Lysosome lipid signalling from the periphery to neurons regulates longevity

Abstract

Lysosomes are key cellular organelles that metabolize extra- and intracellular substrates. Alterations in lysosomal metabolism are implicated in ageing-associated metabolic and neurodegenerative diseases. However, how lysosomal metabolism actively coordinates the metabolic and nervous systems to regulate ageing remains unclear. Here we report a fat-to-neuron lipid signalling pathway induced by lysosomal metabolism and its longevity-promoting role in Caenorhabditis elegans. We discovered that induced lysosomal lipolysis in peripheral fat storage tissue upregulates the neuropeptide signalling pathway in the nervous system to promote longevity. This cell-non-autonomous regulation is mediated by a specific polyunsaturated fatty acid, dihomo-γ-linolenic acid, and LBP-3 lipid chaperone protein transported from the fat storage tissue to neurons. LBP-3 binds to dihomo-γ-linolenic acid, and acts through NHR-49 nuclear receptor and NLP-11 neuropeptide in neurons to extend lifespan. These results reveal lysosomes as a signalling hub to coordinate metabolism and ageing, and lysosomal signalling mediated inter-tissue communication in promoting longevity.

Fig. 1. Peripheral lysosomal lipolysis upregulates neuropeptide signalling.

a, Transgenic strains constitutively expressing lipl-4 in the intestine (lipl-4 Tg) show lifespan extension. b, Gene Ontology of upregulated genes in lipl-4 Tg compared with WT worms. c, Genes encoding neuropeptide-processing enzymes egl-3, egl-21, sbt-1, pgal-1 and pghm-1 are transcriptionally upregulated by lipl-4 Tg. d, Schematic diagram of the neuropeptide processing and maturation pathway. e, List of significantly upregulated neuropeptide genes (P < 0.05) by lipl-4 Tg in RNA-seq transcriptome profiling. f, egl-21 mRNA is detected by smFISH (Red Quasar 670), while nuclei are stained by DAPI. egl-21 is detected in neurons (boxed region) of WT but not egl-21(lf) worms. The intestine region marked by dashed lines shows no egl-21. Scale bar, 30 µm and 10 µm in the inset. g, GFP driven by the egl-21 endogenous promoter is expressed in neurons but not the intestine (marked by dashed lines). Scale bar, 10 µm. h, The loss-of-function mutation of egl-21(lf) suppresses lipl-4 Tg longevity. daf-16 RNAi knockdown is used to eliminate the influence from ILP reduction in egl-21(lf). In a and h, n = 3 biologically independent samples; NS, not significant (P > 0.05), ***P < 0.001 by log-rank test, 60–120 worms per replicate. See Supplementary Tables 4 and 5 for full lifespan data. In c, error bars represent mean ± standard error of the mean (s.e.m.), n = 3 biologically independent samples, ****P < 0.0001 by two-tailed Student’s t-test, ~2,000 worms per replicate. In e, error bars represent mean ± s.e.m., n = 3 biologically independent samples, DESeq2 |fold change| ≥1.5; P < 0.05 by two-sided Wald test (lipl-4 Tg versus WT), ~3,000 worms per replicate. Source numerical data are available in source data. Source data

Fig. 2. NLP-11 neuropeptide acts in neurons to promote longevity.

a, Knockdown of nlp-11 in a neuronal RNAi-sensitive background suppresses lipl-4 Tg longevity. b, The loss-of-function mutation of nlp-11(lf) suppresses lipl-4 Tg longevity. c, Constitutive expression of nlp-11 driven by its endogenous promoter extends lifespan. d, RNAi knockdown of nlp-11 selectively in the intestine shows no suppression of lipl-4 Tg longevity. e,f, Neuron-specific overexpression of nlp-11 prolongs lifespan (e), but intestine-specific overexpression has no such effect (f). In a–f, n = 3 biologically independent samples, ***P < 0.001 by log-rank test, 60–120 worms per replicate, EV, empty vector control for RNAi. Lifespan of one transgenic strain (#1) is shown, and the others are in Extended Data Fig. 2. See Supplementary Tables 4 and 5 for full lifespan data.

Fig. 3. Lysosome-derived PUFAs in the periphery regulate neuropeptide and longevity.

a, Relative levels of FFAs, saturated fatty acids (SFAs), MUFAs and PUFAs are quantified by liquid chromatography–mass spectrometry in lipl-4 Tg versus WT. b, Percentage of lipid species showing significant changes (P < 0.05) in the lysosome purified from lipl-4 Tg versus WT worms. Out of 186 TAG species decreased in lipl-4 Tg, 117 contain PUFAs. Sphingophospholipids (SPH), phosphatidylcholines (PC), lysophosphatidylcholines (LPC), phosphatidylethanolamines (PE), lysophosphatidylethanolamines (LPE), phosphatidylserines (PS), lysophosphatidylserines (LPS), phosphatidylinositols (PI), diacylglycerols (DAG) and TAG. c, Schematic diagram of PUFA biosynthesis in C. elegans. PUFAs enriched in lipl-4 Tg are highlighted in blue, while desaturase enzymes are marked in green. d, The loss-of-function mutation of the fatty acid desaturase fat-1(lf) or fat-3(lf) suppresses the transcriptional upregulation of egl-3 and egl-21 neuropeptide genes by lipl-4 Tg. e,f, RNAi knockdown of either fat-1 (e) or fat-3 (f) selectively in the intestine suppresses lipl-4 Tg longevity. In a, error bars represent mean ± s.e.m., n = 6 (WT) and n = 9 (lipl-4 Tg) biologically independent samples, *P = 0.028 for C18:4n3, P = 0.023 for C14:0 and P = 0.026 for C15:0, **P = 0.007 for C18:2n6 and P = 0.002 for C18:3n3/6, ****P < 0.0001 for C16:1n7, C20:2n6, C20:3n6 and C20:4n6 by two-way ANOVA with Holm–Sidak correction (lipl-4 Tg versus WT), ~40,000 worms per replicate. In d, error bars represent mean ± s.e.m., n = 4 (WT, fat-1(lf), fat-3(lf) and lipl-4 Tg;fat-1(lf)) and n = 3 (lipl-4 Tg and lipl-4 Tg;fat-3(lf)) biologically independent samples, ***P < 0.001 by two-way ANOVA with Holm–Sidak correction, ~3,000 worms per replicate. In e and f, n = 3 biologically independent samples, ***P < 0.001 by log-rank test, 60–120 worms per replicate. See Supplementary Table 5 for full lifespan data. Source numerical data are available in source data. Source data

Fig. 4. Peripheral lipid chaperone LBP-3 regulates neuropeptide and longevity.

a, The loss-of-function mutation of the lipid chaperone lbp-3(lf) but not lbp-2 (lf) suppresses the transcriptional up-regulation of egl-3 and egl-21 by lipl-4 Tg. b, lbp-3(lf) suppresses lipl-4 Tg longevity. c, Constitutive expression of lbp-3 driven by its own endogenous promoter (lbp-3 Tg) prolongs lifespan. d, The transcriptional levels of egl-3, egl-21 and nlp-11 are induced in the lbp-3 Tg worms. e, Out of 39 upregulated neuropeptide genes by lipl-4 Tg, 22 are also significantly induced by lbp-3 Tg (P < 0.05). f, The loss-of-function mutation of egl-21(lf) suppresses lbp-3 Tg longevity. daf-16 RNAi knockdown is used to eliminate the influence from ILP reduction in egl-21(lf). g, The loss-of-function mutation of nlp-11(lf) suppresses lbp-3 Tg longevity. h, Intestine-specific lbp-3 overexpression upregulates the transcriptional levels of egl-3, egl-21 and nlp-11. i, Overexpression of lbp-3 selectively in the intestine extends lifespan. In a, error bars represent mean ± s.e.m., n = 3 (lbp-2(lf), lipl-4 Tg and lipl-4 Tg;lbp-2(lf)) and n = 6 (WT, lbp-3(lf) and lipl-4 Tg;lbp-3(lf)) biologically independent samples, ***P = 0.0001 by two-way ANOVA with Holm–Sidak correction, ~2,000 worms per replicate. In b, c, f, g and i, n = 3 biologically independent samples, ***P < 0.001 by log-rank test, 60–120 worms per replicate. Lifespan of one transgenic strain (#1) is shown and the others are in Extended Data Fig. 4. See Supplementary Tables 4 and 5 for full lifespan data. In d, error bars represent mean ± s.e.m., n = 5 (WT and lbp-3 Tg for egl-3 and nlp-11) and n = 3 (WT and lbp-3 Tg for egl-21) biologically independent samples, **P = 0.007 for egl-3, **P = 0.007 for nlp-11 and ***P = 0.0002 for egl-21 by two-way ANOVA with Holm–Sidak correction, ~2,000 worms per replicate. In h, error bars represent mean ± s.e.m., n = 4 biologically independent samples, *P = 0.0497, ****P < 0.0001 by two-way ANOVA with Holm–Sidak correction, ~2,000 worms per replicate. Source numerical data are available in source data. Source data

Fig. 5. LBP-3 secreted from the periphery regulates neuropeptide and longevity.

a, The gene expression of lbp-3 is indicated by polycistronic GFP, while the LBP-3 protein is visualized by its RFP fusion. Secreted LBP-3–RFP fusion is detected in coelomocytes marked with arrowheads. Scale bars, 30 µm and 10 µm in the inset. b,c, The level of secreted LBP-3–RFP fusion is increased by lipl-4 Tg. Secreted LBP-3–RFP fusion proteins in coelomocytes are marked by arrowheads (b), and their levels are quantified in lipl-4 Tg versus WT (c). Scale bar, 10 µm. Representative images from three biological repeats. d, Intestine-specific overexpression of lbp-3ns, which does not carry the secretory signal sequence, decreases the expression of the neuropeptide genes. e, Intestine-specific overexpression of lbp-3ns fails to extend lifespan. f, RFP fusion of the LBP-3ns protein is not detected in coelomocytes, marked by arrowheads. Scale bar, 10 µm. In c, error bars represent mean ± s.e.m., n = 11 (WT) and n = 15 (lipl-4 Tg) biologically independent samples, ****P < 0.0001 by two-tailed Student’s t-test. In d, error bars represent mean ± s.e.m., n = 4 biologically independent samples, ****P < 0.0001 by two-way ANOVA with Holm–Sidak correction ~2,000 worms per replicate. In e, n = 3 biologically independent samples; NS, P > 0.05 by log-rank test, 75–100 worms per replicate. Lifespan of one transgenic strain in shown, and the others are in Extended Data Fig. 5. See Supplementary Table 4 for full lifespan data. Source numerical data are available in source data. Source data

Fig. 6. DGLA regulates LBP-3 secretion, neuropeptide and longevity.

a,b, In WT conditions, LBP-3–RFP secretion is not affected by fat-3 RNAi knockdown. However, the increased secretion of LBP-3–RFP fusion by lipl-4 Tg is suppressed by fat-3 RNAi knockdown. Secreted LBP-3–RFP fusion proteins in coelomocytes are marked by arrowheads (a), and their levels are quantified (b). Scale bar, 10 µm. Representative images from three biological repeats (a). c, Fluorescence signals derived from 1,8-ANS bound to LBP-3 are decreased, when 1,8-ANS is outcompeted by the increasing amount (M, molar) of AA, DGLA and EPA, but not ETA. d, The supplementation of DGLA but not AA or EPA restores the transcriptional induction of egl-3 and egl-21 by lipl-4 Tg in the fat-3(lf) mutant. DMSO serves as the vehicle control. e,f, DGLA supplementation restores LBP-3–RFP secretion in lipl-4 Tg with fat-3 RNAi knockdown. Secreted LBP-3–RFP proteins in coelomocytes are marked by arrowheads (e), and their levels are quantified (f). Scale bar, 10 µm. Representative images from three biological repeats (e). g, DGLA supplementation restores the lifespan extension in lipl-4 Tg with fat-3 RNAi knockdown. In b, error bars represent mean ± s.e.m., n = 10 biologically independent samples, ***P < 0.001 and ****P < 0.0001 by one-way ANOVA with Holm–Sidak correction. In c, n = 16 biological replicates, one-site Fit Ki, R² = 0.9945 for AA, R² = 0.9954 for DGLA, R² = 0.9903 for EPA and R² = 0.04758 for ETA. In d, error bars represent mean ± s.e.m., n = 5 (lipl-4 Tg;fat-3(lf) on EV, AA and EPA) and n = 8 (lipl-4 Tg;fat-3(lf) on DGLA) biologically independent samples, **P = 0.009 for AA and P = 0.008 for DGLA, ****P < 0.0001 by two-way ANOVA with Holm–Sidak correction, ~2,000 worms per replicate. In f, error bars represent mean ± s.e.m., n = 12 biologically independent samples, ****P < 0.0001 by one-way ANOVA with Holm–Sidak correction. In g, n = 3 biologically independent samples, P < 0.001 by log-rank test, 80–100 worms per replicate. See Supplementary Table 5 for full lifespan data. Source numerical data are available in source data. Source data

Fig. 7. DGLA binding specificity of LBP-3 mediates its effects.

a, DGLA supplementation fails to restore the induction of egl-3 and egl-21 by lipl-4 Tg in the fat-3(lf) and lbp-3(lf) double-mutant background. DMSO serves as the vehicle control. b, Constitutive expression of lbp-2 (lbp-2 Tg) does not affect the transcription of egl-3 or egl-21. c, After incubation with C. elegans liposome, fatty acids bound to LBP-2 or LBP-3 proteins were analysed with mass spectrometry. DGLA shows a high percentage of occupancy in LBP-3 but not LBP-2. d, In the C. elegans liposome, the percentage of EPA is more than ten times higher than that of DGLA or AA. e, LBP-2 (blue) and LBP-3 (grey) superposition structures predicted using AlphaFold2. f, An LBP-2 and LBP-3 protein alignment generated using t-coffee. Secondary structures are displayed above the alignment. The LBP-2 sequence utilized for the replacement in the LBP-3 chimeric protein is highlighted in purple. g,h, Intestine-specific overexpression of chimeric lbp-3(chim) does not affect the transcription of egl-3 and egl-21 (g) nor prolongs lifespan (h). In a and b, error bars represent mean ± s.e.m., n = 3 biologically independent samples; NS, P > 0.05 by two-way ANOVA with Holm–Sidak correction, ~2,000 worms per replicate. In c and d, n = 3 biologically independent samples. In g, error bars represent mean ± s.e.m., n = 4 biologically independent samples; NS, P > 0.05 by two-way ANOVA with Holm–Sidak correction, ~2,000 worms per replicate. In h, n = 3 biologically independent samples; NS, P > 0.05 by log-rank test, 98–120 worms per replicate. See Supplementary Table 4 for full lifespan data. Source numerical data are available in source data. Source data

Fig. 8. Neuronal transduction of peripheral lipid signals to regulate longevity.

a, Schematic design using neuron-expressing GBP to detect the uptake of intestine-secreted LBP-3 into neurons. b, In the transgenic line expressing LBP-3–GFP in the intestine and GBP polycistronic mKate in neurons, GFP signals from secreted LBP-3 are detected in neurons (mKate positive). Scale bars, 30 µm and 10 µm in the inset. c, Constitutive expression of lbp-8 (lbp-8 Tg) does not affect the transcription of egl-3, while the induction of either egl-21 or nlp-11 is negligible (<35%). d, lbp-3 Tg and lbp-8 Tg have an additive effect on lifespan extension. e, The loss-of-function mutation of the nuclear receptor nhr-49(lf) fully suppresses the induction of the neuropeptide genes by lipl-4 Tg, while the loss-of-function of nhr-80(lf) only decreases the induction by less than 17%. f, nhr-49(lf) fully suppresses the induction of the neuropeptide genes by lbp-3 Tg, which is rescued by neuronal restoration of nhr-49. g, nhr-49(lf) fully suppresses the lifespan extension in the lbp-3 Tg worms, which is rescued by neuronal restoration of nhr-49. h, In the transgenic line expressing nlp-11 polycistronic GFP and mKate-fused NHR-49 driven by their endogenous promoters, GFP and mKate signals overlap in many neurons. Scale bar, 10 µm. i, Cartoon illustrating the overall model. In c, error bars represent mean ± s.e.m., n = 4 biologically independent samples, *P = 0.034, **P = 0.002 by two-way ANOVA with Holm–Sidak correction, ~2,000 worms per replicate. In d and g, n = 3 biologically independent samples, ***P < 0.001 by log-rank test, 80–101 worms per replicate. See Supplementary Table 4 for full lifespan data. In e, error bars represent mean ± s.e.m., n = 4 biologically independent samples, ***P = 0.0004 and ****P < 0.0001 by two-way ANOVA with Holm–Sidak correction, ~2,000 worms per replicate. In f, error bars represent mean ± s.e.m., n = 4 biologically independent samples, *P = 0.019, **P = 0.009 and ***P = 0.0006 for WT versus lbp-3 Tg, P = 0.0003 for lbp-3 Tg versus lbp-3Tg;nhr-49(lf) and P = 0.0003 for lbp-3Tg;nhr-49(lf) versus lbp-3Tg;nhr-49(lf);neu-nhr-49, ****P < 0.0001 by two-way ANOVA with Holm–Sidak correction, ~2,000 worms per replicate. Source numerical data are available in source data. Source data

Extended Data Fig. 1. Peripheral lysosomal lipolysis up-regulates neuropeptide signaling (related to Fig. 1).

a) Principal Component Analysis (PCA) of lipl-4 Tg and WT worms falling in two distinct clusters. b) The loss-of-function mutant of egl-21(lf) shows lifespan extension, and lipl-4 Tg cannot further enhance this lifespan extension as it does in the WT condition. c) In the background of daf-16 RNAi inactivation, the longevity effect caused by egl-21(lf) is suppressed. daf-16 RNAi is thus used to eliminate the contribution from ILP reduction in the egl-21(lf) mutant. d) Inactivation of daf-16 does not affect the longevity effect of lipl-4 Tg. (a) PCA analysis using two-sided statistical test, (b-d) n=3 biologically independent samples, n.s. p>0.05 and *** p<0.001 by long-rank test, 60-120 worms per replicate. See Supplementary Tables 4-5 for full lifespan data.

Extended Data Fig. 2. NLP-11 neuropeptide acts in neurons to promote longevity (related to Fig. 2).

a) A schematic representation of the nlp-11(rax51) loss-of-function mutation. White boxes represent the nlp-11(rax51) deletion, black lines represent introns. The mutant lacks the three exons and the transcriptional start site. b) nlp-11 is transcriptionally up-regulated by lipl-4 Tg. c) Constitutive expression of nlp-11 driven by its endogenous promoter extends lifespan. d) The transgenic strains expressing nlp-11 under its endogenous promoter (nlp-11p::nlp-11::sl2-GFP) reveals the expression of nlp-11 in intestinal and neuronal cells. Scale bar 100µm. e, f) Neuron-specific overexpression of nlp-11 prolongs lifespan (e), but intestine-specific overexpression has no such effect (f). (b) Error bars represent mean ± s.e.m., n=3 biologically independent samples, **** p<0.0001 by two-tailed Student’s t-test, ~2000 worms per replicate. (c, e, f) n=3 biologically independent samples, n.s. p>0.05 and *** p<0.001 by long-rank test, 60-120 worms per replicate. See Supplementary Table 4 for full lifespan data. Source numerical data are available in source data. Source data

Extended Data Fig. 3. Lysosome-derived PUFAs in the periphery regulate neuropeptide and longevity (related to Fig. 3).

a) The level of TAG is decreased in the purified lysosomes from lipl-4 Tg compared to those from WT worms. b) The loss-of-function mutant fat-1(lf) suppresses lipl-4 Tg longevity. c) In the loss-of-function mutant of fat-3(lf), the lifespan-extending effect of lipl-4 Tg is shortened. (a) Error bars represent mean ± s.e.m., n=3 (WT) and n=4 (lipl-4 Tg) biologically independent samples, **p=0.0025 by two-tailed Student’s t-test, ~ 200,000 worms per replicate. (b, c) n=3 biologically independent samples, n.s. p>0.05 and *** p<0.001 by long-rank test, 60-120 worms per replicate. See Supplementary Table 4 for full lifespan data. Source numerical data are available in source data. Source data

Extended Data Fig. 4. Peripheral lipid chaperone LBP-3 regulates neuropeptide and longevity (related to Fig. 4).

a, b) The induction of egl-3 (a) and egl-21 (b) in lipl-4 Tg is suppressed by RNAi inactivation of either lbp-2 or lbp-3, but not lbp-1. c) Schematic representation of lbp-2 and lbp-3 loss-of-function mutants. White boxes represent the lbp-2(rax63) and lbp-3(rax60) deletions, black boxes represent insertions, while black lines represent introns. The mutants lack the entire first exon and transcriptional start site. d) The transgenic strain that carries constitutive expression of lbp-3 (lbp-3 Tg) driven by its own endogenous promoter prolongs lifespan. e) PCA analysis of lbp-3 Tg and WT worms falling in two distinct clusters. f) Inactivation of daf-16 does not affect the longevity effect of lbp-3 Tg. daf-16 RNAi knockdown used to eliminate the influence from ILP reduction in egl-21(lf). g) Overexpression of lbp-3 selectively in the intestine extends lifespan. (a, b) Error bars represent mean ± s.e.m., n=4 biologically independent samples, n.s. p>0.05, *p=0.0104, **p=0.0018, ***p=0.0001 and ****p<0.0001 by two-way ANOVA with Holm-Sidak correction, ~2000 worms per replicate. (e) PCA analysis using two-sided statistical test. (d, f, g) n=3 biologically independent samples, n.s. p>0.05 and *** p<0.001 by long -rank test, 60-120 worms per replicate. See Supplementary Tables 4-5 for full lifespan data. Source numerical data are available in source data. Source data

Extended Data Fig. 5. LBP-3 secreted from the periphery regulates neuropeptide and longevity (related to Fig. 5).

a) LBP-3 and the lysosomal membrane protein LMP-1 are visualized by their GFP and RFP fusions, respectively. LBP-3::GFP colocalizes with LMP-1::RFP and Lysotracker Red staining at lysosomes, and is also detected in the cytosol. Scale bar 10µm. b) Intestine-specific overexpression of lbp-3 lacking its secretory signal (lbp-3ns) fails to extend lifespan. n=3 biologically independent samples, n.s. p>0.05 by long -rank test, 72-100 worms per replicate. See Supplementary Table 4 for full lifespan data.

Extended Data Fig. 6. DGLA regulates LBP-3 secretion, neuropeptides and longevity (related to Fig. 6).

a) Intestine was dissected from lipl-4 Tg with fat-3 RNAi knockdown and used for qPCR analysis. The expression of egl-21 is not detectable in the dissected intestine, while the expression of egl-3 is weakly detected. The transcriptional level of either egl-3 or egl-21 is not affected by DGLA supplementation. DMSO serves as the vehicle control. b) egl-21 transcripts are measured in lipl-4 Tg with fat-3 RNAi using smFISH (red Quasar 670). Upon DGLA supplementation, the egl-21 transcript level is increased compared to the vehicle control. Scale bar 10µm. Representative images from three biological repeats. (a) Error bars represent mean ± s.e.m., n=3 biologically independent samples, n.s. p>0.05 by two-way ANOVA with Holm-Sidak correction, ~20 dissected intestines per replicate. (b) n=3 biologically independent samples, ****p<0.0001 by two-tailed Student’s t-test, 12 worms per replicate. Source numerical data are available in source data. Source data

Extended Data Fig. 7. DGLA binding specificity of LBP-3 mediates its effects (related to Fig. 7).

a) After incubation with the C. elegans liposome, fatty acids bound to LBP-2 or LBP-3 proteins were analyzed with mass spectrometry. In addition to PUFAs shown in Fig. 5, SFAs and MUFAs also bind to LBP-2 and LBP-3 with different preferences. b) The LBP-3 chimeric protein structure predicted using AlphaFold2. The two helix regions from LBP-2 are highlighted in purple. c) Western-blot of WT and transgenic worms overexpressing LBP-3 chimeric proteins fused with 3XHA-tag. LBP-3::3xHA fusion proteins are detected in both chimeric lines. ß-actin is used as a control. d, e) Intestine-specific overexpression of the chimeric lbp-3(chim) does not affect the transcription of egl-3 or egl-21 (d) and fails to extend lifespan (e). (d) Error bars represent mean ± s.e.m., n=3 biologically independent samples, n.s. p>0.05 by two-way ANOVA with Holm-Sidak correction, ~300 worms per replicate. (e) n=3 biologically independent samples, n.s. p>0.05 by long-rank test, 72-100 worms per replicate. See Supplementary Table 4 for full lifespan data. Source numerical data and unprocessed blots are available in source data. Source data

Extended Data Fig. 8. Neuronal transduction of peripheral lipid signals to regulate longevity (related to Fig. 8).

a) In the transgenic line expressing GBP polycistronic mKate in neurons alone, no GFP signals are detected in neurons. Scale bar 30µm and 10µm in the inset. b) In the transgenic line expressing secretable LBP-3::GFP selectively in the intestine alone, GFP signals are only strongly detected in the intestine and pharynx. Scale bar 30µm and 10µm in the inset. c) In the transgenic line expressing GBP tagged with the extracellular domain of SAX-7 (GBP::SAX-7) and polycistronic mKate in neurons and secretable LBP-3::GFP in the intestine, strong GFP signals are detected in neurons. Scale bar 30µm and 10µm in the inset. d) Heatmap showing averaged nlp-11 and nhr-49 gene expression per neuronal cell type (threshold equal to two). The color of circles represents relative gene expression, while the size of circles indicates the percentage of cells expressing the gene within each neuron cluster. The heatmap is generated using the online webtool CenGeneApp. Neuronal cell types shared between nhr-49 and nlp-11 are marked by black arrows.