Phylogenetic analysis revealed two orthologs of thyrostimulin GPA2 and GPB5 subunits in the C. elegans genome (8, 9). These are encoded by the genes flr-2 and T23B12.8 (15, 21), which we called gpla-1 and gplb-1 (glycoprotein hormone-like alpha and beta), respectively. GPLA-1 and GPLB-1 resemble vertebrate GPA2 and GPB5 and are widely conserved in nematodes ( Figures S1 , S2 ). In C. elegans, mutations in gpla-1 have been reported to suppress slow growing of several growth-defective mutants (15, 47). We therefore hypothesized that thyrostimulin-like signaling may influence C. elegans body size. To test this, we generated complete gene knockouts of gpla-1 and gplb-1 using CRISPR/Cas9 gene editing ( Figure 1A ) and investigated if these null mutants show body size defects at day 1 of adulthood (65 hours post L1 arrest). The generated gpla-1 (ibt1) and gplb-1 (ibt4) knockout mutants display a significantly shorter body length compared to wild-type animals ( Figures 1B, C ). This size defect persists for at least 5 days into adulthood, suggesting that it results from defective growth rather than a delay in the timing of development ( Figure S3A ). Additionally, worms lacking gpla-1 and gplb-1 have a reduced body width compared to wild type ( Figure S3B ). We also measured body sizes of two independently generated gpla-1 mutants, carrying the ut5 and ok2127 alleles ( Figure 1A ). The gpla-1 (ut5) mutant carries a substitution of the glycine residue between the second and third cysteine (G63E), which is highly conserved and important for proper cysteine knot formation (45). The gpla-1 (ok2127) mutant contains a large deletion, removing the start codon and the complete first exon of the gene ( Figure 1A ). We found that these mutants also display a decreased body length compared to wild-type animals. Mutants carrying the ut5 allele show a reduction in body width as well ( Figures S3C, D ). The defect in length, however, is most pronounced in all gpla-1 and gplb-1 mutants. Therefore, we quantified body length in further experiments. Expressing wild-type gpla-1 and gplb-1 under control of their endogenous promoter in their respective mutant background fully restored body length ( Figures 1B, C ), indicating that both GPA2 and GPB5 orthologs are required for normal body size.

To determine whether GPLA-1 and GPLB-1 act in the same growth-regulating pathway, we quantified the body length of double mutants. Mutants lacking both genes did not significantly differ in length from the single mutants ( Figure 1D ). This suggests that gpla-1 and gplb-1 indeed act in the same genetic pathway to control body size. Overexpression of each individual gene increased the length of wild-type worms ( Figure 1E ), suggesting that GPLA-1 and GPLB-1 are sufficient to induce growth.

To unravel how thyrostimulin-like signaling regulates C. elegans body size, we aimed to identify its target receptor. Since glycoprotein hormones typically bind to LGRs (8), we hypothesized that GPLA-1 and GPLB-1 may signal through FSHR-1, which is the sole glycoprotein hormone receptor ortholog in C. elegans (48). To determine whether C. elegans thyrostimulin can activate FSHR-1, we used a cAMP-based receptor activation assay in human embryonic kidney (HEK) cells. We co-expressed FSHR-1 with a cAMP response element (CRE)-luciferase reporter in HEK cells and quantified its activation by assessing bioluminescence levels in the presence of luciferin substrate ( Figure 2A ). Consistent with previous reports (48), we observed basal activity of FSHR-1, as bioluminescent cAMP signals were higher in receptor-expressing cells than in cells transfected with an empty control vector ( Figure S4A ). However, challenging cells with recombinant GPLA-1 and GPLB-1 proteins significantly increased cAMP signaling by FSHR-1 ( Figures 2B–D ). To generate recombinant GPLA-1/GPLB-1, HEK293 cells were first transfected with expression vectors encoding His-tagged GPLA-1 and non-tagged GPLB-1. Tagging of GPLA-1 did not affect protein function as transgenic C. elegans expressing endogenous His-GPLA-1 showed normal body size ( Figure S4B ). Recombinant proteins purified from cells expressing both subunits significantly increased cAMP signaling by FSHR-1, suggesting that GPLA-1/GPLB-1 is a ligand of FSHR-1 ( Figure 2B ). We also tested whether GPLA-1 and GPLB-1 proteins could activate FSHR-1 individually. For this, recombinant proteins were purified from HEK293 cells transfected with an expression vector encoding only one of the two subunits. We found that GPLA-1 as well as GPLB-1 increased cAMP signaling by FSHR-1 ( Figures 2C, D ), which indicates that each subunit alone is capable of activating the receptor. None of the purified recombinant proteins elicited a significant cAMP response in cells transfected with an empty control vector instead of the FSHR-1 expression plasmid ( Figure S4C ). Taken together, these results suggest that GPLA-1 and GPLB-1 are cognate ligands of FSHR-1.

As FSHR-1 is activated by GPLA-1 and GPLB-1 in vitro, we asked if this receptor is involved in growth regulation. We observed that a loss-of-function mutant of fshr-1 suffers from severe growth defects that persist throughout adulthood ( Figures 2E , S3A ). Expressing wild-type fshr-1 under the control of its endogenous promoter fully restores this growth defect, indicating that this receptor, like its activating ligands GPLA-1 and GPLB-1, is required for growth ( Figure 2E ). Although body length was more strongly reduced in fshr-1 animals than in mutants lacking the thyrostimulin subunits, double and triple mutants of gpla-1, gplb-1 and their receptor fshr-1 were not shorter than the fshr-1 single mutant ( Figure 2F ). This suggests they all act in the same growth-regulating pathway in vivo, consistent with our finding that FSHR-1 is a receptor for GPLA-1 and GPLB-1 in vitro.

To gain further insight into the role of thyrostimulin-like signaling in growth regulation, we investigated the expression patterns of gpla-1, gplb-1 and fshr-1. Fluorescent reporter transgenes, which rescue body length defects ( Figures 1B, C , 2E ), showed expression in multiple pharyngeal neurons and tissues associated with feeding and digestive processes. Consistent with single-cell RNA sequencing data and previous reporter analyses (15, 49, 50), we found that gpla-1 is predominantly expressed in neurons associated with the gastrointestinal tract ( Figures 3A, B ), which are part of the enteric nervous system in C. elegans (50–52). These include the pharyngeal neurons M1, M5, I5 and NSM, as well as the excitatory motor neurons AVL and DVB that control enteric muscle contractions in the hindgut during defecation (53–55). We also observed expression of gplb-1 in the enteric DVB neuron ( Figure 3D ). In addition, both gpla-1 and gplb-1 reporter transgenes show fluorescence in the RME motor neurons that innervate the head muscles ( Figures 3A, C ). Using a reporter transgene for GABAergic neurons, we confirmed that gpla-1 and gplb-1 are both expressed in the GABAergic DVB and RME neurons ( Figure S5 ). In addition, the gplb-1 reporter transgene shows expression in non-neuronal tissues, including the head mesodermal cell (hmc) and the enteric muscles of the hindgut, which are involved in defecation (53–57). Taken together, these results indicate that expression of gpla-1 and gplb-1 co-localizes in the RME neurons in the head and the enteric neuron DVB in the tail, whereas no overlapping expression is seen in the pharyngeal neurons (gpla-1) or hmc and the enteric muscles (gplb-1).

A reporter construct expressing fshr-1 under its endogenous promoter recapitulated the reported expression of the receptor in the intestine and in multiple neurons in the head (49, 58–60). In addition, we observed expression of fshr-1 in several cells close to the anterior pharyngeal bulbus that, based on position and morphology of their projections, are most likely glial cells ( Figures 3E, F ). Together, these expression patterns suggest that thyrostimulin-like signaling may influence body size by controlling intestinal or neural function.

Next, we asked in which tissues fshr-1 is required for normal body size. Expressing wild-type fshr-1 under control of the intestine-specific ges-1 (61–63) or the pan-glial mir-228 (64) promoters fully rescued the body size defect of fshr-1 mutants ( Figure 4A ). By contrast, pan-neuronal expression using the rab-3 (65, 66) promoter partially restored body size ( Figure 4A ). Since additional expression of this promoter has been reported in the intestine (67, 68), we generated a second transgene for pan-neuronal expression of wild-type fshr-1 using the rgef-1 promoter (69). This transgene did not restore the body size defect of fshr-1 mutants ( Figure 4A ); however, it remains possible that fshr-1 expression levels under control of the rgef-1 promoter were not sufficient to rescue body size. Taken together, our results show that FSHR-1 regulates growth in the intestine and glial cells, although we cannot exclude that FSHR-1 signaling in neurons may influence body size as well.

Our expression analysis and rescue experiments suggest that thyrostimulin-like signaling regulates intestinal function. To test this hypothesis, we investigated the anatomy of the intestine and measured the size of the intestinal lumen. The lumen of the intestine was bloated to a similar extent in single and double mutants of gpla-1 and gplb-1 ( Figures 4B, C ). The bloating phenotype was fully rescued by expressing wild-type copies of gpla-1 and partially restored by expressing wild-type gplb-1 ( Figures 4D, E ). Mutants lacking fshr-1 also displayed a severely bloated intestinal lumen, which was rescued by restoring wild-type fshr-1 expression under its endogenous promoter ( Figures 4B, F ). The lumen of double and triple mutants of gpla-1, gplb-1 and their receptor fshr-1 was not wider than that of the fshr-1 single mutant ( Figures S6D–F ), suggesting that they act in the same pathway controlling distention of the intestinal lumen. Tissue-specific expression of wild-type fshr-1 in either the intestine or glial cells fully rescued luminal bloating, whereas pan-neuronal expression under control of the rab-3 or rgef-1 promoter partially restored the luminal size ( Figure 4F ). Distention of the intestinal lumen is thus regulated by fshr-1 in the intestinal and glial cells, and in part by FSHR-1 signaling in neurons.

Since gpla-1 and gplb-1 are predominantly expressed in enteric neurons and tissues controlling the defecation motor program, we hypothesized that thyrostimulin signaling may influence growth by regulating defecation behavior. To test this, we first examined whether mutants defective in the defecation motor program also show growth defects. In C. elegans, the defecation motor program is a rhythmic behavior with a period of about 50 seconds, which ultimately results in contraction of the enteric muscles in the hindgut and the subsequent expulsion of digested food from the intestine (53–55). Mutants for unc-25 and exp-1 are deficient in synaptic GABAergic transmission, which drives enteric muscle contraction, and show severe defecation defects (70–72). We found that unc-25(e156) and exp-1(ox276) mutants are also significantly smaller than wild-type animals ( Figure S6A ), suggesting that defecation defects correlate with impaired growth in these mutants. Next, we asked whether mutants defective in thyrostimulin-like signaling show defects in defecation behavior. We found that the contraction frequency of the anterior body wall muscles (aBoc) did not differ from wild-type aBoc frequency ( Figure S6B ). However, the average length of the defecation cycle between consecutive contractions of the posterior body wall muscles (pBoc) was significantly longer in mutants for gpla-1, gplb-1, and fshr-1 compared to the cycle length of wild-type animals ( Figure S6C ). Taken together, these findings suggest that thyrostimulin-like neuroendocrine signaling regulates growth by controlling intestinal function.

In vertebrates, glycoprotein hormone signaling is regulated by hypothalamic factors that control the release of hormones from the pituitary (2, 73). TRH is a highly conserved releasing factor that regulates growth and metabolism across bilaterian animals (2, 43, 73–75). In C. elegans TRH-related neuropeptides promote growth and mutants lacking the TRH-like precursor gene trh-1 have a smaller body size than wild-type animals (43). We therefore asked whether TRH- and thyrostimulin-like signaling systems may interact to control C. elegans growth.

To test this hypothesis, we compared the body size of single and double mutants lacking trh-1 and the thyrostimulin subunit orthologs gpla-1 and gplb-1. Double mutants defective in both signaling systems did not significantly differ in length compared to the single mutants ( Figure 5A ). Similar to thyrostimulin mutants, worms lacking TRH-1 also displayed intestinal bloating ( Figures 5B, C ). Although the effect is more severe in mutants lacking thyrostimulin-like signaling, the extent of luminal bloating in double mutants of gpla-1, gplb-1 and trh-1 did not significantly differ from single gpla-1 or gplb-1 mutants ( Figures 5D, E ). These results indicate that trh-1 and gpla-1/gplb-1 act in the same growth-stimulating pathway in vivo.

All C. elegans strains were maintained at 20°C on nematode growth medium (NGM) plates seeded with Escherichia coli OP50 bacteria. All experiments were performed using 1-day adult hermaphrodites, unless mentioned otherwise. Wild-type (N2-Bristol), IBE7 flr-2 (ut5), IBE24 flr-2 (ok2127), IBE1 fshr-1 (ok778) and EG1285 lin-15B&lin-15A (n765) oxIs12 [unc-47p::GFP + lin-15(+)] strains were obtained from the Caenorhabditis Genetics Center (CGC, University of Minnesota). CB156 unc-25 (e156) and EG276 exp-1 (ox276) strains were a kind gift from the lab of W.R. Schafer (Laboratory of Molecular Biology, Cambridge, UK). Deletion alleles for gpla-1 (ibt1) and gplb-1 (ibt4) were obtained by CRISPR/Cas9 genome editing, as described below. A full list of strains used in this study can be found in Table S1 .

All fluorescent reporter and rescue constructs were made using the MultiSite Gateway Three-Fragment cloning system (Invitrogen). Genomic DNA or cDNA sequences of fshr-1 (a isoform), gpla-1, and gplb-1 (a isoform) were cloned in front of a SL2 trans-splicing site and fluorescent reporter gene.

Putative promoter sequences for fshr-1 (3110 bp), gpla-1 (4058 bp), and gplb-1 (3589 bp) were cloned from genomic DNA of wild-type C. elegans. Tissue-specific rescue transgenes were generated using promoter regions of rab-3 (1208 bp) (65, 66), rgef-1 (3660 bp) (69), ges-1 (3313 bp, kind gift from the lab of W.R. Schafer, Cambridge, UK) (61–63), and mir-228 (2217 bp) (64). All constructs have a let-858 3’UTR except the rgef-1 rescue construct, which was made using a tbb-2 3’UTR.

The plasmid for heterologous expression of fshr-1 in HEK cells was obtained by directionally cloning the fshr-1 cDNA into the pcDNA3.1/V5-His-TOPO vector (Invitrogen). The cDNA sequence of the fshr-1a gene isoform was amplified by PCR using cDNA from mix-staged wild-type C. elegans as template. The forward primer included a ‘CACC’ sequence at the 5’ end that introduced a partial Kozak sequence for increased translation efficiency in mammalian cells.

The gpla-1(ibt1) and gplb-1(ibt4) knockout alleles were generated by CRISPR/Cas9-mediated deletion of the gpla-1 or gplb-1 open reading frames. For each gene, two crRNA sequences were designed by looking for PAM sites (NGG) that were in the vicinity of the double stranded break site. After analysis of putative off-target sites (http://crispor.tefor.net/), we selected the highest scoring sequences based on their predicted on-target activities. Repair templates were designed to be in-frame and were codon-optimized for C. elegans (https://www.genscript.com/tools/codon-frequency-table). To simplify selection of the successfully edited worms, we used a Co-CRISPR technique in which the dpy-10 gene was also knocked out by CRISPR editing (108). We used the CRISPR/Cas9 protocol as described by Mello et al. (108) for the preparation of injection mixes containing all required components ( Table S2 ). After injection of young adult hermaphrodites, F1 progeny that showed the rolling and/or dumpy phenotype were transferred to individual NGM plates and checked for gene deletion by PCR and subsequent gel-electrophoresis. Successful deletion of the genes was confirmed by sequencing.

The gpla-1 (ibt3 [His::GPLA-1]) knockin allele was made by CRISPR/Cas9 mediated insertion of a His-tag at the N-terminus of the gpla-1 open reading frame, using a similar strategy as described above and the crRNA and repair template listed in Table S2 .

Below, the sequences of the gene knockouts and knockins are shown with flanks on the (-) strand. Exons are in grey, with start and stop codons in red, the remaining gene sequences after CRISPR gene editing are underlined, and with the inserted gene sequence in blue.

gpla-1(ibt1) deletion:

ttctttaaattctatttttaacttttcagcctataccttctcaaatgATGGGCTCCAAAGCACGAGCACGACGACGTTTAAGTTGTTTTTTAAGCGTTTTTGTTGTGACATGCTTATTACAGTACTGCACAGCAGGTGTTACTAAGAATAATAGTTGCAAAAAAGTTGgtacgtcacgaaatctactaaacttcgatcagtgtcctttaaatttttttttcagGAGTGGAGGAACTTATAGATGAAGAAGGCTGTGATTTGATGATAATTCGAATCAATCGATGCAGTGGGCATTGCTTCTCATTTACATTTCCTAATCCCTTAACGAAAAAATATTCAGTGCATGCGAAGTGCTGCCGGATGGTTGAATGGGAAATGgttagtattttttaactacagaaatgacttctgaaaatttataacaaagttattacggaagccaaaattctgggaatgtgttttgcgcaacatgtaaaaaaaatctcgtagcgaaagctacagtaattctctaaataactactgtagtgcttgcgtcgaattacggacttgatttgcgatatatccttcgtttctctgtattactttctcatttttgttttttttttaaacattctatcgaaaaattgatgattaattcatttcgaaagccgagcccgtaaatcgtcacaagcgctacagtagtcatgtaaagaattactgtatcgctacgagatgttttaatattgttttccaagaatcaatttttatttttcagCTTGAAACAGAATTAAAATGTTCCAAAGGAAACCGAAATCTTCGAATACCATCTGCAACACAATGTGAATGTTTTGATTGTCTTGTTCGATAGttgcagtttccatctctttcatttctcattcttgaatctcttgagttttacctgcataatagatttaatttattttctctcctccagtgccacattcatccacatttccataaaatctcgttcccttaattatttctataataatctcgttgatacgggcatctaaaaatgcaaaaaatcaaagcgaaagaaggatgactgacataaattgtctcaa

gplb-1(ibt4) deletion:

tgttgacgttaagctcttttgaaaaaaATGCTTATATTTCCCGTGATCACTATACTTCATATATTTTTGgtaactttttgaaatatcttctaaaatgaatttaatatgtaacaatattctaacaaaatttttaatattgatcagatagtttttccaaatttaagtattaaccaggcgcgaaattttccgaattttaggccaaaaatacggtgcccggtctcgacacgaatttttttattaggtaaaaatgggtgtgtgcctttaaagagtactgtaactttaaactttcgttgctgcggaacttttgtcgacttttcatagctattagataaaaataaaaaaatattcaatattttcaacaaatctttagaaaaactatgtaaaatcgataaaaattctgcaacaaaaatttgaagttacagtactctttaaaggcacacactcgattgtatttaacaaaaaaagtcgtgtcgagaccggttaccgtaatttttgcgcaaatcggaataatttcgcgcttgggtaataagcatcacatctccaactaatttaaaagcaaaagtgtgatttttaaattcagATTTCGGTTGAATCTGGAAAAGAATGCGAGTTTGCAATGCGATTGGTCCCAGGATTCAATCCACTTCGTCAAGTTGATGCAAATGGAAAAGAATGCCGAGGAAACGTGGAATTGCCATTTTGCAAGGGTTACTGTAAGACTAGCGAGgtgaatttccttttttttccgattcaaaaataatccaattaaatttcagAGTGGCACCCATGGCTTTCCACCACGAGTTCAAAATAGTAAAGTGTGCACATTGGTCACCACTTCAACTCGAAAAGTAGTTCTTGATGATTGTGATGATGGAGCCGATGAGAGTGTCAAGTTTGTAATGGTTCCACATGGAACTGATTGTGAATGTTCTGCAGTTCCACTTGAACAACATCATTCATAAattatcatacattcattcaaaaattcatcgaataaataaaagttttgtg

gpla-1(ibt3 [His::GPLA-1]) insertion:

ttctttaaattctatttttaacttttcagcctataccttctcaaatgATGCATCACCATCACCATCACGGCTCCAAAGCACGAGCACGACGACGTTTAAGTTGTTTTTTAAGCGTTTTTGTTGTGACATGCTTATTACAGTACTGCACAGCAGGTGTTACTAAGAATAATAGTTGCAAAAAAGTTGgtacgtcacgaaatctactaaacttcgatcagtgektcctttaaatttttttttcagGAGTGGAGGAACTTATAGATGAAGAAGGCTGTGATTTGATGATAATTCGAATCAATCGATGCAGTGGGCATTGCTTCTCATTTACATTTCCTAATCCCTTAACGAAAAAATATTCAGTGCATGCGAAGTGCTGCCGGATGGTTGAATGGGAAATGgttagtattttttaactacagaaatgacttctgaaaatttataacaaagttattacggaagccaaaattctgggaatgtgttttgcgcaacatgtaaaaaaaatctcgtagcgaaagctacagtaattctctaaataactactgtagtgcttgcgtcgaattacggacttgatttgcgatatatccttcgtttctctgtattactttctcatttttgttttttttttaaacattctatcgaaaaattgatgattaattcatttcgaagccgagcccgtaaatcgtcacaagcgctacagtagtcatgtaaagaattactgtatcgctacgagatgttttaatattgttttccaagaatcaatttttatttttcagCTTGAAACAGAATTAAAATGTTCCAAAGGAAACCGAAATCTTCGAATACCATCTGCAACACAATGTGAATGTTTTGATTGTCTTGTTCGATAGttgcagtttccatctctttcatttctcattcttgaatctcttgagttttacctgcataatagatttaatttattttctctcctccagtgccacattcatccacatttccataaaatctcgttcccttaattatttctataataatctcgttgatacgggcatctaaaaatgcaaaaaatcaaagcgaaagaaggatgactgacataaattgtctcaa

Germline transformations were carried out by injecting constructs into the syncytial gonad of young adult worms together with a co-injection marker (myo-2p::mCherry, unc-122p::dsRed, or unc-122p::GFP) and 1-kb DNA ladder (Thermo Scientific) as carrier DNA.

Expression patterns of fluorescent reporter transgenes were visualized by an Olympus confocal microscope (FluoView FV1000, IX81) and resulting Z-stack projections were created and analyzed with ImarisViewer (v9.7.2) software. Adult worms were immobilized by mounting in 50 mM Sodium Azide solution in M9 buffer on a 2% agarose pad and covered with a glass cover slip. Expression patterns were confirmed in at least two independent transgenic strains. DVB and RME expression was confirmed by crossing with marker strain EG1285, which marks GABAergic neurons (109). Other cells expressing gpla-1, gplb-1, or fshr-1 were identified based on their position and morphology.

Body size parameters (length, width, volume) of day one adults were measured using the custom WormSizer MATLAB script (https://github.com/jwatteyne/WormSizer). Synchronized L1 larvae were placed onto freshly seeded NGM plates one day post-synchronization and kept at 20°C until assaying. After 65 hours on food, 20 to 30 well-fed day one adults were transferred from the plates and anesthetized in 20 µL of 10 mM tetramisole hydrochloride solution (Sigma-Aldrich) in Milli-Q water on a 2% agarose pad. Images were captured with a ZEISS Axio Observer.Z1 at 5x magnification. Body size was measured from images of anesthetized day 1 adult worms on at least two independent days. After skeletonization, delineating the worm’s outline and midline, the worm’s total length and middle width were calculated using the calibration factor (pixels/µm) of the pictures. For the volume estimation, the midline of the worm was separated in 30 segments of the same size. The volume for each segment was determined by using the formula for the frustum of a cone, taking the natural shape of a worm into consideration. The summation of all these volumes was described as the worm’s total volume. To assess growth throughout adulthood, body size measurements were executed for 5 subsequent days (day 1 to day 5 adulthood) for at least two independent time periods.

Intestinal lumen diameter was quantified from images taken for body size measurements of day 1 adult worms. Using the ImageJ software (version 1.53s), luminal width was determined by measuring the average width (in µm) of the intestine at two different points in the posterior lumen. Body size parameters and luminal widths were plotted, and significance levels were calculated with GraphPad Prism 8 software.

Recombinant proteins were synthesized and purified by OriGene™ Technologies (Rockville, USA). To generate recombinant GPLA-1/GPLB-1, HEK293 cells were transfected with expression vectors encoding C. elegans N-terminal 6xHis-tagged GPLA-1 and non-tagged GPLB-1. Recombinant proteins were purified from cell lysate by affinity chromatography in PBS buffer (PBS and 10% glycerol, pH 7.3) and analyzed by Tricine-SDS-PAGE on a 8 – 20% gel in Tris-Glycine running buffer and by subsequent Western blot using anti-His monoclonal antibodies. A similar approach was used to synthesize individual GPLA-1 and GPLB-1 proteins. N-terminal 6xHis-tagged GPLA-1 was expressed in HEK293 cells and purified by affinity chromatography. Non-tagged GPLB-1 was expressed in HEK cells and purified using ion exchange chromatography. Both recombinant GPLA-1 and GPLB-1 protein samples were analyzed by Tricine-SDS-PAGE.

The cAMP bioluminescence assay quantifies ligand-induced bioluminescent responses by measuring changes in intracellular cAMP levels after receptor activation. HEK293T cells were cultured in monolayer at 37°C in a humidified atmosphere with 5% CO₂, in Dulbecco’s Modified Eagle’s Medium (DMEM)/Nutrient F-12 Ham (Gibco) supplemented with 10% heat inactivated Fetal Bovine Serum (FBS, Sigma-Aldrich) and 1% Penicillin/Streptomycin (Gibco). Co-transfection with the cAMP indicator CRE(6x)-luciferase and pcDNA3.1/fshr-1a was done in a 1:1 ratio when cells reached 60-70% confluency. Transfection medium contained Opti-MEM (Gibco, Life Technologies), CRE(6x)-luciferase and receptor plasmid DNA, Plus Reagent (Invitrogen) and Lipofectamine LTX (Invitrogen). One day post-transfection, fresh culture medium was added to the transfected cells. Two days post transfection, each well of a Bio-One CELLSTAR™ 96-well, Flat Bottom Microplate (Greiner) was loaded with a dilution series of recombinant protein compounds in 200 µM 3-isobutyl-1-methylxanthine (IBMX), which inhibits cAMP hydrolysis. Compound buffer (PBS with 10% glycerol) without hormone was added to the IBMX medium as a ligand-free negative control. Cells were detached, counted, pelleted and resuspended to a concentration of 10E-06 cells/mL in IBMX medium. Each well of the compound plate was supplemented with 50 µL of the cell suspension (50 000 cells/well) and the plate was incubated at 37°C for 3.5 h. Cells were then loaded with 100 µL SteadyLitePlus substrate (PerkinElmer) and incubated on a shaking plate for 15 minutes under dark conditions at RT. Finally, luminescence was measured twice for 5s (at 0s and 5s) per well at 469 nm on a Mithras LB 940 luminometer (Berthold Technologies). Measurements were performed in triplicate in at least 4 independent experiments. Luminescence values were plotted and significance levels were calculated with GraphPad Prism 8 software.

The defecation motor program (DMP) was analyzed as previously described (93). Animals were synchronized by letting day one adults lay eggs for 2-3 hours, before removing them from the NGM plates, and allowing the eggs to develop into day one adults. One day prior to the DMP assay, L4 animals were picked onto a freshly seeded NGM plate and all strains were blinded. After 15-17 hours, worms were acclimated on the microscope stage for ~10 minutes. At least 10 consecutive defecation cycles were observed from each animal on a Nikon stereomicroscope. Cycle length, aBoc and expulsion events were scored, and defecation behavior was logged using BORIS (Behavioral Observation Research Interactive Software). Animals were assayed alongside wild type and at least one defecation-defective control. The custom DefecationAnalysis R script was applied to measure the defecation parameters of day one adults (https://github.com/NathanDeFruyt/DefecationAnalysisBeetsLab). The length of one DMP cycle was defined by the time elapsed between two subsequent contractions of the posterior body wall muscles (pBocs). The average cycle length was calculated over 10 defecation cycles for each worm. aBoc frequency was defined as the ratio of aBoc over pBoc. Defecation parameters were plotted and significance levels were calculated with GraphPad Prism 8 software.

The H. Sapiens GPA2 and GPB5 and C. elegans GPLA-1 and GPLB-1 sequences were used as queries to identify sequences with highest similarity in selected model systems representative for their phylum by the Basic Local Alignment Search Tool (BLAST) of the National Center for Biotechnology Information (NCBI) database. The multiple sequence alignment program Clustal Omega (https://www.ebi.ac.uk/Tools/msa/clustalo/) was used to align sequences, and conserved residues were identified using BoxShade (https://embnet.vital-it.ch/software/BOX_form.html).