Drosophila melanogaster has been studied for over a century, serving as a prominent model organism in genetics, biochemistry, molecular biology, and bioinformatics. Even though the D. melanogaster genome was sequenced in 2000 (Adams et al., 2000), there are no available phenotypic data to support the characterization and annotation of a number of computed genes (CGs). One example is CG18135, which was identified in 2004 via computational techniques (Gramates et al., 2022).

CG18135 is located on 3L:18990026–18998283 on the genomic minus strand and is known to encode five transcripts that are translated into proteins having between 636 and 770 amino acids. CG18135 sequence is not only conserved in higher eukaryotes such as fish, reptiles, birds, and mammals but also in invertebrates and yeasts, according to DIOPT v9.1 (Hu et al., 2011).The CG18135 gene is orthologous to human glycerophosphocholine phosphodiesterase 1 (GPCPD1). CG18135 gene function was not previously genetically studied in D. melanogaster, but the associated protein was evidenced to facilitate myosin filament binding during embryogenesis via the interaction with the Myo10A protein at the plasma membrane level (Liu et al., 2008) and participate in glycerophospholipid catabolism (Gaudet et al., 2011).

CG18135 has been documented to participate in embryogenesis and to be active through all the developmental stages in wild-type individuals of D. melanogaster. Increased levels of CG18135 RNA were identified in both early (0–4 h) and late (20–24 h) embryonic stages (Fisher et al., 2012). Highly increased levels of CG18135 RNA were found in pupae and adult flies compared to embryos, suggesting that the gene plays an important role beyond the embryonic stage (Casas-Vila et al., 2017). In previous studies, it has been shown that 1-week-old adult male flies have significantly higher CG18135 gene expression than similarly aged female flies (Brown et al., 2014; Casas-Vila et al., 2017).

Several transposon insertions in CG18135 were reported in FlyBase (FB 2023_05), namely, nine P elements, one PiggyBac (PBac), and two Minos-mediated integration cassette (MiMIC)-derived artificial mobile elements (Buels et al., 2016; Gramates et al., 2022; FlyBase Curators, 2004). The P and PBac artificial transposons as well as one MiMIC (Venken et al., 2011) are located in the 5’ half of CG18135, defining a local insertional hotspot with a total of 11 transposons in a 1267-bp span. Interestingly, none of them determine noticeable phenotypic consequences.

During one of our previous experiments, we achieved mutagenesis mediated by the mobilization of the P{lacW}γCop ⁰⁵⁷³⁰² transposon (Ratiu et al., 2008) located in the 5′UTR of γCop. The idea backing this study was that transposon mobilization starting from an essential gene, such as γCop, would lead to conservative reinsertions in other essential genes, eventually functionally linked. One of the lines generated in this experiment is MZ4CM3, which contains both P{lacW}γCop ⁰⁵⁷³⁰² and P{lacW} ^{CG18135.MZ4CM3} transposons located on the 3R and 3L chromosomes, respectively. The latter P{lacW} resides in the first intron of the CG18135-RB transcript at the genomic coordinates 3L:18996755–18996762 (GenBank/NCBI accession number: HQ695001.1). This nucleotide interval is the one with the maximum resolution for P{lacW} ^{CG18135.MZ4CM3} mapping, as debated by Ecovoiu et al. (2016), due to the elusive nature of the octet duplication. Since P{lacW}γCop ⁰⁵⁷³⁰² determines embryonic lethality in homozygous conditions (Deak et al., 1997; Ecovoiu et al., 2002), the actual phenotypic effect of P{lacW} ^{CG18135.MZ4CM3} was impossible to assess as long as the two alleles were linked. We successfully performed a series of genetic crosses, allowing us to separate the P{lacW}γCop ⁰⁵⁷³⁰² and P{lacW} ^{CG18135.MZ4CM3} transposons by crossing-over and constructing the CG18135 ^{P{lacW}CG18135} -Sep1H line, which contains only the novel P{lacW} insertion in CG18135. To out-cross P{lacW} ^{CG18135.MZ4CM3} , we used the w¹¹¹⁸ strain; thus, the third chromosome containing solely P{lacW} ^{CG18135.MZ4CM3} has a mosaic structure consisting of fragments from both the original and isogenic w¹¹¹⁸ genotypes. Considering the new chromosome makeup, we assumed that the specific phenotypes of the CG18135 ^{P{lacW}CG18135} -Sep1H line are caused by perturbations of CG18135 gene functions. Accordingly, the reversion test confirmed that P{lacW} ^{CG18135.MZ4CM3} excision rescues the specific mutant phenotypes.

Herein, we report the first phenotypic and functional characterization of this original insertional allele of the CG18135 gene, symbolized CG18135 ^{P{lacW}CG18135} . Our main objectives were to investigate the viability of heterozygous and homozygous CG18135 ^{P{lacW}CG18135} mutant flies during various developmental stages, score the phenotypic features of the elusive homozygous mutant adult escapers, and evaluate the CG18135 expression level in adult male and female mutant heterozygotes. Moreover, we used a set of bioinformatics tools in order to investigate the putative regulatory consequences induced by P{lacW} ^{CG18135.MZ4CM3} .

In this study, we investigated the mutant phenotypes of offspring from CG18135-P{lacW}/GFP and CG18135-P{lacW}/TM6B lines. We found that, for CG18135 ^{P{lacW}CG18135} /CG18135 ^{P{lacW}CG18135} mutants, the larval and pupal developmental stages were strongly associated with lethality. To the best of our knowledge, this is the first viability analysis of any CG18135 allele in D. melanogaster.

The lack of statistically significant differences between the observed and expected lethality of mutant embryos suggests that the CG18135 ^{P{lacW}CG18135} allele is not associated with embryonic lethality. This result could be attributed to the parental delivery of functional CG18135 proteins either in the unfertilized egg (Fisher et al., 2012) or via the seminal fluid (Findlay et al., 2008), fulfilling the CG18135 protein requirements in early embryonic stages.

Previous proteomics studies identified increased CG18135 protein levels in D. melanogaster larvae, pupae, and adults compared to embryos (Brown et al., 2014; Casas-Vila et al., 2017). In addition, high expression of CG18135 was found in the Bolwig’s organ in larvae (Fisher et al., 2012). The corresponding protein was identified within the plasma membrane of D. melanogaster adults’ heads (Aradska et al., 2015). Taken together, these data suggest that the CG18135 protein requirement could be higher in larvae, and in the absence of wild-type protein production, the maternal or paternal CG18135 delivery in embryos is not sufficient to support normal development, thus increasing larval lethality.

We observed that a small percentage of CG18135 ^{P{lacW}CG18135} /CG18135 ^{P{lacW}CG18135} pupae developed to the adult stage. The CG18135 protein participates in myosin filament binding within muscular tissues and cell division (Liu et al., 2008). Therefore, it is expected that the protein requirement increases during pupal development, which implies organogenesis and high rates of cellular division. Our previous preliminary data revealed that the pupal lethality of CantonS and individuals heterozygous for the CG18135 ^{P{lacW}CG18135} allele is ≈6.7% and ≈10.3%, respectively. For CG18135 ^{P{lacW}CG18135} /CG18135 ^{P{lacW}CG18135} mutants, almost all the surviving individuals that get through the embryonic and larval stages would die inside or attached to the puparium.

These results reveal two major lethality thresholds in the development stages: one in the early larval stage (but we do not exclude a certain degree of mortality in the late pre-larval embryonic stage) and the second in the late stages of pupal development. The latter could be considered partial lethality since rare imago escapers are active for a maximum of approximately 2 days, but beyond that age, the lethality induced by the CG18135 ^{P{lacW}CG18135} /CG18135 ^{P{lacW}CG18135} genotype is absolute.

We hypothesize that the CG18135 gene participates in various biological and molecular processes in D. melanogaster, similar to its structural orthologous gene in higher eukaryotes, GPCPD1. In humans, GPCPD1 knockdown of malignant cells showed an increased intracellular glycerophosphocholine/phosphocholine (GPC/PC) ratio, a decreased lipid metabolism, and cell migration inhibition (Steward et al., 2012). In higher eukaryotes, GPCPD1 facilitates GPC degradation, an essential phospholipidic membrane constituent (Eibl, 1980; Okazaki et al., 2010), into glycerol-3-phosphate (G3P) and choline (Steward et al., 2012). In eukaryotes, the G3P requirement is obtained from glycolysis (Li et al., 2019; Nguyen et al., 2019), and together with choline, it is further used for GPC synthesis via the Kennedy pathway (Veldhuizen et al., 1989; Fagone and Jackowski, 2013). These observations suggest that the CG18135 protein may be involved in the regulation of a negative feedback loop between G3P, choline, and GPC (Figure 5). Furthermore, choline contributes to the de novo synthesis of phosphatidylcholine and phosphatidylethanolamine membrane constituents via the Kennedy pathway (Gibellini and Smith, 2010; Steward et al., 2012) and participates in the Krebs cycle (Steward et al., 2012), being an essential component in energy metabolism.

The observed phenotypes of homozygous mutant imago escapers are consistent with the proposed involvement of the CG18135 gene in either muscular activity or energy metabolism.

The participation of CG18135 in embryonic myosin filament binding has been previously demonstrated (Liu et al., 2008). Being localized in the cellular membrane vicinity, Myo10A participates in myosin filament binding during embryogenesis together with several cargo proteins, including DE-cadherin, alpha-tubulin, Katanin-60, Milton, atypical protein kinase C (aPKC), NEDD4, and the CG18135 protein. This interaction might be, at least partially, responsible for the unsynchronized movements observed in mutant adult escapers, as an abnormal delivery of the CG18135 protein could result in muscle contraction defects. In addition, the structural orthologous gene GPCPD1 is associated with muscle development in mice (Okazaki et al., 2010; Cikes et al., 2021). Okazaki et al. (2010) showed that GPCPD1 gene expression was downregulated in atrophied skeletal muscles, and GPCPD1 gene knockout enhances myoblastic differentiation and limits the progression of skeletal muscle atrophy (Okazaki et al., 2010). Thus, a lack of sufficient CG18135 protein could increase the lethality rates in larvae or pupae, as observed in our study.

A recent study suggests that the inactivation of GPCPD1 also results in early muscle aging due to GPC accumulation (Cikes et al., 2021). More specifically, the “aged-like” muscles of GPCPC1 deficient mice were caused by abnormal glucose metabolism and not by muscle development per se (Cikes et al., 2021). Interestingly, increased GPC levels were found only in the muscle (but not in the global lipidome) and did not occur in GPCPD1 deficiency in other tissues, such as fat, brown fat, or liver (Cikes et al., 2021). These results suggest that GPCPD1 is associated with glucose metabolism within the muscle tissue.

In addition, from the spasmodic leg movement of imago escapers, we observed a general lack of activity, which could be caused by either an abnormal energy metabolism or a decreased response to environmental stimuli. We would rather overlook the latter, as CG18135-P{lacW}/GFP is derived from a line that did not show any phenotypes that might suggest an abnormal response to the stimuli. Considering the previously discussed functions of GPCPD1 in muscle contractions and energy metabolism, the sedentary phenotype might be caused by a lack of global energy availability in the imago escapers or by a lack of coordinated muscle contractions that require ATP. These phenotypes were evident in imago individuals, but we did not notice such phenotypes in larvae. Although more investigations are required, preliminary data might suggest that CG18135 protein necessity is higher in the later stages of development. Further experiments might not only focus on the expression of the CG18135 gene in mutant larvae or pupae but also on the GPC/PC ratio in mutants compared to wild-type.

Considering the high lethality of CG18135 ^{P{lacW}CG18135} /CG18135 ^{P{lacW}CG18135} mutant flies prior to the adult stage, we decided to evaluate the CG18135 gene expression in adult flies, both in male versus female groups and in mutant versus control groups. Following the quantification of CG18135 gene expression in different D. melanogaster adult groups, we found that CG18135 ^{P{lacW}CG18135} /TM6B mutant female flies and CantonS control female flies displayed a similar expression level, even if the mutant female flies had only one functional allele of CG18135.

Interestingly, we found that CG18135 was significantly overexpressed in heterozygous mutant male flies compared to control male flies. It is plausible that in the absence of one functional allele, there might be an increase in the GPC/PC ratio (Steward et al., 2012). In this context, in order to reduce the GPC/PC ratio, an upregulation of CG18135 and correspondingly greater protein production are necessary. The biochemical pathways involving phospholipids are complex, and because we did not observe similar expression profiles in mutant female flies compared to control female flies, we consider that it is not possible to infer a definitive conclusion.

In male flies, the CG18135 protein is present in the seminal fluid of D. melanogaster individuals at mating (Findlay et al., 2008), which could partially explain the overexpression of the CG18135 gene in mutant or wild-type male flies compared to female flies. Additionally, the reproductive system of male flies is associated with intense cellular division and cellular differentiation during spermatogenesis (Fabian and Brill, 2012). These processes are dependent on the availability of glycerophospholipids and may influence the expression of CG18135.

The observed CG18135 overexpression in male flies compared to female adult flies is consistent with previous transcriptomics studies (Brown et al., 2014; Casas-Vila et al., 2017). The CG18135 overexpression in heterozygous mutant male flies versus heterozygous mutant female flies (FC = 4.927) could be caused by a combination of factors that drive overexpression in mutant male flies compared to control male flies (FC = 1.85) and in control male flies compared to control female flies (FC = 2.724).

Genomic context analysis showed that the P{lacW} insertion in CG18135 resides adjacently to a putative binding site for SP1/KLF TFs. Several members of this family of zinc finger proteins, such as Sp1, buttonhead, and cabut (cbt), act as activators or repressors of various developmental processes involving head structures and sensory organs (Brown et al., 2005). If the respective binding site is not efficiently accessed by cbt, a series of downstream interactions could be influenced, including dorsal closure and neuronal development in embryos, wing disc morphogenesis, muscle degeneration, and energy metabolism (Bejarano et al., 2008; Belacortu et al., 2011; Rodriguez, 2011). Abnormal cbt activation might partially explain the specific wing phenotype in mutant imago escapers, but it could also influence the mechanism of flight and jump muscle control as well as the regulation of the number of tergal depressors in the trochanter (TDT) jump muscles (Giedd, 2018). Considering that cbt takes part in energy metabolism via a regulatory link between sugar sensing and the circadian clock (Bartok et al., 2015), cbt interaction with the CG18135-located binding site might further aggravate the sedentary behavior or sporadic movements of mutant adult escapers.

In humans, the homologs of cbt are Krüppel-like factor 10 (KLF10) and KLF11 (Bartok et al., 2015). Both are part of a TF family that regulates CG-rich promoters (Lomberk et al., 2013) and influences energy metabolism. KLF11 is upregulated during starvation (Zhang et al., 2013), while KLF10 is overexpressed in response to high sugar levels (Iizuka et al., 2011) and is considered a true cbt functional ortholog (Bartok et al., 2015). These TFs are also involved in TGF-β-mediated cell proliferation, differentiation, and apoptosis (Lomberk and Urrutia, 2005).

In our view, the mere presence of P{lacW} ^{CG18135.MZ4CM3} might exert profound effects on the overall transcription regulation of CG18135 since CRM14763 is split and various neighboring CRMs that could share functional roles are forcibly separated.

In this study, we characterized mutant phenotypes produced by the CG18135 ^{P{lacW}CG18135} insertional allele for the first time in D. melanogaster. We performed several viability assessments, which allowed us to establish that the early larval and late pupal developmental stages are critical for lethality onset. Rare imago escapers that live a very short life exhibit very specific phenotypes, affecting the eye and wing morphology as well as their locomotor proficiency. These severe phenotypes are probably the consequence of the regulatory perturbations that are likely to be caused by the insertion of the P{lacW} ^{CG18135.MZ4CM3} transposon inside a group of CRMs containing binding sites recognized by wing and energy metabolism-specific, as well as ubiquitous, TFs. The heterozygous mutant flies are normally developing, but we found that the gene expression of CG18135 is significantly different in male flies compared to female flies and in mutant flies compared to control male flies. Our findings support the essential role of the CG18135 gene in D. melanogaster and define a novel experimental platform for modeling the interactions of the GPCPD1 structural and functional orthologous genes in higher eukaryotes.