Glia‐specific expression of neuropeptide receptor Lgr4 regulates development and adult physiology in Drosophila

Similar to the human brain, Drosophila glia may well be divided into several subtypes that each carries out specific functions. Glial GPCRs play key roles in crosstalk between neurons and glia. Drosophila Lgr4 (dLgr4) is a human relaxin receptor homolog involved in angiogenesis, cardiovascular regulation, collagen remodeling, and wound healing. A recent study suggests that ilp7 might be the ligand for Lgr4 and regulates escape behavior of Drosophila larvae. Here we demonstrate that Drosophila Lgr4 expression in glial cells, not neurons, is necessary for early development, adult behavior, and lifespan. Reducing the Lgr4 level in glial cells disrupts Drosophila development, while knocking down other LGR family members in glia has no impact. Adult‐specific knockdown of Lgr4 in glia but not neurons reduce locomotion, male reproductive success, and animal longevity. The investigation of how glial expression of Lgr4 contributes to this behavioral alteration will increase our understanding of how insulin signaling via glia selectively modulates neuronal activity and behavior.

each carry out specific functions, such as CNS pruning, controlling synaptic signals, forming the blood-brain barrier, and providing neuroprotection.Recent studies in Drosophila suggest that glia supports the systemic metabolism for brain health (Backer & Kadow, 2022;Logan, 2017).For instance, glial glycolysis, which is mediated by glycolytic enzymes mostly produced by blood-brain barrier (BBB) glia, is crucial for neuronal survival and is involved in mediating sugar import into the Drosophila brain (Volkenhoff et al., 2015).
Leucine-rich repeat containing G-protein-coupled receptors (LGRs) make up a group of transmembrane proteins that are distinguished by having a large N-terminal extracellular domain.Drosophila Lgr4 (dLgr4) shares 40% amino acid similarity with human relaxin receptors (Hiel et al., 2012).Humans possess four relaxin receptor family genes RXFP1, RXFP2, RXFP3, and RXFP4 (Yegorov et al., 2014), these receptors and their ligands signaling are involved in a variety of activities, including angiogenesis, cardiovascular control, collagen remodeling, and wound healing (Gundlach et al., 2009).Lgr4 is strongly expressed in the adult male brain, the thoracic-abdominal ganglion, the crop, and particularly in the male midgut (Hiel et al., 2015), in contrast to Lgr3, which is highly expressed in larval CNS and regulates the timing of developmental transitions through its ligand Insulin-like peptide 8 (Ilp8) binding (Garelli et al., 2015;Gontijo & Garelli, 2018).A recent study suggests that Insulin-like peptide 7 (Ilp7) might be the ligand for Lgr4 and ilp7-Lgr4 neuropeptidergic circuit gates selective escape behavior of Drosophila larvae (Imambocus et al., 2022).However, the precise function of Lgr4 in Drosophila is still unknown.In this study, we demonstrated that Drosophila Lgr4 expression in glial cells, not neurons, is necessary for early development, adult locomotion, and lifespan.

| Fly rearing and strains
Drosophila melanogaster were raised on cornmeal-yeast medium at similar densities to yield adults with similar body size.Flies were kept in 12 h light: 12 h dark cycles (LD) at 25°C (ZT0 is the beginning of the light phase, ZT12 is the beginning of the dark phase) except for some experimental manipulation (experiments with the flies carrying RNAi reared in 20 or 29°C for experimental purposes).Wild-type flies were Canton-S.To reduce the variation from genetic background, all flies were backcrossed for at least three generations to CS strain.All mutants and transgenic lines used here have been described previously.
We are very grateful to the colleagues who provided us with many of the lines used in this study.We obtained the following line from Dr. Marc Freeman (Vollum Institute, OHSU, USA): NP2222-GAL4; from Dr. Qingzhong Ren and Tzumin Lee (HHMI Janelia Research Campus, USA): Eaat1 P88 -GAL4, Mz709-GAL4: from Dr.

| Mating assays for successful copulation
Mating assay was performed as previously described (Kim et al., 2012(Kim et al., , 2013a)).At the fifth day after eclosion, males of the appropriate strain and CS virgin females were mildly anesthetized by CO 2 .After placing a single female into the mating chamber, we inserted a transparent film and then placed a single male to the other side of the film in each chamber.After allowing for 1 h of recovery in the mating chamber in a 25°C incubator, we removed the transparent film and recorded the mating activities.Only those males that succeeded in mating within 1 h were included for analyses.All assays were performed from noon to 4 p.m.We conducted blinded studies for every test.

| Lethality assay
We evaluated the lethality of various RNAi strains expressed by tissuespecific GAL4s using two separate methods (Snow et al., 2022).First,

Significance
Expression of dLgr4 in glial cells is crucial for various aspects of development, adult behavior, and lifespan.Only Lgr4 reduction in glia will cause the impairment of development.And that there exists the potential for ilp7 as a potential ligand for Lgr4 without co-localization.Sb [1] then measures the ratio of Ser [1] and Sb[1] marker phenotype.
All the tests were done in triplicate vials and then combined into one merged sample for presentation.Second, the effects of Lgr4-RNAi expressed by various Lgr4-GAL4s were evaluated based on the number of eggs deposited and the percentage of flies that eclosed throughout development.The number of eggs generated by a particular cross was measured 24 h after the cross was done.The adults were transferred to a second vial to generate a replica, which was then destroyed after 2 days.Eggs were counted in the vial using visual discrimination under the microscope.The percentage of egg to adult viability was determined by dividing the number of flies of each cross that eclosed by the number of eggs deposited, then multiplying by 100.Male and female offspring were counted separately.By using this second method, we examined the lethality of each cross shown in Figure 6.

| Climbing assay
For climbing assay, we modified the conventional RING assay (Gargano et al., 2005).In brief, 40-50 aged flies were placed in an empty vial and were tapped to the bottom of the tube.We used 10-day-old adults reared in different temperatures.After tapping of flies, we recorded 10 s of video clip.This experiment was done five times at 5-min intervals.With recorded video files, we captured the position of flies 10 s after tapping the vial.This captured image file was then be loaded in ImageJ to perform particle analysis.

| Particle analysis with climbing assay data
For quantifying the location of flies inside vial, we used the "analyze particles" function of ImageJ (Grishagin, 2015).The position of pixels was normalized by height of vial then only the particles above the midline (4 cm) of vial were counted.

| Lifespan assay and statistical analysis
For lifespan analysis, we used conventional procedure (Piper & Partridge, 2016).Briefly, 50 flies were aged by sex before being raised in typical 12 h light: 12 h dark cycles at either 29, 25, or 20°C for each experimental objective.The number of dead flies was recorded daily for 40 days.Every 3-4 days, the surviving flies were transferred to fresh vials.To compare the survival curves of each genotype, we used the software Graph Pad Prism's Kolmogorov-Smirnov Test.

| Drosophila CNS dissection and immunostaining
As described before (Kim et al., 2013), wandering third instar larval CNS or adult brain was dissected then fixed in 4% formaldehyde for 30 min at room temperature, washed with 1% PBT three times (30 min each), and blocked in 5% normal donkey serum for 30 min.
Fluorophore-conjugated secondary antibodies are the same as above.
Fluorescence was quantified in a manually set region of interest (ROI).For quantifying fluorescence signals, we used the threshold/ color threshold and measure function of ImageJ.Threshold was used to remove background spontaneous fluorescence and color threshold was used to select overlapping signals.Relative body area and percent area were calculated as threshold area/ROI area.All specimens were imaged under identical conditions.

| Statistical analysis
Statistical analysis of climbing assay was similar to our previous studies (Kim et al., 2012(Kim et al., , 2013(Kim et al., , 2016)).Forty to fifty adults were used for climbing assay.Statistical comparisons were made between  Figures 7, 8, and S9) (Lee et al., 2023).In short, "estimation statistics" is a simple framework that-while avoiding the pitfalls of significance testing-uses familiar statistical concepts: means, mean differences, and error bars.
More importantly, it focuses on the effect size of one's experiment/ intervention, as opposed to significance testing (Claridge-Chang & Assam, 2016).In comparison to typical NHST plots, estimation graphics have the following five significant advantages such as (1) avoid false dichotomy, (2) display all observed values (3) visualize estimate precision (4) show mean difference distribution.And most importantly (5) by focusing attention on an effect size, the difference diagram encourages quantitative reasoning about the system under study (Ho et al., 2019).In 2019, the Society for Neuroscience journal eNeuro instituted a policy recommending the use of estimation graphics as the preferred method for data presentation (Bernard, 2021).

| Single-nucleus RNA-sequencing analyses-Data and code availability
snRNAseq dataset analyzed in this paper is published in (Li et al., 2022) and available at the Nextflow pipelines (VSN, https:// github.com/ vib-singl ecell -nf), the availability of raw and processed datasets for users to explore, and the development of a crowd-annotation platform with voting, comments, and references through SCope (https:// flyce llatl as.org/ scope ), linked to an online analysis platform in ASAP (https:// asap.epfl.ch/ fca).

| Gene expression pattern analyses in different tissues
Single-cell RNA sequencing (scRNA-seq) data from the Drosophila melanogaster were obtained from the Fly Cell Atlas website (https:// doi.org/ 10. 1126/ scien ce.abk2432).UMIs data were generated from the loom file of 10x VSN All (Stringent), comprising a total of 506,660 cells.The Seurat (v4.3.0)package (https:// doi.org/ 10. 1016/j.cell.2021.04.048) was utilized for data analysis.The "NormalizeData" function was used to perform automatic data normalization.Heat map is generated using "ComplexHeatmap" (https:// doi.org/ 10. 1093/ bioin forma tics/ btw313).Dot plots were generated using the "Dotplot" function, and the cells were initially categorized into 289 parts according to the "annotation broad extrapolated" (n = 17) and the "tissue" (n = 17) data downloaded with the FCA data.The Triangle threshold method was then applied to filter out categories with small cell numbers (threshold of 1500).The "Dotplot" function was used to calculate the table of dot plots for one gene.For the gene expression pattern of the Lgr1, Lgr3, and Lgr4 in each cell type of different tissues, the cell types are split by FCA.

| Glial knockdown of Lgr4 disrupt the development of Drosophila
To investigate the role of Lgr4 in development, we used pan-glial repo-GAL4 and pan-neuronal elav c155 drivers to knock-down Lgr4.
Control experiments for the lethality testing revealed that in both males and females, the ratio of eclosed animals with TM3 balancer chromosomes and wild-type chromosomes is similar in all temperatures (Figures 1a,b and S1a,b).Flies that expressed Lgr4-RNAi with repo-GAL4 showed severe mortality when reared at 25 and 29°C (Figure 1c,d), but those that expressed with elav c155 developed normally at all temperatures (Figure 1e,f).Males showed a more severe lethality phenotype than females when Lgr4 is knocked down in glial cells (compare the 29°C bars between Figure 1c,d, Šídák's multiple comparisons test p = .0072).We verified three distinct Lgr4-RNAi strains and concluded that repo-GAL4 is effective with all three RNAi TA B L E 1 Description of glial GAL4 drivers used in this study.LGR7 and LGR8, also called RXFP1 and RXFP2 respectively (Hiel et al., 2012).In order to compare the role of other fly LGRs family with Lgr4 in glial cells, we expressed Lgr1-, Lgr2-, Lgr3-RNAi using repo-GAL4.Male and female flies that expressed Lgr1-RNAi, Lgr2-RNAi, or Lgr3-RNAi with repo-GAL4 did not exhibit any abnormalities in the lethality test (Figure 2a-f).
Similar to the results of repo-GAL4, males showed a more severe lethality phenotype than females (compare the Eaat1 8849 -GAL4 bars between Figure 3a,b, Šídák's multiple comparisons test p = .0004).The lethality trait disappeared when flies were raised at 20°C, as evidenced by the repo-GAL4 data in Figure 1 (Figure S2).Next, we used recently developed subtype glial-GAL4 drivers to determine which subtypes of glia are responsible for this phenotype (Kremer et al., 2017).Both

| Glial knockdown of Lgr4 in adults Drosophila shows severe locomotion and lifespan defects
In order to investigate the role of Lgr4 in adult stage, we expressed Lgr4-RNAi at 20°C throughout development to diminish the effects of RNAi.After adult eclosion, we raised the temperature to 29°C to enhance the effects of RNAi throughout the adult stage (Ni et al., 2009).At 20°C, control experiments for the climbing assay revealed that both males and females which were expressing Lgr4-RNAi in association with repo-GAL4 or elav C155 exhibited normal locomotion (Figure S3a,b), male flies with those two genotypes also showed normal mating success ratio (Figure S3c).At 29°C, however, males and females expressing Lgr4-RNAi in association with repo-GAL4 displayed abnormal locomotion, but those expressing elav C155 exhibited normal locomotion (Figure 4a,b).Males expressing Lgr4-RNAi in glia failed to mate, whereas those in neurons mated successfully (Figure 4c).Lack of Lgr1, Lgr2, or Lgr3 in glia was not associated with mating success (Figure 4d).Only males lacking glial Lgr4, but not neuronal Lgr4, had a substantially shorter lifetime at 29°C (Figure 4e 29°C.However, the results already suggested sex differences.All these findings indicate that glial expression of Lgr4 is required for appropriate adult locomotion and lifespan.

| Lgr4 expresses highly in adult glial cells
To validate the glial expression of Lgr4, we made use of Fly Cell Atlas (FCA)-a scRNA sequencing dataset of fruit fly that is available on the SCope website (Li et al., 2022).We first reviewed the adult glia cells, and confirmed that those cells highly expresses Lgr4 both in male and female (Figure 5a,b).Notably, the newly proposed Lgr4 ligand, ilp7, is not coexpressed with Lgr4 nor expressed in adult glial cells (Figure 5c-f).In contrast to Lgr4, adult glial cells do not express Lgr3 (Figure 5g,h).Lgr4-expressing cells partly overlap with annotated "cell body glia" and "adult glial cell" when just "glial cell" is considered (Figure 5i-t).Most Lgr4-expressing glial cells were enriched in the "body," not the "head," both in males and females (Figure 5u-x).Male "testis" and "body wall" of both sexes had more Lgr4-expressing cells than other identified tissues (Figure S4).Lgr4expressing cells were enriched in "epithelial cell," "glial cell," and "muscle" based on functional annotation .Of all the neurons, Lgr4 expression is concentrated in T neurons and ventral nervous system (Figure S6) and MCFO results also show that Lgr4-positive cells are morphologically diverse in the brain and the VNC (Figure S7).
To validate the expression patterns of Lgr4 throughout development and adulthood, we analyzed the expression of existing Lgr4-GAL4 strains (Figure 6a).These Lgr4-GAL4 strains demonstrated differential lethality after being crossed with Lgr4-RNAi (Figure 6b), this lethal phenotype did not exhibit sex-specificity similar to previous results (Figures 1b,3a,b,6b').Meanwhile, we found no direct linear relationship between the GAL4-related lethal phenotype and the number or proportion of labeled glial cells (Figures 6b-d and   S8).These results suggest that the lethality caused by Lgr4 knockdown may not result from a mass expression deficit in glial cells, but rather from an expression deficit in specific glial cell subtypes, such as PNG, SPG, CG, and ALG, as demonstrated in the previous results (Figure 3c).
When paired with membrane and nuclear markers, UAS-mCD8GFP, UAS-RedStinger, putative brain enhancer GAL4 strains (Pfeiffer et al., 2008).R24A08-GAL4 and R25C08-GAL4 labeled a huge quantity of glial cells in the larval CNS (Figures S9e,f and 8a).When repo-GAL80 is coupled to R25C08-GAL4, membrane and nucleus patterns of glia (Figure 8b,c) disappeared (Figure 8e,f).RedStinger fluorescence quantification revealed that about 85 percent of all cells tagged by R25C08-GAL4 were eliminated by repo-GAL80 conjugation (Figure 8g).Lgr4 MB03440 -GAL4 also tagged adult brain surface glia in both sexes (Figure S9g) as shown in FlyLight surface glia GAL4s (Figure S9i,j), and repo-GAL80 conjugation could reduce these expression patterns (Figure S9h).GFP expression was more prevalent in male brains than in female brains (Figure S9k).Repo-GAL80 conjugation with Lgr4 MB03440 -GAL4 lowers GFP expression in the male brain by around 90% (Figure S9l).These findings indicate that Lgr4 is extensively expressed in glia throughout both developmental and adult stages.

| DISCUSS ION
Our study shows that Lgr4 expression in a subset of glial cells is essential for normal development and adult physiology.Reducing the Lgr4 level in glial cells, but not neurons, induces lethality of developing Drosophila (Figures 1 and S1), although knocking down other LGR family members in glia has no impact (Figure 2).Using previously reported glial-GAL4 drivers and newly discovered subtype glia-specific GAL4 drivers, we determined that ALG, CG, PNG, and SPG are the subtype glial populations responsible for the lethality phenotype induced by Lgr4-RNAi knockdown (Figure 3).
Adult-specific knockdown of Lgr4 in glia but not neurons reduce locomotion, male reproductive success, and lifespan (Figures 4   and S3).Using newly published Drosophila scRNA seq data and accessible Lgr4-GAL4 drivers, we showed that Lgr4 is substantially concentrated in adult body glia but not head glia (Figures 5, 7, 8, and S4, S5, S9).
However, when Lgr4 is knocked down by typical glial GAL4 drivers that have been shown to identify this subtype of glia population, no effect is observed.When combined with Lgr4-RNAi, only Eaat1and Mz97-GAL4 drivers displayed the lethal phenotype (Figure 3a,b).
Both Eaat1-and Mz97-GAL4 drivers express numerous subtypes of the glia population, such as astrocytes and other glia (Eaat1) and surface glial cell/wrapping glial cell/lamina anlage glial cell (Mz97) (Table 1).We postulate that distinct glial cell lineages are responsible for the Lgr4-mediated phenotype of lethality.Eaat1 and Mz97 appear to identify at least one of these subtype glia lineages, whereas other GAL4s do not.Future studies should determine if a specific subtype of glia is accountable for Lgr4-mediated lethality.
One of the intriguing features of Lgr4 function is that males are more sensitive to knockdown of Lgr4 in glial cells (Figures 1c,d,d).It has been shown that the level of Lgr4 expression in males is much higher than that in females (Hiel et al., 2015).
Lgr4 transcript is highly expressed in the third instar larvae's midgut, Malpighian tubules, hindgut, central nervous system, salivary glands, and fat body.The adult stages of the midgut, crop, brain, and thoracic-abdominal ganglia express Lgr4 transcript at high levels (Hiel et al., 2015).Mamalian homologues of LGRs are the relaxin-family peptide receptors RXFP1 and RXFP2, which, among other See Section 2 and previous report (Kim et al., 2013(Kim et al., , 2016) ) for detailed quantification methods.
roles, are required for proper reproduction in both sexes (Gundlach et al., 2009).Since Lgr3 in Drosophila neurons was reported to be sex-specifically regulated (Meissner et al., 2016) and controls developmental timing with its ligand ilp8 (Garelli et al., 2015;Gontijo & Garelli, 2018), it will be interesting to determine whether ilp7-Lgr4 signaling through neuron-glia in Drosophila is also sex-specifically regulated.
One of the key discoveries of our study is the significance of Lgr4 distribution of Lgr1, Lgr2, and Lgr4 in the entire body.Our results revealed that the presence of Lgr4-expressing cells was predominantly seen in several anatomical regions such as the body, head, testis, proboscis, and to a lesser extent in the fat body.However, it is important to note that the expression pattern of Lgr4 differed significantly when compared to Lgr1 and Lgr3 (Figure S10b-d).We also examined and compared the expression levels of Lgr1, Lgr3, and Lgr4 in male and female.The expression of Lgr4 is seen to be more abundant in neurons compared to glial cells in many tissues, irrespective of sex.In contrast, the expression of Lgr4 is higher in male glial cells compared to female glial cells (Figure S10e).The observed variations in the expression patterns of Lgr4 in glial cells, which are sexually dimorphic, could potentially provide an explanation for the heightened sensitivity of males to Lgr4 knockdown in numerous instances (Figures 1c,d,d).
Ilp7 encodes a secreted insulin-like peptide that binds to the receptor encoded by InR to activate the insulin signaling pathway.
Previous studies have shown that ilp7 is a ligand for Lgr4 (Imambocus et al., 2022).Ilp7+ neurons are located in the larval VNC, providing output mostly in the subesophageal zone (SEZ) and brain lobe region along its dorsally projecting axon.They also synthesize and secrete this neuropeptide into downstream cells, thus there is a possibility of co-localization or non-co-localization with Lgr4.We explored this possibility via the data from SCope single-cell sequencing and found that Ilp7 is not co-expressed with Lgr4, nor expressed in adult glial cells.
Relaxin subfamily in vertebrates plays diverse roles in tissue homeostasis and remodeling, behavior, and reproduction (Gundlach et al., 2009).Recent reports indicate that ilp7-Lgr4 signaling works as circuit gates for escape behavior (Prince et al., 2021).The investigation of how glial and non-CNS tissue expression of Lgr4 contributes to this behavioral alteration will increase our understanding of how insulin signaling via glia selectively modulates neuronal activity and behavior.report (Kim et al., 2013(Kim et al., , 2016) ) for details.See Section 2 and previous report (Kim et al., 2013(Kim et al., , 2016) ) for detailed quantification methods.report (Kim et al., 2013(Kim et al., , 2016) ) for detailed quantification methods.
effects of each cross for lethality, we used TM3 or TM6B balancer chromosomes.As shown in Figures 1 and S1-S3, virgin RNAi/TM6B female flies and male GAL4/TM3 flies were crossed and the proportion of progeny with TM3 or TM6B balancer was counted and compared to those without TM3 or TM6B.For control experiments shown in Figure 1a,b, we used male TM3, y[+] Ser[1]/ F I G U R E 1 The impact of Lgr4 knockdown on neuronal and glial cell populations, as determined by the percentage of lethality from egg to adult.(a and b) Experiments serving as controls for the lethality test used in this study.The proportion of (a) male or (b) female flies with the TM3 balancer chromosome (black bar) or RNAi (gray bar) is shown in the donut graph.All results were collected from the triplicate vials.The donut graph represents the accumulated data from the triplicate tests.Total genotypes refer to the percentage of each of the two genotypes in the graph out of the total number of genotypes.20°C male n = 1000, female n = 1088; 25°C male n = 1041, female n = 1023; 29°C male n = 1281, female n = 1183.(a and b) Šídák's multiple comparisons test of male versus female 20°C = .2364,25°C = .1616,29°C = .6599.(c and d) The accumulated (donut) proportion of (c) male or (d) female flies with TM3 balancer chromosome (black) or RNAi (gray) conjugated with repo-GAL4 driver.20°C male n = 1320, female n = 1397; 25°C male n = 778, female n = 1020; 29°C male n = 540, female n = 718.(c and d) Šídák's multiple comparisons test of male versus female 20°C = .8477,25°C = .1053,29°C = .0072.(e and f) The accumulated (donut) proportion of (e) male or (f) female flies with TM3 balancer chromosome (black) or RNAi (gray) conjugated with elav c155 driver.The sex of each animal is labeled above the graph.20°C male n = 1820, female n = 1563; 25°C male n = 1003, female n = 1120; 29°C male n = 1251, female n = 1406.(e and f) Šídák's multiple comparisons test of male versus female 20°C = .9158,25°C = .8356,29°C = .8537.Genotypes of each animal are labeled below the graph.The numbers inside the donut graphs represent the proportion of an eclosed animal.The p values in Šídák's multiple comparisons test presented are adjusted p values.See Section 2 for detailed experimental methods and statistical analysis used in this study.
naively reared or sexually experienced within each experiment.As climbing data of adults showed normal distribution (Kolmogorov-Smirnov tests, p > .05),mean values were compared by one-way ANOVA.Each figure shows the mean ± standard deviation (SD) (****p < .0001,***p < .001,**p < .01,*p < .05).All analysis was done in GraphPad (Prism8, RRID:SCR_002798).Individual tests and significance are detailed in figure legends.Besides traditional t-test for statistical analysis, we added estimation statistics for all climbing assays and two group comparing graphs ( The impact of LGRs knockdown on glial cell populations, as determined by the percentage of lethality from egg to adult.(a-f) The accumulated (right donut) proportion of (a, c, and e) male or (b, d, and f) female flies with TM3 balancer chromosome (black) or repo-GAL4 (gray) conjugated with (a and b) Lgr1-RNAi (c and d) Lgr2-RNAi (E-F) Lgr3-RNAi.The sex of each animal is labeled above the graph.Genotypes of each animal are labeled below the graph.The numbers inside the donut graphs represent the proportion of an eclosed animal.Total genotypes refer to the percentage of each of the two genotypes in the graph out of the total number of genotypes.(a-f) n = 300.Šídák's multiple comparisons test of male versus female in (a and b) 20°C = .9977,25°C = .3345,29°C = .8564(c and d) 20°C = .6868,25°C = .9816,29°C = .0945(e and f) 20°C = .9956,25°C = .9388,29°C = .9994.The p values in Šídák's multiple comparisons test presented are adjusted p values.
S1c,d).All these data suggest that glial expression of Lgr4 is essential for normal development in Drosophila.Drosophila genome contains four LGRs family, Lgr1, Lgr2, Lgr3, and Lgr4.Fly Lgr1 belongs to type A LGRs, which contain mammalian glycoprotein hormone receptors (FSHR, LH/CGR, and TSHR).Fly Lgr2 belongs to type B LGRs, which contain mammalian LGR4, 5, and 6. Fly Lgr3 and Lgr4 belong to type C LGRs and contain mammalian males and females expressing Lgr4-RNAi conjugated with ALG-, CG-, PNG-, and SPG-GAL4 drivers had a lethal phenotype, with males exhibiting a more severe phenotype than females (Figure 3c,d), as shown by repo-GAL4 (Figure 1c,d) and other glial-GAL4 drivers (Figure 3a,b).All these findings show that the expression of Lgr4 in certain subtypes of glia is crucial for lethality of animal.F I G U R E 3 Testing the functionality of glial GAL4 drivers with Lgr4-RNAi.(a and b) The average (left bar) proportion of (a) male or (b) female flies with CyO or TM6B balancer chromosome (black) or Lgr4-RNAi (gray) conjugated with (a and b) previously described glial-GAL4 drivers (c and d) recently developed subtype glial-GAL4 drivers.The sex of each animal is labeled above the graph.Genotypes of each animal are labeled below the graph.The numbers inside the bar graphs represent the proportion of an eclosed animal.Šídák's multiple comparisons test of male versus female in (a and b) Alrm(II) = .9941,Alrm(III) = .9914,NP2222 > .9999,Eaat = .0004,Eaat1 = .9957,Mz709 > .9999,GCM = .9109,17A = .9922,A8C > .9999,A58 > .9999,Mz97 = .5061(c and d) ALG = .3216,CG = .2001,EGN = .9733,EGT > .9999,PNG = .8749,SPG = .8751.The p values in Šídák's multiple comparisons test presented are adjusted p values.
-g).Females' lifetime, however, displayed none statistically significant difference regardless of whether lack glial Lgr4 or neuronal Lgr4 in all temperatures (FigureS3d-f).If the flies' survival rate was monitored for an extended period of time, perhaps females would also show a statistically significant reduction in lifespan at F I G U R E 4 The impact of Lgr4 knockdown in adult behavior and physiology.(a and b) Bar graphs with dot plots represent the percent of (a) male or (b) female flies crossed midline.All experiments were performed five times after a sufficient recovering period.Asterisks represent significant differences revealed by unpaired Student's t test (*p < .05,**p < .01,***p < .001,****p < .0001).n.s.represents nonsignificant differences revealed by unpaired Student's t test.The same notations of climbing assay for statistical analysis are used in other figures.Each dot represents 50 animals.See Section 2 and previous report Snow et al., 2022 for detailed quantification methods of climbing assay and mating assay.(c and d) Bar graphs with dot plots represent the percent of male flies successfully mated within 1 h with wild-type virgin female.In (a-d) Each dot represents 50 animals.(e-g) Each colored dot represents the percentage of survived flies at that day.The difference between lines was inferred by linear regression analysis.Genotypes of all above data are labeled below the graph.Rearing temperature, age, and sex of animals are labeled within the graph.Asterisks represent significant differences revealed by unpaired Kolmogorov-Smirnov test (*p < .05,**p < .01,***p < .001,****p < .0001).n.s.represents non-significant differences revealed by unpaired Kolmogorov-Smirnov test.F I G U R E 5 t-distributed stochastic neighbor embedding (t-SNE) representation of scRNA-seq datasets from individual cells of the D. melanogaster-generated as part of the Fly Cell Atlas (10× stringent dataset)(Li et al., 2022)-colored for cell types and expression of the indicated genes shows that.Annotations of cell clusters are adapted fromLi et al. (2022).(a-h) t-SNE plot showing Lgr4 (red), ilp7 (blue) expression overlapping with adult glial cell (green) in whole body of male and female.(i-t) t-SNE plot showing Lgr4 (red) expression overlapping with different annotations in the glia of both male and female, including (i and j) ensheathing glial cell, (k and l) CNS surface associated glial cell, (m and n) perineurial glial sheath, (o and p) subperineurial glial sheath, (q and r) cell body glial cell and (s and t) adult glial cell.(u-x) t-SNE plot showing Lgr4 (red) expression overlapping with adult glial cell (green) in (u and v) head and (w and x) body of male and female.Dotted lines indicate glial cell populations that express Lgr4.
activity in glial cells for several biological processes, including life span, courting behavior, and developmental lethality.To support the credibility of our experimental results, we employed a publicly available single-cell RNA sequencing dataset of Drosophila melanogaster from the SCope website (https:// flyce llatl as.org/ ).The investigation focused on assessing the levels of expression of Lgr1, Lgr2, and Lgr4 across different tissues.The findings revealed that Lgr4 and Lgr3 display comparable patterns of tissue-specific expression, which contrast with the expression pattern observed for Lgr1 (FigureS10a).Following this, a thorough examination was conducted on the overall F I G U R E 7 Sex differences of Lgr4-GAL4s labeled glial cells.(a and b) Third instar (a) male or (b) female larval CNS expressing Lgr4[MB014850] with UAS-CD4tdGFP were immunostained with anti-GFP (green) and anti-nc82 (magenta) antibodies.Gray panel represents GFP signals for quantification to set threshold shown in the right panels.Scale bars represent 100 μm.(c) Percent of body area quantified from (a) and (b).GFP signals represent GAL4 expressing "Head" or "Body" region were calculated by ImageJ threshold and measure function then quantified.p = .4082.(d-f)The same experiments were performed as described in (a-c) except the Lgr4[MB03440] was used as GAL4 driver.p = .0009.See Section 2 and previous report(Kim et al., 2013(Kim et al., , 2016) ) for detailed quantification methods.F I G U R E 8The most of Lgr4-GAL4-labeled cells are glia.(a-f) Third instar larval CNS expressing UAS-mCD8GFP, UAS-RedStinger with (a-c) R25C08-GAL4 or (d-f) R25C08-GAL4, repo-GAL80 were immunostained with anti-GFP (green), anti-DsRed (red), and anti-nc82 (blue) antibodies.Gray panel represents GFP and RedStinger signals for quantification to set threshold shown in (g).Scale bars represent 100 μm.Red dotted lines represent nuclei of glia and white dotted lines nuclei of neurons.(g) Percent area of RedStinger signals quantified from (c) and (f).RedStigner signals represent the number of GAL4 expressing cells that were calculated by ImageJ threshold and measure function then quantified.The right panel represents RedStinger signals of set threshold by ImageJ for quantification.p < .0001.See Section 2 for detailed quantification methods.
cells of the D. melanogaster-generated as part of the Fly Cell Atlas (10× stringent dataset) (Li et al., 2022)color for adult glial cell and expression of Lgr4 and ilp7.(a-x) t-SNE plot showing Lgr4 (red) and ilp7 (blue) expression overlapping with adult glial cell in (a and b) antenna, (c and d) testis, (e and f) wing, (g and h) trachea, (i and j) proboscis and maxillary palps, (k and l) malpighian tubule, (m and n) leg, (o and p) heart, (q and r) haltere, (s and t) gut, (u and v) fat body, and (w and x) body wall of male and female.Dotted lines indicate glial cell populations that express Lgr4.

Fig. S5 .
Fig. S5.Adult glial cells express Lgr4 in some of the annotated cell populations, including the lepithelial cell, glial cell, and muscle.t-SNE plot representation of scRNA-seq datasets from individual cells of the D. melanogaster-generated as part of the Fly Cell Atlas (10× stringent dataset) (Li et al., 2022)-color for adult glial cell and expression of Lgr4 and ilp7.(a-o) t-SNE plot showing Lgr4 (red) and ilp7 (blue) expression overlapping with adult glial cells in different annotated cell populations: (a) adipose system, (b) cardiac cell, (c) epithelial cell, (d) excretory system, (e) germline cell, (f) gland, (g) glial cell, (h) gut cell, (i) hemocyte, (j) muscle, (k) neuron and sensory organ cell, (l) oenocyte, (m) pigment cell, (n) stem cell, and (o) tracheal cell of male and female.Dotted lines indicate glial cell populations that express Lgr4.

Fig. S10 .
Fig. S10.Fly cell atlas data showing the expression pattern of Lgr1, Lgr3, and Lgr4.(a) heat map showing the expression level of Lgr1, Lgr3, and Lgr4 in different tissues.The scaled normalized UMIs are indicated by the intensity of the dot color (Average Expression).(b-d) the dot plot illustrates the expression levels of (b) Lgr4, (c) Lgr3, and (e) Lgr1 in various cell types of several tissues.The size of each dot reflects the percentage of cells expressing genes within each cell type (Percentage Expressed).The average normalized expression level is indicated by the intensity of the dot color (Average Expression).(e) the dot plot illustrates the expression levels of Lgr1, Lgr3, and Lgr4 in various cell types of several tissues between male and female.The size of each dot reflects the percentage of cells expressing genes within each cell type (Percentage Expressed).The average normalized expression level is indicated by the intensity of the dot color (Average Expression) https:// doi.org/ 10. 1371/ journ al. pgen.10107 53.g012.Data S1.Transparent Science Questionnaire for Authors