching is the common term used to describe dysbiosis in the coral holobiont, specifically, the breakdown of symbiosis (xenophagy and/or expulsion) between the cnidarian host and the dinoflagellate endosymbionts (i.e., dinoflagellates provide most of the coral tissue pigmentation and as dysbiosis progresses, the tissue becomes transparent, thereby revealing the white calcium carbonate skeleton) (Suggett & Smith, 2020). While the full signaling cascade leading to dysbiosis is still poorly defined, we know it leads to damage to cell membranes, lipids, proteins and DNA via nitro-oxidative stress (i.e., the accumulation of free radicals, reactive oxygen species and reactive nitrogen species), a failing antioxidant machinery (e.g., catalase, ascorbate peroxidase, superoxide dismutase) and the organisms’ innate immune response (Weis, 2008; Lesser, 2011; Suggett & Smith, 2020).

Nitro-oxidative stress is common across aerobic systems and, in the case of the coral holobiont, has been associated with heat-damaged chloroplasts (Tolleter et al., 2013; Alderdice et al., 2022) and other damages to the photosynthetic mechanism through heat and light (Gleason & Wellington, 1993; Lesser & Farrell, 2004; Tolleter et al., 2013; Downs et al., 2013; Alderdice et al., 2022), the composition of the thylakoid membrane lipids (Tchernov et al., 2004), the potential for upregulation of ROS scavenging capacity and molecular chaperons during periods of thermal-, light-, or osmotic stress and hypoxia (Gardner et al., 2016; Levin et al., 2016; Ochsenkühn et al., 2017; Aguilar et al., 2019; Alderdice et al., 2021, 2022), seawater trace metal concentrations (Shick et al., 2011; Ferrier-Pagès, Sauzéat & Balter, 2018; Biscéré et al., 2018; Reich et al., 2023) and N:P ratios (Fabricius et al., 2013; Pogoreutz et al., 2017). Furthermore, bleaching has also been associated with the seawater carbonate saturation horizon and dissolved CO₂ levels (i.e., ocean acidification) (Anthony et al., 2008; Crawley et al., 2010).

In the context of coral bleaching induced by heat stress, a study involving the tropical sea anemone Aiptasia pallida identified over 500 up-regulated genes, categorized into Cluster I linked to immunity and apoptosis and Cluster II related to protein folding, with potential regulators influenced by transcription factors NFκB and HSF1. A total of 337 genes in symbiotic anemones exhibited declining expression levels before visible bleaching, suggesting their involvement in algal symbiosis loss (Cleves et al., 2020). These findings hint at potential interactions of these genes with the INS signaling pathway, considering known roles of INS signaling in apoptosis and immune responses (Yuyama et al., 2018). Furthermore, experiments inducing ERK activity in corals via UV radiation and thermal stress (Courtial et al., 2017) and heat-shock experiments on Aiptasia (Sloan & Sawyer, 2016) contribute to our understanding of ERK and AKT phosphorylation and MAPK activities in these organisms, potentially implicating the INS signaling pathway in coral bleaching.

Glucose regulation in corals: an opportunity for understanding INS action in non-model organisms

The symbiotic algae provide as much as 90% of the energy corals consume by light harvesting and photosynthesis (Gierz, Forêt & Leggat, 2017). Thus, corals must be able to measure and regulate nutrient balance (Cunning et al., 2017). Given the crucial role of INS signaling for this task in other organisms, we here hypothesize that INS signaling may also exist in corals, although this hypothesis is purely theoretical and remains to be experimentally validated. Support for this hypothesis comes from transcriptomic studies (Yuyama et al., 2018). A comparison between the expression of INS signaling related genes in the presence and absence of the symbiotic algae strongly suggests that INS signaling is induced at the transcriptomic level in response to algal density in the tissue. A likely interpretation of this finding is that corals need to respond to the sugars produced by the algae through light harvesting and perhaps too much sugar could have detrimental effects on corals, similar to the diabetic response through aberrant INS signaling in humans. The symbiotic interaction between algae and coral involves algae entering the host, and the facilitation of energy and metabolite exchange. Algae utilize seawater substrates to synthesize a spectrum of organic compounds, effectively transferring vital nutrients, including amino acids, small peptides, sugars, carbohydrates, and lipids, to coral cells with glucose being a major metabolite transferred in this exchange as demonstrated by Burriesci and colleagues (Burriesci, Raab & Pringle, 2012). It is also possible that the mechanism for bleaching (loss of symbiotic algae from the coral holobiont) involves an imbalance in nutrient regulation and possible involvement of the INS signaling pathway. This raises an interesting speculation: could corals have diabetes, and could insulin resistance be related to the bleaching that is threatening coral species survival? While corals of course do not have Langerhans islets nor blood, the diabetes analogy at the molecular level may stimulate new ways of thinking about coral and human health (see below).

Indeed, there is evidence for INS signaling in corals at the molecular level. First, remote homology detection using HHblits have identified homologues for human IR and INSR in corals (Roger et al., 2022). HHblits is a so-called Hidden Markov Model (HMM)-based alignment approach developed by Remmert et al. (2011). Unlike traditional HMM profiles, in HHblits, both query and template are HMMs. The search for homologues is through an HMM-HMM alignment and the query HMM is generated by using amino acid distributions which makes this method extremely sensitive. It has been shown that, in many instances, HHblits successfully outperforms the identification and alignment of remote homologues, as compared to the traditional profile HMM approach, such as HMMER3 (Remmert et al., 2011). Given the 700 million years of evolution between corals and humans, this enhanced sensitivity of HHblits is instrumental to the comparison between corals and humans. We have already described the sequence alignments of the ligand-receptor pair, INS with INSR for human and for coral (Roger et al., 2022). In both cases, the alignments were identified with high confidence and cover a large fraction of the sequences: 1,164 out of 1,382 amino acids in the case of INSR and 101 out of 110 in the case of INS. The comparison of the sequences of human insulin (UniProt ID P01308) and Pocillopora damicornis (pdam) protein pdam_00006633, and the extracellular domain of the human INSR (UniProt ID P06213) and pdam_00013976 are shown in Tables 3A and 3B, respectively.

The finding of a human INS homologue in Pocillopora damicornis has prompted us to test the effect of human INS on corals experimentally (Roger et al., 2022). An average 20% reduction in viability at 100 µg/mL INS concentration was observed in line with its proteotoxicity in other systems (Rege et al., 2020). Due to the importance of INS administration in diabetes, its folding and stability has been studied extensively (Weiss & Lawrence, 2018; Liu et al., 2018). High concentrations of salts are known to promote INS aggregation and misfolding (Grudzielanek et al., 2007; Chatani et al., 2014), and the use of seawater in our experiments may induce similar effects, which may be the cause for the observed cytotoxicity.

As shown in Fig. 2, not only INS and INSR, but a total of 75 proteins (excluding isoform counting) are involved across the three INS related pathways in humans. Application of the non-model organism pipeline described above (Kumar et al., 2023) reveals that the majority of downstream signaling proteins, namely 67 of the 75 human proteins, are likely conserved in Pocillopora damicornis. In Fig. 2, all human proteins shown in black have a predicted Pocillopora damicornis homologue, while those shown in red do not. Crosstalk between the SST, CGC and INS pathways is mediated by several proteins that are common to two or even all three pathways. We were not able to identify suitable Pocillopora damicornis homologues for eight proteins: GLP-1, GLP-1R, CGC, CGCR, SST, SSTR, BAD, and CALM, as judged by their poor e-values as well as low percent alignments of amino acids. It is important to realize that GLP-1R, CGCR and SSTR are all GPCR’s and thus it is difficult to differentiate GPCR variation within organisms as compared to across organisms. This complication has been discussed in detail, and it was proposed that the GPCR repertoire of Pocillopora damicornis is 151 as compared to 825 in human (Kumar et al., 2023). The results of the remote homology search can be accessed through supplementary file S4 of that article, where all three human GPCR sequences (CGCR, SSTR and GLP-1R) are never ranked first for any of the coral GCPR candidates. The most closely related GPCR is pdam_00008152-RA, which is more similar to the GLP-2 receptor than those three. Thus, it may be possible that GLP-2 is a more ancient modulation of the INS pathway than the GLP-1, CGC and SST pathways (Amato, Baldassano & Mulè, 2016). This finding suggests that the lack of interest in GLP-2 in previous human studies reviewed above is perhaps unjustified. It is important to note that these new evidences are computational only and await future experimental validation. While SST, GLP-1 and GCG are the ligands initiating their respective signaling pathways through their respective GPCR’s, BAD is located at the effector end of the main INS pathway, indicating that this pathway is mostly functional. Similarly, CALM functions only to stimulate PYGL in the GCG signaling cascade, and TSC1/2 in INS signaling. Importantly, there is a clear homologue of both INS and the INSR present.

Outlook: coral pharmacology

In this review, we have pointed out numerous possible analogies between human and coral INS biology. The enormous pharmacological importance of treating INS resistance in diabetes makes it tempting to speculate that we can translate what we know about human INS pharmacology to corals, coining a new field of “coral pharmacology” which opens the door to thinking about drug discovery and treatments for corals. While at present we do not know if and how we can deliver medicines to corals in the vast ocean in practical terms, ideas include coating surfaces on which coral larvae settle, or feeding corals or dispersing such compounds into the ocean in proximity to coral reefs. This is not unthinkable given that there are many known examples of small molecules secreted into the ocean used in communication between different inhabitants of a reef, e.g., for attracting fish to anemones (Kamio & Derby, 2017; Saha et al., 2019; Kamio, Yambe & Fusetani, 2022; Morgan et al., 2022) or to mediate biological interactions with surfaces during settling of coral larvae (Petersen et al., 2023). Nanocarriers could also assist with this purpose (Roger et al., 2023). Given the fluid nature of the ocean environment, such small molecules can be dispersed easily and thus the environmental impact of treatment of corals with small molecule “coral drugs” will need to be carefully addressed. Nonetheless, the idea of “coral pharmacology” may open new avenues to think about how to tackle the coral bleaching crisis. How might coral drug discovery look like? The high quality of sequence alignments of INS and INSR with respective coral homologues shown in Table 3 provides an opportunity to exploit the large amount of INSR structural data that has become available recently (McKern et al., 2006; Menting et al., 2013; Gutmann et al., 2018; Weis et al., 2018), especially due to the advances in cryoelectron microscopy (Uchikawa et al., 2019). Shown in Fig. 3 are homology models for the various domains in human used to predict coral INSR using the sequences shown in Table 3B. These models open the door to the first step of future exploration of potential drug targets in coral, exemplified here by the INSR as a drug target, extrapolated from its role as a human drug target (Kumar, Vizgaudis & Klein-Seetharaman, 2021). The concept of “coral pharmacology” aims to develop pharmacological approaches towards potentially treating corals who have been harmed by human activities. Using membrane receptors as a proof of concept, we developed a pipeline for establishing the functional similarities between human and coral membrane receptor signaling systems (Kumar et al., 2023). This pipeline extends to the INS-INSR pair and its related signaling pathway (Fig. 2). Given the role of INS signaling for regulation of nutrient concentration in humans, we surmise that the coral homologues will likely carry out a similar function in corals. This suggests that early metazoans such as corals use the INS system despite their simple organization. This may have major implications for coral bleaching and the communication across cnidarian host and symbiotic algae. Transcriptomic analysis has revealed that INS signaling is clearly affected by the establishment of symbiosis between cnidarian host animals and algal symbionts (Yuyama et al., 2018). Given that one of the major benefits of symbiosis is the delivery of sugars obtained through photosynthesis of the algae to the host, we can expect that the role of INS signaling is analogous in corals to that in humans, despite their evolutionary distance. It is tempting to speculate that under high light conditions, when the algae synthesize excess sugars, that the cnidarian host may experience INS resistance, a hypothesis that remains to be validated experimentally. By inference, pharmacological treatment of INS resistance may allow coral rescue. We have already shown that human INS does have an effect on corals (see above), and in fact is cytotoxic (Roger et al., 2022). The presence of receptors such as IR (described here) and GPCR (described in Kumar et al., 2023) suggests that many other functions of corals could potentially be targeted by pharmacological means to help prevent their extinction predicted under current climate trajectories.