{"paper_id":"febb25a9-3c32-41e5-9091-4085ecd6fe4d","body_text":"Congenital heart disease comprises defects in cardiac morphogenesis; it is the most common type of birth defect [ 1 ], with its incidence globally calculated at 8–9.5 per 1000 live births, though regional variability can exist [ 2 ]. Defects are often classified based on the presence/absence of cyanosis due to right-to-left shunting. Cyanotic defects include Tetralogy of Fallot, transposition of the great arteries, and total anomalous pulmonary venous return [ 3 , 4 , 5 , 6 ]. Non-cyanotic defects, on the other hand, can include both obstructive disease, including aortic stenosis and coarctation of the aorta [ 7 , 8 ], as well as left-to-right shunts. The latter comprises defects such as atrial and ventricular septal defects, patent ductus arteriosus, and atrioventricular canal defect [ 9 , 10 , 11 , 12 ].\nMany defects are often diagnosed before or right after birth. Some, including Tetralogy of Fallot [ 13 ], may require surgical correction to maintain proper communication between right and left circulations [ 14 ]. Regarding causality, different factors can be identified, including inherited or de novo genetic mutations [ 15 , 16 , 17 , 18 ] and larger chromosomal aberrations such as aneuploidy and polyploidy syndromes [ 19 , 20 , 21 ]. For example, aneuploidy syndromes such as Turner syndrome or various Trisomy Syndromes can be associated with cardiac septum and outflow tract defects [ 22 , 23 ]. Aneuploidy syndromes can also often present with mosaicism; in this case, a set of cells exhibits chromosomal aneuploidy, while the remaining cells possess a normal karyotype [ 24 ]. Mosaic aneuploidy syndromes often result in longer survival with less severe characteristics [ 24 , 25 , 26 ] compared to their non-mosaic counterparts, as with, for example, mosaic Trisomy 13 [ 24 ]. They may also present with a normal phenotype altogether, as is the case with mosaic Trisomy 22 [ 27 ]. Subchromosomal aberrations, including 22q11.2DS deletion syndrome, also known as DiGeorge syndrome, can present with cardiac defects as well [ 20 , 28 ].\nSince genetic causes are identifiable in about 35% of patients with the disease [ 29 ], the employment of models used to discover associations between individual candidate genes and their phenotype in simple organisms is indispensable [ 30 ]. Compared to other in vivo models, particularly mammalian models, which can often be more expensive and difficult to deploy for large-scale candidate gene screening due to ethical reasons [ 30 ], simple invertebrate models offer a better and often cheaper alternative. Specific genes for which a cardiovascular association has been found can then be evaluated further in more complex models [ 30 , 31 ]. The evolutionary conservation in the genetic networks that direct cardiac development allows for the use of distantly related animal species to study cardiac development [ 32 ]; however, the greater the evolutionary gap between organisms, the greater the difference in form and function, as opposed to species that relatively recently diverged [ 33 ]. Thus, while distantly related species can be used to evaluate the function of orthologous genes, more closely related species are chosen when recapitulation of relevant aspects related to form and function is required as well [ 1 ]. On the other hand, regarding  D. melanogaster  and other similar invertebrate organisms, they are often characterized not only by relative simplicity in form, at least compared to mammalian vertebrates/mammals, but can also exhibit simpler genetic networks. Around 13.601  D. melanogaster  genes have been identified [ 34 ], with ~70 genes associated with congenital heart disease [ 35 ]. The simplicity in the cardiac gene networks governing  D. melanogaster  heart formation better facilitates the identification of single gene effects during the progression of cardiac morphogenesis; in addition, they can be used to discern whether a particular gene or locus may have any effect on heart development at all, if disrupted, in the case of candidate gene evaluations [ 36 ].\nThe purpose of this comprehensive review is to group and summarize studies that evaluate the effects of gene mutations on  D. melanogaster  cardiac development. A review of relevant literature ranging from 1990 to 2025 (present) has been carried out. Studies employing  D. melanogaster  as a screening tool, when these include genes affecting cardiac development, are also included. Overall, target genes have been grouped into categories based on their function; this includes genes comprising the cardiac regulatory networks that orchestrate heart development; genes involved in cellular metabolism and protein synthesis/trafficking; genes and factors involved in the migration and alignment of cardiac progenitors during assembly of the cardiac tube/dorsal vessel; genes involved in the establishment of segmentation, polarity, and other relevant signaling pathways; homeotic genes involved in the establishment of the animal body plan; and finally, genes encoding for histone modifiers and other chromatin-regulating enzymes with an effect on heart development. Orthologs of these genes in vertebrates/mammals, along with their effects on cardiac development. Finally, any associations between gene orthologs and human congenital heart defects have been incorporated as well.\nIn general, this text thus aims to compare the ancient cardiac gene network, along with any additional genes involved in heart development in  D. melanogaster , to the more developmentally complex vertebrate cardiac gene networks. Similarly, phenotypes resulting from perturbations in these genes in  D. melanogaster  will be compared to the corresponding perturbations seen when orthologs of these same genes are disrupted in vertebrates/humans.\n\nArthropod body plans exhibit a segmented architecture, which consists of a variable number of similar units, known as segments, along the anteroposterior axis [ 37 ]. Regarding the cardiovascular system, all arthropods generally possess an open circulatory system composed of a dorsally located contractile heart, arteries, and a hemocoel. Hemocoels are body spaces through which hemolymph circulates freely [ 38 , 39 ]. The contractile heart, in particular, is a solenoid structure composed of a posterior cardiac chamber and an anteriorly placed aorta from which it is separated by an aortic valve; this structure is known as the dorsal vessel. In the dorsal vessel, the aorta is situated in the thoracic segments, while the cardiac chamber is found within the abdominal segments. As with all arthropods, the circulatory system is open; hemolymph enters the heart via specialized ostia functioning as inflow tracts, flows through the dorsal vessel in a retrograde/anterograde manner depending on the conditions [ 40 ], and is finally pumped toward the body cavity. In this space, the hemocoel, the hemolymph flows and intermixes with both interstitial and extracellular fluids [ 30 , 41 ]. As with most arthropods [ 32 ], these systems are low-hydrostatic-pressure systems since the circulating hemolymph is not separated from interstitial fluids and thus not responsible for tissue oxygenation [ 42 ]. This contrasts with circulatory systems in mammalian vertebrates/humans, where interstitial fluids and blood are clearly separated, flowing through high-pressure circuits [ 43 , 44 ].\nIn  D. melanogaster , molecular distinction between the anterior dorsal vessel (aorta) and the posterior dorsal vessel (aorta, heart chamber), as well as variation in genetic expression along the anteroposterior axis, can be attributed to differential  Hox  gene expression [ 45 ]. In the embryonic/larval heart, the anterior dorsal vessel spans both thoracic and abdominal segments (T3–A4), while the posterior dorsal vessel is found in abdominal segments (A4–A7) [ 46 ]. In the adult fly, on the other hand, the anterior dorsal vessel is limited within thoracic segments (T1–T3), while the posterior dorsal vessel spans both thoracic and abdominal segments (T3–A5) [ 47 ]. Histolysis of abdominal segments A6 and A7 has already occurred during metamorphosis, altering the overall length of the dorsal vessel in the adult [ 46 , 47 ]. Regarding cellular composition, it is composed of a single cardiac cell layer derived from cardioblasts, surrounded by non-contractile pericardial cells [ 48 ]. Some of these pericardial cell populations, including odd-skipped ( Odd )+ pericardial cells (OPCs), will eventually fulfill nephrocytic functions in the adult fly, participating in hemolymph filtration [ 49 , 50 ]. Since these cell types are closely related to hemopoietic lymph gland cells, they also participate in immune cell responses [ 51 ]. During embryonic life, there are three paired openings or ostia along the dorsal vessel, located at the boundaries of segments A5/A6, A6/A7, and A7/A8. They are abutted by specialized ostial cardiac cells and form passive inflow tracts [ 40 , 52 , 53 ]. No such openings can be found in the anterior dorsal vessel [ 47 ]. Intracardiac valves can also be identified, facilitating the unidirectional flow of hemolymph with each heartbeat. Intracardiac valves are composed of valve cells, specialized cardiac cells originating from tinman ( tin )+ cardiac cells, which are characterized by their large intracellular volume and unique sarcomere arrangement [ 54 , 55 ].\nIn the anterior-most dorsal vessel, an outflow tract can be observed, formed by specialized Even-skipped ( Eve )+ tinman+ ( tin )+ pericardial cell (EPC) populations; these cells assemble into an outflow hanging structure, which attaches the anterior-most dorsal vessel cardioblasts to the cuticle [ 56 ] via further association with ladybird ( lb ) +  heart-anchoring cells derived from the dorsal epidermis, generating a funnel-shaped tip [ 56 ]. Additional structures observed in this area include a pair of cardiac outflow muscles on either side, derived from the pharyngeal mesoderm [ 57 ]. EPC populations in the thoracic segments, under the influence of  Hox  genes such as Antennapedia ( Antp ), differentiate into wing heart pericardial cells (WHPCs), while EPC groups not subjected to  Hox  gene expression comprise an outflow hanging structure previously described in [ 56 ]. Additional pericardial cell groups include end-of-the-line pericardial cells (ELPCs) [ 51 , 58 ], found only in abdominal segment A7 [ 51 ] ( Supplementary Table S1 ).\nThe dorsal vessel enlarges mainly via cellular growth, alongside the enlarging animal body, during larval stages L1, L2, and L3; loss of some mononucleated pericardial cells occurs during this period [ 59 , 60 ]. Transition from pupa to adult form, also known as metamorphosis, is associated with a reduction in overall cardiac cell numbers (~104 to 84) and an increase in ostia pairs from three to five. There may also be evidence for the existence of a posteriorly located terminal opening in the adult, allowing for the retrograde flow of hemolymph during episodes of heartbeat reversal [ 40 , 52 , 61 ]. During the transition from larva to adult, the number of intracardiac valves also increases from one to three. Eventually, the adult structure is a four-chambered heart with an anteriorly located aorta and heart chambers arranged in series [ 62 ] ( Figure 1 ).\nA ventral longitudinal muscle layer inferior to the cardiac tube, derived from larval alary muscles, also emerges during metamorphosis. This layer morphologically separates the pericardial and abdominal cavities [ 63 ]. Alary muscles attach to the dorsal vessel indirectly via extracellular matrix proteins such as pericardin [ 64 ] and Viking, both of which are collagen IV-like proteins [ 65 ]. This allows the dorsal vessel to anchor to the outer epidermis and, as such, stabilize it within the body cavity of the fly [ 66 ]. Cardiac chamber remodeling also occurs during metamorphosis; posterior-most segments undergo histolysis, as has already been stated [ 46 ], and new neural connections are formed [ 41 ]. Accessory pulsatile organs facilitating hemolymph flow toward the limbs, wings, and antennae also emerge during this transitional period [ 52 , 67 ].\nComparison of  D. melanogaster  and  H. sapiens  embryonic heart structures, with cells carrying out similar functions highlighted. General cell group/tissue types found in each organism are also noted. Structures compared represent late-stage embryonic forms, and as such, the  D. melanogaster  dorsal vessel shown possesses only one aortic valve and three ostia on either side of the heart chamber. The single heart tube stage of cardiac development found in  H. sapiens  is shown to facilitate comparison, as later human embryonic stages diverge into more species-specific structures. For a complete list of all gene abbreviations, see  Supplementary Table S9 . CM, cardiomyocyte; CTPC, cut+/tinman+ pericardial cell; DV, dorsal vessel; ELPC, end-of-the-line pericardial cell; EPC, even-skipped+/tinman+ pericardial cell; OPC, odd-skipped+ pericardial cell; PC, pericardial cell; WHPC, wing heart pericardial cell; gCB, generic cardioblast; oCB, ostial cardioblast. [ 32 , 56 , 62 , 68 , 69 , 70 ]. Created in BioRender. Stougiannou, T. (2025)  https://BioRender.com/wpq4alf .\n\nD. melanogaster  embryos possess centrolecithal eggs that undergo superficial cleavage due to the presence of a centrally located mass of yolk [ 71 , 72 ]. Eventually, the blastoderm emerges during early development, comprising peripherally localized cells arranged around a central yolk; this structure is considered homologous to the blastula observed during embryonic development in other species [ 73 ]. Invagination of cells from this peripheral layer toward the center commences during gastrulation, during which stage, development of the cardiovascular system also commences [ 51 , 74 ]. This is facilitated by a complex interplay of transcription factors that comprise the core cardiac gene regulatory network, which directs specification of the cardiac mesoderm, leading to the eventual derivation of the various cardiac cell groups, as illustrated in  Figure 2 a, b, and described in detail in the associated figure captions ( Figure 2 ). Additional details regarding the function of each gene described in this section, as well as all other sections of this text, along with information about their mammalian/human orthologs, can be found in  Supplementary Table S8 . As cardioblasts and pericardial cells diversify, they also undergo defined movements in space; though cardiac cell group aggregations can initially be seen bilaterally, these will eventually converge toward the center and align with those of the contralateral side. This migrational movement, observed as cardiac leading-edge activity, parallels the movement of the overlying ectodermal epithelium converging toward the midline as well. This movement is regulated by distinct signaling pathways, including FGF and Slit/Robo signaling [ 75 ], explored in more detail in  Section 4 . Eventually, a solenoid structure positioned in the dorsal midline of the animal body will form [ 32 ].\n\nPrimitive organs resembling a heart first appear in the tree of life about 500 million years ago; earlier hearts are generally simpler in structure, with more complex cardiac systems eventually appearing, reflecting specific adaptations depending on species and environment [ 36 ]. The appearance of bilaterians is associated with the emergence of the mesoderm [ 91 ], and with it, the cardiac mesoderm and the heart [ 92 ]. Development of the heart in these organisms is dependent upon a genetic regulatory network, although genes associated with this network, such as  Mef - 2 , have been found in species without bilateral symmetry as well. This only highlights the primordial origins of this regulatory network [ 93 ]. Simple tubular heart structures are usually found in animals that have evolved earlier in the phylogenetic tree, including invertebrates and arthropods. Tubular hearts are also observed during the early stages of vertebrate embryonic development [ 92 ]. The occurrence of morphologically similar structures in developmental stages across different species represents an example of convergent evolution, often reflected in the model of the “developmental hourglass”. While initial developmental stages differ, many animals eventually reach a period that is morphologically conserved among different phyla, also known as the phylotypic period. Once this period is complete, development once again becomes less conserved among different species, leading to the emergence of different animal forms. The presence of a phylotypic period represents the need to generate a viable animal body plan, a process directed by gene regulatory networks conserved among different species [ 94 ]. Accumulation of genetic changes, such as gene duplications, has contributed to genomic evolution and the emergence of more complex genetic networks, in turn orchestrating the development of more complex vertebrate cardiac systems [ 33 , 36 ].\nRelationships between organisms can be defined based on the presence of the most recent common ancestor; this allows for the correct identification of genealogical relationships. These relationships in the evolutionary history (phylogeny) of an organism [ 95 ] can be summarized with phylogenetic trees; in this manner, a graphical depiction of the evolutionary history of a particular species in relation to other closely related species can be carried out [ 96 ]. The phylogeny of a species [ 96 ] can differ from the genealogies of specific genes (gene trees) within the species [ 95 ]. Often, gene tree topologies, even when part of the same species, may differ, exhibiting a topology different from that of the organism where they are found [ 97 ]. This may reflect alterations in gene sequences due to various events, including horizontal gene transfer, which allows genetic material to flow between organisms in a manner other than vertical transfer; the latter usually occurs in the context of traditional relationships of descent [ 98 ]. In eukaryotic organisms characterized by a membrane-enveloped nucleus and germ-line segregation of genetic material, however, the rate of non-vertical gene transfer is greatly diminished compared to Bacteria or Archaea [ 99 ], though this topic is still under debate [ 98 ].\nHomology, a term used for over 150 years, often with variations in definition, has come to describe the degree of similarity derived from an evolutionary relationship, owing to the presence of a common ancestor [ 100 , 101 ]. It is important to note that similarities alone are not enough to characterize homology [ 102 ], as similarities between organisms can be attributed to events such as convergent or parallel evolution [ 103 ], allowing similar features or characteristics to develop in non-closely related animals [ 104 ]. The concept of homology can be applied to genes as well; though a particular gene possesses a specific function in the ancestral species, over time, with the accumulation of genetic changes and eventual evolutionary divergence, homologous genes emerge across different species, exhibiting additional functions or new functions compared to the ancestral gene [ 100 ]. While homologous genes can be characterized by percent sequence similarity, sequence similarity itself is not a defining characteristic of homology, as once again, an evolutionary relationship must be present [ 105 ].\nHomology can have many different forms, including partial homology [ 100 ], paralogy [ 106 ], xenology [ 107 ], and orthology [ 106 , 108 ]. Terms such as xenology and partial homology describe homologous characters due to horizontal gene transfer (xenology) [ 107 ], as well as differing homology relationships occurring between different areas within the same gene, a phenomenon often due to genetic recombination or other events affecting gene subregions (partial homology) [ 100 ]. Paralogous genes result from gene duplication and can often be found in the same species, as is the case with hemoglobin and myoglobin in  H. sapiens  [ 109 ] or the neuromancer 1 ( nmr1 ) ( H15 )/neuromancer 2 ( nmr2 ) ( mid ) and dorsocross  1 / 2 / 3  ( doc1 ,  doc2 ,  doc3 ) genes in  D. melanogaster  [ 86 , 108 ]. Gene duplication has contributed to the emergence of more complex traits along the evolutionary timeline; for example, duplications in ancient Homeobox ( Hox ) gene clusters have been associated with the emergence of complex cardiac forms and cardiac chambers [ 110 , 111 ].\nOrthologous genes are homologous genes that have emerged due to speciation events that led to the splitting of the evolutionary lineage [ 100 ]; they often retain similar or equivalent functions among different organisms [ 108 ]. They can be classified based on the number of orthologs that exist within each compared species, a phenomenon described as cardinality; 1-to-1 (1:1) pairwise orthology refers to the presence of one orthologous gene in each species; 1-to-N (1:N) refers to the presence of more than one ortholog in the other species, possibly due to duplication events in a previous ancestor and finally; and many-to-many (N:N) orthology refers to many orthologs found in both species under comparison [ 112 ]. To predict orthology between genes across different species, specialized tools can be used, including the Drosophila RNAi Screening Center Integrative Ortholog Prediction Tool (DIOPT) ( https://www.flyrnai.org/cgi-bin/DRSC_orthologs.pl , accessed on 15 July 2025), which has also been used to describe orthologous relationships in this review as well. Through this tool, orthology predictions from other similar tools (Homologene, OMA, Isobase, Phylome, RoundUp, InParanoid, orthoMCL, TreeFam, and Ensembl Compara) can be integrated into one score, reflecting the number of tools that support an orthologous relationship [ 113 ].\nExamples of 1:N orthology include the  tin : NKX2  pairing; the  NKX2  gene family comprises the genes  NKX2.1 ,  NKX2.2 ,  NKX2.3 ,  NKX2.4 ,  NKX2.5 , and  NKX2.6  [ 114 ], all considered paralogous genes originating from an ancient duplication event in vertebrates [ 115 ]. However,  NKX2  family genes are orthologous not only with  tin , but also with genes such as scarecrow ( scro ) and ventral nervous system defective ( vnd ), which are involved in  D. melanogaster  neurogenesis [ 116 , 117 ], as well as bagpipe  (bap) , which is involved in development of  D. melanogaster  midgut musculature [ 118 ]; this further highlights the complexity of orthologous gene relationships [ 108 ]. Though  tin  possesses an orthologous relationship with all  NK2  genes, the orthologous relationship between  tin : NKX2.5 , in particular, generates the highest DIOPT score [ 113 ]. Furthermore, while  tin  and the  NKX2  genes are all orthologous, their functional contribution to the developing cardiac system is different. Thus, though the fly  tin  is important for the emergence and specification of cardiac mesoderm, in vertebrates, the ortholog  NKX2.5  is not required for the very early stages of cardiac mesoderm specification, though it is, nevertheless, indispensable for physiological heart development [ 119 , 120 ]. In addition, there is functional redundancy between different  NKX2  paralogues, for example, with  NKX2.3  and  NKX2.5 , during cardiac development in vertebrates [ 121 ]. Despite these orthologous relationships,  NKX2  genes cannot generally substitute for  tin  with regard to its cardiogenic function in  D. melanogaster , with the exception of zebrafish  NKX2.3;  in return, neither can  tin  substitute for  NKX2.5  in associated assays [ 122 ]. This showcases the functional divergence between these two gene groups in different species; alternatively, these observations may also reflect the loss of some cardiogenic functions that could have been present in the ancestral gene [ 122 ].\nDoc , another gene involved in early cardiac development, comprises three paralogues,  doc1 ,  doc2 , and  doc3  [ 123 ]; these are all orthologous to the vertebrate T-box transcription factor genes  TBX2 ,  TBX3 , and  TBX6 , an example of N:N orthology [ 123 ].  TBX6  is involved in mesoderm and paraxial mesoderm induction, as well as indirect induction of cardiovascular cell lineages in pluripotent stem cell lines in vitro [ 124 ].  TBX2 ,  TBX3 ,  TBX4 , and  TBX5  in vertebrates are all orthologous to the  D. melanogaster  gene  bifid  (optomotor blind [ omb ]); in associated studies, removal or deficiency of  bifid  results in lethality and various defects in eye [ 125 ] and wing development [ 126 ]. Interestingly, however, the function of  bifid  in the fly cannot be rescued with  TBX2 , another example of the functional divergence observed between gene orthologs across different species. However, in the same experiment,  D. melanogaster  bifid protein administration also does not restore function, which may point to sensitivity in protein dosage as the explanation for the lack of phenotype rescue [ 127 ]. Additional genes identified in both  D. melanogaster  and  H. sapiens  include  Mef2 , with  MEF2  in vertebrates comprising the paralogues  MEF2A  through  MEF2D  [ 90 ]. Another example of gene paralogues in  D. melanogaster  includes the bric-a-brac ( bab ) locus with paralogues  bab1  and  bab2  [ 128 ];  bab2 , in particular, is implicated in the diversification of cardioblasts/pericardial cell groups [ 129 , 130 ], while both  bab1  and  bab2  function synergistically in imaginal disk development [ 128 ].\nHomology can be evident with morphological stages during development as well. An example of this is the gastrula, an embryonic stage identifiable across many different species, including  D. melanogaster  and  H. sapiens , even though the developmental processes that lead to its emergence are different. In flies, for example, cells from the single-layered blastoderm surrounding the central yolk cell invaginate, a process described as epithelial folding. Eventually, they undergo changes that result in the generation of a three-dimensional embryonic gastrula [ 131 ]. In mammals and humans, on the other hand, the amount of yolk present during embryonic development is considerably less, and cells undergo more complex movements. These include ingression or cell detachment from an epithelial layer leading to individual movement [ 132 ]; involution or cell rolling against a surface [ 133 ]; and finally, convergent extension or convergence and extension, a method of cellular rearrangement aimed at deriving a specific shape via narrowing along one dimension and extending along another [ 134 ]. Thus, it is evident that while different mechanisms are active in each species, the form produced can be identified as a gastrula. The gastrula is found across many different species, from invertebrates such as  D. melanogaster , anamniotes, and amphibians to amniotes such as reptiles, birds, and all the mammalian groups [ 133 ].\nDevelopmental homology is also evident between the two species due to conservation in the genetic programs that culminate in the establishment of the body axes, particularly the anteroposterior and dorsoventral axes. This could be attributed to the presence of a common ancestor in both  D. melanogaster  and  H. sapiens  predating deuterostome and polyphyletic protostome divergence. Deuterostomes include the Chordata, which comprise mammalian species such as  H. sapiens , as well as simpler organisms such as echinoderms and Hemichordata [ 135 ]; polyphyletic protostomes, on the other hand, include phyla such as arthropods, including  D. melanogaster  [ 136 ]. Though a dorsoventral axis exists in both species ( D. melanogaster  and  H. sapiens ), it is inverted between the two [ 137 ], with the nerve chord located ventrally in invertebrates [ 138 ] and dorsally in vertebrates [ 139 ]. This difference has been attributed to axis inversion or, alternatively, the presence of a common ancestor with diffuse dorsoventral axis organization [ 140 , 141 ]. As a result of this difference in orientation, cardiac structures in  D. melanogaster  are located dorsally, as opposed to their ventral localization in many mammalian vertebrates, including humans [ 142 ].\nRegarding anatomical characteristics, there is no homology between  D. melanogaster  and  H. sapiens ; in the former, cardiac structure is tubular, with chambers arranged in series in the dorsal section of the animal body. In the latter, cardiac structure is considerably more complex due to additional events of rightward looping driven by left–right asymmetry [ 36 , 143 ], chamber/endocardial cushion development [ 144 , 145 ], and septae and valve formation [ 146 , 147 ], which occur after the single heart tube stage during embryonic development. Regarding physiological characteristics, however, the two hearts are considered homologous by some [ 148 , 149 ]; cardiac flow in  D. melanogaster  is pulsatile with cycle-dependent hemolymph transport, while characteristics such as flow velocities within the heart chambers and across the aorta, cardiac output, and mechanics relating to the function of the heart as a pump are comparable as well [ 148 , 149 ].\nThe homology in physiological function, coupled with the conservation of gene regulatory networks that govern cardiac development in both  D. melanogaster  and mammalian vertebrates [ 150 ], allows for the use of the former in genetic cardiac disease modeling [ 84 ]. In these models, mutations in genes that drive developmental networks or generally affect the process of cardiac development can result in congenital heart disease [ 151 ]. In addition, the cardiac gene regulatory network is simpler in  D. melanogaster  due to fewer genetic redundancies [ 36 , 152 ]. In short, a subset of cardiac transcription factor genes becomes initially active, eventually activating other similar genes, as well as genes implicated in cardiac structure/function and associated signaling pathways. Transcription factors carry out this function by binding to cis-regulatory elements such as promoters, enhancers [ 153 ], and downstream promoter elements [ 154 ]. Mutations in these early factors can be easily tracked and assessed in  D. melanogaster , as the animal does not depend on a cardiovascular system for oxygen transport; this facilitates the evaluation of phenotypes and the contribution of candidate genes in these phenotypes that would otherwise result in early embryonic lethality in vertebrate models [ 32 , 142 ].\nVarying approaches can be used to investigate gene function in the  D. melanogaster  system, including loss-of-function studies [ 142 ], gene knockout, and gene knockdown. While knockdown comprises transcriptional/translational suppression in gene expression, causing reduced protein production without genome modification, knockout involves ablation of genes or larger genomic loci altogether [ 155 ]. Knockdown can include tools such as ribonucleic acid interference (RNAi) [ 156 ] and morpholino antisense nucleotide knockdown [ 157 ]. These, however, can also be associated with off-target effects [ 158 ] or phenotypes originating due to the toxicity of the products themselves; furthermore, they usually only lead to a partial loss-of-function phenotype [ 155 ]. Knockout, on the other hand, can include targeted nuclease-based approaches, such as ZFN [ 159 ], TALEN [ 160 ], and CRISPR/Cas9 [ 161 ]. While these systems can also be associated with off-target effects, these can be tackled with further refinements, including the use of more than one nuclease to achieve cleavage (Cas9 nickase [Cas9n]), as well as the refinement of short guiding RNA sequences [ 162 , 163 ]. Additional methods involving alteration in gene structure include in vitro mutagenesis; the gene products generated in these cases exhibit a change in function or reduced function [ 164 ]. In vitro mutagenesis can involve site-directed mutagenesis, usually employed in cases where the wild-type target sequence is known and involves the synthesis of an oligonucleotide primer. Changes induced in this manner involve substitutions or deletions [ 165 ]. Others, such as gene disruption mutagenesis or knockout, can involve DNA insertion and recombination techniques to abolish gene function, including the highly specific recombination knockout techniques mentioned previously, as well as techniques involving DNA-alkylating agents or DNA insertion using transposons, both of which lack target specificity [ 166 ]. Mutations that result in the complete absence of a gene product and its associated function are often called null or amorphic mutations [ 155 ]. Often, a continuous region within a chromosome can be absent or deleted, affecting several genomic loci, described as a deficiency. Deficiencies can be used to evaluate phenotype severity associated with a particular allele, constituting definitive null alleles [ 167 ].\nMethods to increase expression levels in a gene of interest can also be applied, as this too can induce perturbation in cellular and molecular processes. While the term “overexpression” is often used interchangeably with terms such as “misexpression” or “ectopic expression”, the latter two are often used in studies involving metazoan models to describe the expression of a particular gene within a cell group, tissue, or developmental time frame that it is not normally found in [ 168 ]. However, many studies with metazoans, including  D. melanogaster , use the term “overexpression” for this purpose as well [ 63 , 169 , 170 , 171 ]. Mechanisms used to induce gene overexpression can include mutations in the enhancer area of a gene, leading to increased gene expression [ 168 ]. Additionally, GAL4 systems can be employed, comprising a transcriptional activator isolated from yeast and modified to drive expression in a tissue-specific manner, along with the gene of interest or a transgene whose expression is controlled by an upstream activation sequence (UAS G ), bound by GAL4. These systems have been very commonly employed in  D. melanogaster  studies to evaluate gene function and associated phenotypes [ 172 , 173 ]. Tools culminating in overexpression can also be employed to increase the expression of mutated genes that, when produced, still retain the ability to interfere with the function of other proteins, including the function of a wild-type protein. The phenotype produced with such mutations is usually dominant, hence the term “Dominant Negative” mutation [ 168 ], considered a non-loss-of-function effect [ 174 ].\nAdditional strategies for inducing gene overexpression include the use of heat-shock systems, comprising transgenes that include the gene of interest, along with a promoter derived from a heat-shock protein. Expression in these cases is dependent on applied temperatures [ 175 , 176 ]. Temperature-sensitive mutations involving genes that encode for a functional product at the permissive (low) temperatures and a non-functional gene product at non-permissive (high) temperatures can also be applied. Temperature instability in these cases is most commonly due to a thermolabile protein product, which can become unstable or exhibit defects in folding under non-permissive temperatures. Temperature-sensitive mutations are useful for inducing changes in gene expression, including loss-of-function, at desired timepoints during an experiment [ 177 ]. These are usually classified as conditional mutations [ 178 ].\nMany mutations contributing to congenital heart defects, both as de novo mutations and inherited or syndromic mutations, can be attributed to disruptions in cardiac transcription factor genes. As mentioned earlier and throughout this text, these genes encode for transcription factors that will, in turn, regulate the expression of similar or other gene types, all collectively involved in cardiac morphogenesis. Early factors activated during morphogenesis include  msh-2 ,  tin , and tailup ( tup ). The gene  msh-2  encodes a homeobox transcription factor implicated in early mesoderm development; as a result,  msh-2  loss-of-function studies exhibit a complete absence of the dorsal vessel and visceral muscle in  D. melanogaster.  In other cases, although somatic muscles can be identified, they are often abnormal [ 179 ]. In murine models, mutations in the  msh-2  ortholog  MSX-2  affect cardiac mesoderm precursors that will eventually assemble into the outflow tract, leading to defects in morphogenetic rotational movements in the truncus arteriosus [ 180 ].\nAmong the earliest cardiac transcription factors identified in  D. melanogaster  experiments,  tin  is involved in dorsal mesoderm and cardiac mesoderm specification, as well as cardiac development and cardioblast diversification [ 181 , 182 , 183 ]. Null mutations involving  tin  usually result in the complete absence of cardiac/dorsal somatic muscle, with disruption in somatic muscle arrangement in each segment [ 181 ]. Mutations also affect the expression of another early transcription factor gene,  doc , from Stage 12 of embryonic development and onward [ 182 ]. The  tin  gene also possesses a downstream core promoter element, which, when affected by site-directed mutagenesis, exhibits reduced expression; additional targets also affected include  doc ,  svp ,  Mef2 , and  Odd.  This eventually leads to specification of fewer cardioblasts with functional defects in the dorsal vessel [ 161 ]. In line with similar experiments [ 181 , 182 , 183 ], somatic and visceral muscles are not as affected [ 161 ].\nNKX2.5  is also indispensable for cardiac development in vertebrate models. However,  NKX2.5  is not necessary for the initial stages of cardiac mesoderm specification [ 161 , 184 ], even though ablation of  NKX2.5  in early developmental stages still leads to embryonic lethality and cardiac defects in murine models [ 185 ].  NKX2.5  ablation in later stages affects the development of the ventricles [ 186 ], ventricular septum, and cardiac conduction system and can lead to arrhythmias [ 185 ]. Various  NKX2.5  variants have also been associated with atrial septal defects and hypoplastic left heart syndrome in humans [ 187 ]. Furthermore, though downstream promoter region motifs have been identified in  NKX2.5 , their effect on  NKX2.5  levels is yet to be determined [ 161 ]. The effects of  NKX2.5  variants, including the variant K158N ( D. melanogaster  ortholog  R321N ), have been examined as well. In individual flies, it is associated with defects in differentiation, although initial cardiac specification occurs normally [ 183 ]. Through the  D. melanogaster  model, this variant has been associated with a pathophysiologic mechanism involving disruption between DNA and cofactor binding [ 183 ]. Thus, although the variant is still of unknown clinical significance, the phenotypes demonstrated both in vitro and in vivo may point to some effects that may also be present in vertebrate/human populations as well, requiring further study [ 183 ].\nAdditional transcription factor genes whose perturbation leads to visible cardiac defects in the  D. melanogaster  model include  svp  [ 188 ]; the paralogues  H15  ( nmr1 ) and  mid  ( nmr2 ) [ 113 ]; the paralogues  doc1 ,  doc2 , and  doc3  [ 78 ];  Eve  [ 189 ];  tup  [ 80 ];  Hand  [ 190 , 191 ]; and  D-mef2  [ 113 ]. Loss-of-function mutations in  tup  result in a hypoplastic dorsal vessel with severe morphological defects, including gaps and distortion in the structure, along with a reduction in cardioblast populations [ 80 , 81 ]. This gene is also expressed in valve cells, alary muscles, and thoracic–alary-related muscles; mutations in these cases usually affect the myofibrillar organization of valve tissue [ 62 , 192 ]. The vertebrate orthologs  ISL1  and  ISL2  are also similarly required during early development as part of the early cardiac transcription factor network. They are involved in the regulation of second heart field progenitor groups as these emerge and expand, contributing to the development of the outflow tract [ 193 ].  ISL1  further contributes to the development of the atrial septum, the sinoatrial and atrioventricular nodes [ 194 ], and the endothelial and vascular smooth muscle cell groups [ 195 ]. In vertebrate models,  ISL1  knockout in mice has been associated with the complete absence of the atria, the right ventricle, and the outflow tract [ 193 ], while defects in cardiac looping and development of the arterial pole have been described with  ISL2  mutations in zebrafish [ 196 ]. In human genetic studies,  ISL1  variants and mutations have been described in cases of ventricular septal defects and double outlet right ventricles [ 197 ].\nThe  TBX20  transcription factors  H15  and  mid  also participate in early cardiac development, with mutations in  D. melanogaster  affecting the expression of other transcription factors. More specifically,  H15 / mid  mutations are associated with a reduction in  tin ; upregulation in  Eve  and  Odd  expression; and finally, effects on cardioblast/pericardial cell diversification divisions, cardioblast alignment, and the general spatial arrangement of the cells in the midline [ 113 , 198 , 199 ]. If mutations are reproduced in adult animals, these usually bring about functional disruption in cardiac structure/myofibrillar arrangement [ 200 ]. In  D. melanogaster  studies, this is often measured as the effect on cardiac function (heart failure) induced by a stressor, in this case, in the form of electrical pacing [ 201 ]. Similar to the interaction between  H15 / mid  and  tin ,  TBX20  interacts with  NKX2.5 , revealing a genetic association that has persisted throughout multiple lineage diversifications and across different species [ 202 ].  TBX20  similarly interacts with  GATA4 / 5  and  TBX5  [ 203 ]. In vertebrates,  TBX20  contributes to the development of the atrioventricular canal and ventricular cells, while  TBX20  knockdown in murine models is associated with hypoplasia of the right ventricle and outflow tract, valvular defects, and outflow tract septation anomalies [ 204 ].  TBX20  mutations and variants have also been associated with heart defects in humans, including septal defects, double outlet right ventricle [ 205 ], congenital mitral valve prolapse/regurgitation, congenital defects in the conduction system [ 203 ], bicuspid aortic valves, and hypoplastic left heart syndrome [ 205 ].\nAs previously mentioned,  doc  comprises three paralogues,  doc1 ,  doc2 , and  doc3 ; the absence of these genes in  D. melanogaster  is associated with embryonic death [ 78 ].  Doc  genes exhibit orthologous relationships with  TBX6 ,  TBX2 , and  TBX3 ; in vertebrates,  TBX6  is involved in left–right patterning during early mouse development [ 206 ], the regulation of skeletal musculature development [ 207 ] via effects on axial and paraxial mesoderm development, and regulation of cardiac progenitor differentiation in vitro [ 124 ].  TBX2  is involved in the development of the outflow tract and atrioventricular canal, while  TBX3  is associated with the development of both atrial and ventricular cardiomyocytes, as well as the cardiac conduction system [ 78 , 208 ].  TBX6  disruption has been associated with skeletal defects [ 209 ], while a deletion in the genomic locus that also contains  TBX6  has been associated with pulmonary atresia with ventricular septal defect, a severe form of Tetralogy of Fallot in humans, along with other candidate genes [ 210 ].  TBX2  mutations are associated with defects in outflow tract septation and atrioventricular canal development in animal models [ 208 ] and contribute to the development of Tetralogy of Fallot, single ventricle, and single atrium defects in humans. Both  TBX2  and  TBX3  have been associated with craniofacial defects in animal models [ 211 ].  TBX3  has also been implicated in congenital heart defects in  H. sapiens , including Tetralogy of Fallot, here as well, along with transposition of the great arteries [ 212 ].\nEve  is mostly associated with the diversification of cardioblast/pericardial cell populations in  D. melanogaster  models, and related defects include disruption in pericardial cell populations [ 189 ]. In vertebrates, the corresponding orthologs,  EVX1  and  EVX2 , are involved in the development of limbs and genitalia [ 213 ], but no cardiac defects have yet been associated with either, as most cases described in the literature describe defects in limb development [ 213 , 214 ]. The transcription factor gene  svp  is another factor that contributes to the diversification of cardioblast/pericardial cell groups, with loss of expression usually associated with a corresponding loss of cardioblasts that express  svp  [ 113 , 188 ]. These cardioblasts normally go on to form specialized cardiac cells that line the ostia in the dorsal vessel, functioning as inflow tracts for the circulating hemolymph [ 113 , 188 ]. In vertebrates, one of the  svp  orthologs corresponds to  NR2F2 , a gene that regulates epithelial-to-mesenchymal transition, and contributes to and later maintains atrial cardiomyocyte identity [ 215 ].  NR2F2  is expressed in the developing atria, aorta, and coronary vessels [ 216 ] and also contributes to the development of the atrioventricular canal [ 215 ] and coronary vessels [ 217 ]. Since ostia can be thought of/function as inflow tracts [ 113 , 188 ], similar to atrial chambers in the vertebrate heart, this could point to a conserved function across different cardiac systems.  NR2F2  mutations in humans have been associated with various septal defects, including atrioventricular canal defects [ 216 ], double outlet right ventricle, and Tetralogy of Fallot.  NR2F2  variants/mutations that affect the cooperation of  NR2F2  with  GATA4  have also been associated with congenital bicuspid aortic valve [ 218 ].\nFinally,  Hand  is a bHLH transcription factor, and  D-mef2  encodes transcription factors that are mainly associated with activation of structural and functional genes in cardioblasts/pericardial cells and hematopoietic progenitors [ 78 ]. Mutations in these genes are associated with dorsal vessel hypoplasia ( Hand ) [ 191 ] and cardiac tissue differentiation defects ( D-mef2 ) [ 190 ]. In vertebrates,  HAND2  interacts with Notch signaling and is involved in the development of the endocardium, ventricular trabeculation and septation, and coronary vessel maturation [ 219 ].  MEF2C  and  MEF2A , vertebrate orthologs of  D-mef2 , are involved in the development of the right ventricle, cardiomyocyte development and differentiation, and cardiac looping [ 220 ]. The contributions of these factors to cardiac development is further evident by the effects of their mutations, as in animal models,  HAND2  mutations are associated with defects in ventricular myocardial tissue, along with reduced trabeculation and defects in septation [ 219 ]; the  MEF2C  and  MEF2A  mutations are also associated with the failure of right ventricular development and cardiac looping defects [ 220 , 221 ] ( Table 1  and  Supplementary Table S2 ).\nD. melanogaster  genes comprising the core cardiac regulatory network and corresponding orthologs with the highest DIOPT score, along with any associations with congenital heart defects in humans. ASD, atrial septal defect; BAV, bicuspid aortic valve; DORV, double outlet right ventricle; HLHS, hypoplastic left heart syndrome; MR, mitral regurgitation; MVP, mitral valve prolapse; PDA, patent ductus arteriosus; PFO, patent foramen ovale; PTA, persistent truncus arteriosus; TOF, Tetralogy of Fallot; VSD, ventricular septal defect. For a complete list of all gene abbreviations, see  Supplementary Table S9 .\nAdditional genes involved in lipid [ 229 , 230 , 231 ] and glucose metabolism [ 232 ], as well as genes implicated in proteostasis [ 151 , 233 ], can also contribute to heart development and thus be implicated in the pathological heart phenotypes observed in the  D. melanogaster  model. HMG-CoA reductase (HMGCR), along with other enzymes in the mevalonate pathway and the G protein Gγ1, are all implicated in cardioblast–pericardial cell associations; in particular, modification of Gγ1 by geranylgeranylation allows for its appropriate intracellular localization, facilitating adhesion between cardioblasts and pericardial cells. Mutations in these enzymes result in the “broken-hearted” phenotype in flies, with cardioblast–pericardial cell dissociation and embryonic lethality [ 229 ]. HMGCR inhibition in humans has been reportedly associated with both cardiac (atrial and ventricular septal defects, hypoplastic aorta) and central nervous system malformations [ 234 , 235 ]. Glucose metabolism can also lead to derangements in cardiac development via an effect on endothelial nitric oxide synthase transcription. More specifically, hyperglycemia can reduce transcription at the  Nos3  locus encoding for endothelial nitric oxide synthase, leading to increased expression of Jarid, a regulator of histone methyltransferase. As a result, there is reduced nitric oxide production. Eventually, Notch expression is inhibited, and with it, the progression of cardiac development [ 232 ]. Hyperglycemia, in concert with genetic mutations, can affect cardiac development in the  D. melanogaster  system, with effects on myofibril arrangement and fibrosis, further shedding light on the mechanisms implicated in the cardiac malformations in infants of hyperglycemic mothers [ 232 ]. Evaluation of genes involved in proteostasis with unknown function in the context of congenital heart disease has shown variable defects in cardiac development in the fly, ranging from complete absence of the dorsal vessel to minimal effects on myofibril and actin organization [ 233 ], as well as partial to complete dorsal vessel atrophy [ 151 ]. Evaluation of the genes found to be implicated in hypoplastic left heart syndrome in humans has also been carried out, with relevant fly phenotypes ranging from cardiac dilation and disruption in adenosine triphosphate synthesis to mitochondrial defects [ 230 , 231 , 236 , 237 ]. Genes associated with Tetralogy of Fallot and hypertrophic cardiomyopathy in mammalian vertebrates/humans have also been evaluated in  D. melanogaster , with results ranging from cardiac constriction to cardiac dilation and effects on embryonic survival [ 152 ].\nWhile  D. melanogaster  exhibits distinct progenitor populations after cardioblast diversification events, no grouping analogous to the first heart field and second heart field present in mammalian vertebrates can be identified. Instead, genes with homology to these populations are distributed across all cardiac progenitors in the fruit fly [ 51 ]. Recent evidence, however, suggests that the ventral longitudinal muscle may be an appropriate model for the study of genetic interactions implicated in second heart field [ 90 ] development, as derivation of the ventral longitudinal muscle is facilitated by the  Org-1 -mediated suppression of  tup  [ 63 , 238 ], a genetic interaction mirrored in second heart field development with the  Org-1  ortholog  TBX1 , the  tup  ortholog  ISL1 , and FGF/FGFR signaling [ 63 , 90 ]. Despite similarities in the genetic network, however, this interaction leads to modified skeletal muscle formation in  D. melanogaster  and cardiac muscle formation in mammalian vertebrates [ 90 ] ( Supplementary Table S3  and  Figure 3 ).\nAs cardiac mesoderm becomes specified and differentiates to eventually generate progenitors such as cardioblasts and pericardial cells, it also undergoes defined movements in space. Cardiac mesoderm can be initially seen bilaterally, appearing as segmented sections, owing to the combined action of Dpp and segmented Wg expression [ 182 ]. Though it has been previously thought that the cardiac mesoderm moves passively as a result of its attachment to the overlying ectoderm, it is now known that cardiac progenitors move autonomously as a result of cellular and intercellular events [ 84 ]. Migrating cellular groups move dorsally, eventually making contact with contralateral populations, an event associated with dorsal closure of the embryo [ 84 ]. Migration, alignment, and positioning of cells across one another are mediated via conserved pathways employing Slit/Roundabout (Robo) and Roundabout2 (Robo2) signaling [ 77 , 85 ]. Both Slit and Robo are expressed in the same cell, acting in an autocrine manner [ 245 ]; the proteins accumulate between rows of migrating cells, facilitating their alignment [ 77 ]. Furthermore, Slit/Robo facilitates apicolateral cell polarization in relation to the presumptive lumen by cooperating with disks-large ( dlg ), dystroglycan ( dg ), and shotgun ( shg ) [ 246 ]. Slit/Robo is further regulated by  nmr  [ 77 , 199 ].\nSlit/Robo signaling also involves integrins and their transmembrane receptors; in general, integrins in  D. melanogaster  comprise three alpha (α) subunits (αPS1, αPS2, and αPS3) encoding for the proteins known as multiple edematous wings (Mew), Inflated (If), scab (scb), and 2 beta (β) subunits (βPS, βν). βPS encodes for myospheroid (mys) [ 247 ]. Integrins localize on the presumptive luminal aspect of migrating cardioblasts, guiding their alignment and polarization [ 83 ]; they also facilitate intercellular connections between alary muscle and pericardial cells [ 247 , 248 ]. Integrins accumulate apically in the cell due to the effects of Robo, and in return, apical localization of Slit/Robo is facilitated/stabilized by integrins [ 83 ]. Usually, sites between contralateral cardioblasts that will eventually form the lumen are repulsed due to Slit/Robo interactions, while areas in the dorsal and ventral areas attach via DE-Cadherin interactions, facilitated by Shg [ 249 ]. Slit/Robo also facilitate the formation of the outflow tract [ 57 ], while integrins further regulate the localization of pericardin, which, under physiological conditions, is found in the basal cardioblast domain between adjacent cardioblasts and pericardial cells [ 83 , 250 ]. Once cells reach the midline, dorsal interconnections between cells are generated, and afterward, ventral interconnections. The latter are usually mediated by cell division control protein 42 (Cdc42), a small GTPase protein that is part of the actomyosin network, along with other proteins that regulate actin polymerization [ 251 ]. Migrating cardioblasts exhibit cellular protrusions rich in actin, which are regulated by actin regulator proteins such as Enabled (Ena) [ 84 , 252 ]. Migration is facilitated by matrix metalloproteinases, mutations in which usually lead to variable defects in lumen formation and disruption in the collective cardioblast migration, resulting in “cardia–bifida” [ 253 , 254 ]; similarly, matrix metalloproteinase mutations contribute to “cardia–bifida” in vertebrates as well [ 255 ].\nTo achieve this regulation, Cdc42 interacts with  tin , Zipper (non-muscle myosin) [ 256 ], and dishevelled-associated activator of morphogenesis (dDAAM) [ 257 ], a member of the diaphanous-related formin (DRF) family [ 258 , 259 ]. Cdc42 also facilitates heart function in adult flies via its effect on the K+ channels, and the interaction between  cdc42 / tin  is conserved in mammalian vertebrates, with disruptions usually leading to increased QRS intervals and other arrhythmias [ 260 ] ( Supplementary Table S4 ).\nMesoderm migration in  D. melanogaster  occurs in response to FGF signaling mediated via the FGF8-like ligands Pyramus, Thisbe, and the FGFR receptor Heartless; Pyramus and Thisbe originate in the ectoderm, and Heartless is found in the mesoderm [ 261 ]. In vertebrates, FGF signaling is similarly involved in the coordination of cellular movements during gastrulation, specification of axial/paraxial mesoderm, and dorsoventral patterning with specification of dorsal and posterior cellular fates, always in coordination with other similar morphogens [ 262 ]. Mutations in the FGF signaling pathway thus affect early mesoderm migration and disrupt cardioblast and pericardial cell diversification in later stages [ 261 ]. In vertebrates, FGF10 has been implicated in cardiomyocyte proliferation during the elongation phase of the linear heart tube via recruitment of second heart field cardiac progenitors. While FGF10 mutations are associated with defects in pulmonary arteries/veins and ventricular apex localization,  Fgfr2b  mutations affecting FGFR function are implicated in ventricular septal defects, poor ventricular trabeculation, and defects in the alignment of the outflow tract [ 263 ]. In addition, FGF8, along with BMP2/4, also has a place in vertebrate cardiac development, participating in second heart field proliferation, migration, and formation of the arterial pole in the developing linear heart tube [ 264 ].\nEventually, the cardiac mesoderm emerges, a process that involves signaling via Dpp for specification of the dorsal mesoderm and later via combined Wg/Dpp signals for eventual cardiac mesoderm derivation.  Wg  is a segment polarity gene, expressed in the overlying ectoderm in a segmental pattern; it is implicated in the development of the nervous system, body segmentation, and heart morphogenesis [ 265 ].  Dpp  encodes for a BMP-like protein and participates in the dorsoventral patterning of the  D. melanogaster  embryo; Dpp when combined with Wg signaling, culminates in the eventual specification of cardiac mesoderm [ 176 , 266 , 267 ]. Disruption of Wg/Dpp signaling affects mesoderm and cardiac mesoderm specification and can cause ectopic heart tissue formation in cases of overexpression [ 266 , 268 ]. Disruption of Wg/Dpp signals in later developmental stages leads to disruption of cardioblast and pericardial cell diversification [ 267 ].\nThe need for Wg signaling is mirrored in mammalian vertebrates, albeit in a more complex manner; in these animals, canonical Wnt signaling can both induce and suppress mesoderm specification. While mesoderm induction requires Wnt signaling, the cardiac mesoderm specification that follows does not; on the contrary, it is suppressed by Wnt signaling. This helps to more clearly demarcate areas where cardiogenic tissue will eventually appear in the embryo [ 269 ]. In zebrafish, expression of Wnt8 right before gastrulation increases the number of cardiac progenitors that will eventually be generated afterward, while expression of Wnt8 after gastrulation onset, during which time cardiac development also transpires, prevents the further generation of cardiac progenitors [ 270 ]. Regarding other Wnt ligands, Wnt8a is expressed throughout the developing vertebrate heart; Wnt2a/Wnt2b are associated with the atria and inflow tracts; and finally, Wnt5a and Wnt11 are expressed mainly in the outflow tract [ 271 ]. Furthermore, while canonical Wnt signaling is associated with the development of cardiac valve cells in mammalian vertebrates, in  D. melanogaster , pygopus has been associated with this event instead [ 55 ]. Although pygopus is a component of the canonical Wnt signaling pathway, with its product functioning alongside Wg, Armadillo, and T cell factor/lymphoid-enhancer factor (TCF), no interactions have been observed between it and other components of the pathway during D. melanogaster heart development. This may suggest that pygopus functions independently of Wnt signaling, via a mechanism that affects actin organization and arrangement [ 55 ]; in mammalian vertebrates, similar mechanisms are mediated via non-canonical Wnt/planar cell polarity (PCP) signaling [ 272 ]. Evidence of non-canonical signaling in  D. melanogaster  may also be found during  svp + cardioblast during specification [ 170 , 171 ]. BMP signaling is also implicated in heart development in vertebrates, including the maintenance of  NKX2.5  expression [ 273 ], while in zebrafish, BMP signaling can also facilitate cardiac tissue regeneration [ 274 ]. Most  Wg / Wnt  mutations in  D. melanogaster  disrupt early stages of dorsal vessel development, with severe cases leading to absence of heart formation; the early pattern of activation of Wg signaling in the migration of the mesoderm also translates into widespread defects resulting in embryonic lethality, affecting both somatic and visceral muscles, as well as variable defects in ectoderm and endoderm development [ 176 , 275 , 276 ]. In later stages, disruptions in the diversification of cardioblast and pericardial cell populations also occur [ 265 ], particularly affecting the expression of  svp ,  Eve , and  Odd  [ 170 , 171 ], along with defects in cardiac valve cell formation [ 55 , 277 , 278 ]. In vertebrates, loss of Wnt5a has been associated with outflow tract defects such as persistent truncus arteriosus [ 279 ], loss of Wnt11 with ventricular septal defects and double outlet right ventricles in mice [ 280 ], and ventricular septal defect and Tetralogy of Fallot in humans [ 281 ]. Both ligands (Wnt5a and Wnt11) normally signal through the non-canonical Wnt pathway [ 282 ].\nHedgehog (Hh) signaling in  D. melanogaster  maintains segmental  Wg  expression [ 283 ] and regulates the development of various heart progenitor groups. This is carried out via RAS/MAPK signaling owing to effects on the EGFR-associated protease rhomdoid, involved in the specification of  eve+  populations. FGF signaling via Heartless also converges on the activation of RAS. Alternatively, Hh inhibits Cubitus interruptus (Ci), which normally inhibits this pathway, thus removing the inhibition and allowing for upregulation of  eve + populations and the suppression of  lb+  populations instead. As a result, Hh mutations can lead to variable effects on heart development, depending on timing, ranging from decreases in heart progenitor populations to disruption in the diversification of cardioblast and pericardial cell groups [ 176 , 284 ]. In vertebrates, Shh signaling regulates the timing of cardiomyocyte differentiation during development via activation of appropriate gene regulatory networks [ 285 ], as well as endocardium and second heart field development [ 286 ]. Disruptions in this pathway have been associated with defects in cardiac looping and left-to-right animal body patterning defects, including situs inversus, dextrocardia, atrioventricular septal defects, transposition of the great arteries, and double outlet right ventricle [ 287 ].\nEGF/EFGR signaling is also conserved in  D. melanogaster  development, facilitating, in concert with other signaling pathways, the generation of diverse cardiac cell fates [ 86 ]. In vertebrates, EGFR signaling via the Erb-B2 Receptor Tyrosine Kinase (RTK) 2 (ERBB) group mediates diverse functions during cardiac development, including proliferation/growth of cardiac progenitors, valvulogenesis, and regulation of intercellular interactions [ 288 ]. As with  D. melanogaster , in vertebrates, Notch signaling restricts cardiac cell fate [ 289 ] via upregulation of  su  ( H ) homologs [ 290 ]. Notch signaling pathways in vertebrates allow non-myogenic cell fates [ 291 ], including cells of the conduction system, to be generated [ 292 ], while experimental upregulation of Notch signaling inhibits cardiomyocyte proliferation [ 293 ]. Signaling pathways regulating the derivation of ventral longitudinal muscle from alary muscle, including Heartless and Notch signaling, have been shown to act in a similar manner in mammalian vertebrates, allowing for the derivation of second heart field cardiac progenitors [ 294 , 295 ]. Notch signaling disruptions, at least through mutations in sanpodo and Numb, affect the diversification of cardioblast and pericardial cells [ 296 ] ( Supplementary Table S5 ).\nHomeodomain or  Hox  genes encode for factors [ 297 ] necessary for the proper development and patterning of organisms; they exhibit evolutionary conservation between animal groups, from  D. melanogaster  and  D. rerio  (zebrafish) to mammalian vertebrates and humans. They are generally characterized by the presence of a conserved DNA sequence termed the homeobox sequence, which encodes for a DNA-binding domain in the final protein [ 298 ]. Based on phylogenetic classification, there are 11 groups of homeodomain-containing genes in animals [ 299 ], namely, ANTP, PRD, LIM, POU, HNF, SINE, TALE, CUT, PROS, ZF, and CERS; the  Hox  gene group is classified within the ANTP group [ 300 ]. While  Hox  genes appear in animal groups after the evolutionary divergence of Cnidaria and Bilaterians, they are arranged in chromosome clusters only in Bilaterians.  Hox  gene expression is spatially and temporally regulated and confers different cellular identities depending on their relevant position with regard to the anteroposterior body axis [ 301 , 302 ]. The correlation between placement within the chromosome and position in the animal body where activity from a particular  Hox  gene dominates is conserved as well [ 303 ].  Hox  genes located posteriorly on each chromosome additionally exhibit spatial regulation along the proximodistal axis in vertebrates [ 301 ] via histone-modifying protein complexes such as the Trithorax group (TrxG) [ 304 ] and Polycomb group (PcG).\nIn  D. melanogaster , as in most insects, there are eight  Hox  genes clustered together, albeit split across two different chromosomes [ 300 ], comprising the Homeotic Complex (HOM-C) [ 305 , 306 ]. The ANTP Complex (ANT-C) and the Bithorax Complex (BX-C) of  Hox  genes can be recognized; ANT-C contains  Antp  and is involved in the specification of T2 (mesothorax) [ 307 ] and A1 [ 308 , 309 ], while BX-C comprises Ultrabithorax ( Ubx ), Abdominal-A ( Abd-A ), and Abdominal-B ( Abd-B ) and is involved in the specification [ 303 ] of T3 and A2-A8 [ 310 ].  Abd-A  is also implicated in the specification of cardiac identity [ 309 ].\nAbd-A  exhibits the highest expression levels in  tin + cardioblasts of A6–A7, the posterior  tin + cardioblasts of segment A5, and the  svp + cardioblasts in the segment borders of A5/A6, A6/A7, and A7/A8. Lower expression levels are observed in some A5  tin + cardioblasts, as well as in  tin + and  svp + cardioblasts in A8; a general range of A5–A8 associated with the posterior dorsal vessel (heart chamber) is thus observed [ 45 , 47 ].  Abd-A  also contributes to alary muscle formation in the posterior dorsal vessel [ 311 ].  Abd-B , on the other hand, exhibits a general range of A6–A7, with its expression suppressing cardiac morphogenesis and contributing to the formation of a heart terminus (A8) during embryonic development [ 45 , 47 ]. During metamorphosis,  Abd-B  expression is regulated by  Nacα , a NAC chaperone subunit. This allows for dorsal vessel remodeling during the larval and pupa stages [ 312 ], culminating in the eventual histolysis of segments A6–A7 in response to ecdysone secretion [ 46 ].  Ubx  exhibits its highest expression in  tin+  cardioblasts of A3, with lower expression in  svp + cardioblasts of the A3/A4 border and  tin + cardioblasts of A2 and A5, with even lower expression in segments T3-A1; a general range of T3 to A1–A5 is thus observed [ 45 , 47 , 307 , 309 ]. It is also expressed in the alary muscles of the anterior dorsal vessel [ 307 , 309 ]. Finally,  Antp , along with other homeotic genes of the ANT-C, contributes to the specification of mesothorax (T2) structures, including lymph glands and the Ring gland (T3, A1). It exhibits its highest expression in  tin + cardioblasts of A2 and  svp + cardioblasts of the A1/A2 border, with lower expression in  tin + cardioblasts of T3 and A2 and in  tin – cardioblasts of A2.  Antp  expression in the posterior dorsal vessel is repressed by  Ubx  [ 45 , 307 , 308 ].\nNull mutations and ectopic expression of homeotic genes of the BX-C and ANT-C groups lead to variable disruptions in the specification of anterior (aorta) and posterior dorsal vessel (heart) identity and heart tube morphogenesis [ 45 , 46 , 47 , 308 , 309 , 311 , 312 , 313 ]. Evolution associated with multiple rounds of duplication and divergence in the ancestral  Hox  gene cluster eventually resulted in the generation of 39 genes in vertebrates. These are arranged into four gene clusters,  HoxA ,  HoxB ,  HoxC , and  HoxD , and comprise seven gene families: the anterior  Hox1  and  Hox2  groups; the  Hox3  group; the central  Hox4 ,  Hox5 , and  Hox6 – 8  groups; and finally, the posterior  Hox9 – 13  groups [ 299 , 314 ]. Furthermore, as a result of these duplication events, paralogous  Hox  genes can be found at the same relevant locations within each cluster and exhibit some functional equivalence [ 110 ].  Hox  genes may also be implicated in vertebrate cardiac development, including migration of cardiac progenitors via binding of the transcription factor Mesoderm Posterior BHLH Transcription Factor 1 (Mesp1) to  Hoxb1  regulatory sequences [ 315 ] and the development of the outflow tract [ 316 ]. Congenital heart disease has also been associated with  Hoxa1  gene mutations in both humans [ 317 ] and mice [ 316 ]. In general, of the ANTP homeotic gene group in vertebrates, anterior  Hox  families such as  Hox1  have been mostly associated with heart defects [ 318 , 319 , 320 , 321 , 322 , 323 ], with  Hox3  groups mostly associated with carotid artery malformations in mammalian vertebrates [ 324 , 325 ]. Experimental deletion of the  Hoxa / Hoxb  clusters in mice results in an atavistic heart phenotype with absence of rightward looping [ 315 ]. On the other hand, in  D. melanogaster , more posteriorly located groups (BX-C) are the ones mainly associated with cardiac development and thus, upon their perturbation, cardiac defects result instead [ 45 , 46 , 47 , 308 , 309 , 311 , 312 , 313 ]. This association is reflected in the localization of the heart chamber in mammalian vertebrates/humans compared to  D. melanogaster  in the animal body plan [ 303 ] ( Supplementary Table S6 ).\nAs with the transcriptional regulation imposed on  Hox  gene expression, gene transcription differences across different tissues and timepoints, in general, can be established via the action of chromatin-binding and chromatin-modifying factors. Differential histone modifications can often distinguish differentially functioning areas of the genome, with high levels of monomethylation at Lysine 4 of Histone 3 (H3K4me) generally associated with enhancer sequences and high levels of trimethylation at H3K4 associated with active promoter sequences. Apart from H3K4me marks, acetylation at Lysine 27 of Histone 3 (H3K27ac) is also associated with activated enhancer sequences [ 326 ]. H3K36 histone marks are also associated with active chromatin [ 327 ]. Methylation of H3K4, H3K36, and H3K27 can be carried out by histone-modifying enzymes, including protein complexes associated with SET-containing domain 1 (Set1) (COMPASS) [ 327 , 328 , 329 ]. The COMPASS series of complexes comprises the core subunits Set1, Trithorax (Trx), and Trithorax-related (Trr). Each of these proteins is the core subunit of a specific COMPASS complex, though all three also share common subunits, including absent, small, or homeotic disks 2 (Ash2); Dpy-30-like 1 (Dpy-30L1); retinoblastoma-binding protein 5 (Rbbp5); and will die slowly (Wds). Other subunits are unique to one specific COMPASS complex, including WD repeat domain 82 (Wdr82) [Set1-COMPASS], Menin 1 (Mnn1) [Trx-COMPASS], and PAX Transcription activation domain-interacting protein (Ptip) [Trr-COMPASS]. Finally, others, such as Host cell factor (Hcf) [Set1, Trx-COMPASS], are found only in specific COMPASS complexes [ 328 ]. Experimental knockdown of these subunits in the  D. melanogaster  system leads to variable effects on cardiac structure and function during larval and adult stages, as well as lethality on emergence from the pupal stage, also known as eclosion [ 35 , 327 , 328 , 330 , 331 ].\nThe proposed mechanism of action for the Set1-, Trx-, and Trr-COMPASS series of complexes during  D. melanogaster  development includes activation of Set1- and Trr-COMPASS during Stages 13 and 14. During this time, cardiac progenitors begin their migration toward the midline. While Set1-COMPASS exhibits steady activity throughout development, Trr-COMPASS is mainly active only during these earlier stages. More specifically,  Trr  exhibits a drop in expression of ~40% during Stage 16 [ 331 ]. In later developmental stages [Stages 16–17], cardiac progenitor migration results in a closed heart tube, and the Trx-COMPASS complex, along with Set1-COMPASS, further contributes to completion of heart development [ 328 ]. Histone methylation is important for physiologic adult heart function as well, as evident from the dysregulation in structure and function in relevant experiments [ 35 , 327 , 328 , 330 , 331 ]. Many of the above genes have been associated with heart defects in vertebrate models as well, including  Lysine methyltransferase 2C  ( KMT2C ) and  Lysine methyltransferase 2D  ( KMT2D ) (encoding for core subunits of the COMPASS complex series in vertebrates as well) [ 328 ], and are associated with defects such as ventricular septal defects, Tetralogy of Fallot [ 332 ], and Kabuki syndrome [ 333 ]. Kabuki syndrome comprises multiple congenital defects, including distinct facial features, skeletal abnormalities, intellectual disability, and congenital heart defects [ 334 ]. DNA methylation may also represent a cause of adult heart dysfunction in vertebrates as well, as upregulation of DNA methyltransferases 1 ( DNMT1 ) and 3 ( DNMT3 ) can upregulate (Wnt1/β-catenin signaling) or downregulate (pERK1/2 signaling) cellular pathways that promote cardiac fibrosis and heart failure with preserved ejection fraction [ 335 ] ( Table 2  and  Supplementary Table S7 ). Additional details for each of the genes described in  Section 4  and their mammalian/human orthologs can be found in  Supplementary Table S8 .\nSummary of D. melanogaster genes presented in this review; associated defects observed in the  D. melanogaster  model; corresponding ortholog with the highest DIOPT score, including weighted scores in parentheses; and associations with any congenital heart defects in animal models/humans. In cases where an ortholog cannot be found based on the DIOPT tool, other sources are employed, including the relevant literature. APOB, apolipoprotein B; ASD, atrial septal defect; BAV, bicuspid aortic valve; DORV, double outlet right; Dpp, Decapentaplegic; Dpy-30L1, Dpy-30 like 1; EGFR, epidermal growth factor receptor; EcR, ecdysone receptor; FGF8, fibroblast growth factor 8; FGFR3, fibroblast growth factor receptor 3; MMP2–14, matrix metalloproteinase 2–14; MR, mitral regurgitation; MVP, mitral valve prolapse; PFO, patent foramen ovale; PTA, persistent truncus arteriosus; TIMP3, tissue inhibitor of metalloproteinase 3; TOF, Tetralogy of Fallot; Trr, Trithorax-related; Trx, Trithorax; VSD, ventricular septal defect; VEGF, vascular endothelial growth factor; N/A, not applicable. For a complete list of all gene abbreviations, see  Supplementary Table S9 .\n\nThe evolutionary timeline of species emergence is complex, and the timeline of new gene emergence is equally or even more so; as species diverge, and new phyla emerge, new complexities arise, driven by the complexities of the genetic networks orchestrating the development of their body plans. Not only do new genes arise via duplication, but these genes can then be inherited and, with the passage of time and across different species, they may lose some of the functions found in the ancestral gene and only carry out some of these original functions in the new organism, an event termed subfunctionalization. They may even acquire new functions altogether, a phenomenon known as neofunctionalization [ 493 ].\nAs these different sets of functions arise, all these homologous genes now participate in updated or completely new regulatory networks; furthermore, interactions between different genes can also be conserved and carried into the new species. An example of this is the conserved interactions between  Org-1  and its suppression of  tup;  in the fly, this interaction leads to the derivation of ventral longitudinal muscles, a skeletal muscle that interacts with and attaches to the dorsal vessel. In vertebrates, this interaction has been conserved between the  Org-1  ortholog  TBX1  and the  tup  ortholog  ISL1 ; however, in this case, the resulting tissue that emerges comprises second heart field progenitors that contribute to elongation of the linear heart tube [ 63 , 90 , 238 ]. These conserved relationships become even more complex, as many of these orthologous relationships between genes connected by a common ancestor exhibit N:N relationship cardinality [ 112 ], which, in turn, makes genetic modeling in distantly related species complex as well [ 494 ].\nAnother caveat of modeling in distantly related species is the obvious divergence in animal form and function; though many animals will exhibit a phylotypic stage or “evolutionary hourglass” during stages of their embryogenesis, the animal body plan may differ greatly, depending on the species [ 495 ]. The fly and the human exhibit no anatomical homology and only some physiological homology in terms of heart structure and function; utilizing  D. melanogaster  and other relatively simple species, however, allows for the evaluation of genes whose deletion would be embryonically lethal to mammalian vertebrates and mammals. This is due to differences in cardiovascular and respiratory function that result in embryonic lethality and the non-viability of development past a certain stage in the latter [ 32 , 142 ]. The widespread effect of early cardiac transcription factors in the development of systems may also be the reason why it is more difficult to identify mutations of these factors in humans; some mutations are embryonically lethal and thus would not be identified in viable individuals [ 496 ].\nIn the end, the simplicity of the gene networks coupled with the simplicity in the anatomy/physiology allows for simple model organisms such as  D. melanogaster  to provide a useful first step in the identification and evaluation of candidate genes involved in cardiac development and, to an extent, cardiac defects when disrupted. These genes can then be further evaluated in other model organisms to fully ascertain their function in development and their contribution to congenital heart disease.","source_license":"CC-BY-4.0","license_restricted":false}