The C. elegans LON-1 protein requires its CAP domain for function in regulating body size and BMP signaling

Abstract

The CAP (cysteine-rich secretory proteins, antigen-5, and pathogenesis-related) proteins are widely expressed and have been implicated to play diverse roles ranging from mammalian reproduction to plant immune response. Increasing evidence supports a role of CAP proteins in lipid binding. The Caenorhabditis elegans CAP protein LON-1 is known to regulate body size and bone morphogenetic protein (BMP) signaling. LON-1 is a secreted protein with a conserved CAP domain and a C-terminal unstructured domain with no homology to other proteins. In this study, we report that the C-terminal domain of LON-1 is dispensable for its function. Instead, key conserved residues located in the CAP domain are critical for LON-1 function in vivo. We further showed that LON-1 is capable of binding sterol, but not fatty acid, in vitro, and that certain key residues implicated in LON-1 function in vivo are also important for LON-1 sterol binding in vitro. These findings suggest a role of LON-1 in regulating body size and BMP signaling via sterol binding.

Keywords: C. elegans; BMP signaling; CAP; LON-1; body size; fatty acid binding; sterol binding.

Fig. 1.

Summary of null and endogenously tagged lon-1 alleles. a) Schematic of the lon-1 gene, including the canonical e185 allele, and the null allele generated in this study, jj284. b) Schematics of WT LON-1 protein and tagged SP::V5::LON-1 protein present in lon-1(jj373) animals. LON-1 contains a predicted SP (signal peptide), a CAP (cysteine-rich secretory protein, antigen-5, and pathogenesis-related 1) domain, a hinge region, and an unstructured CTD with no known homology. c) Relative body lengths of lon-1 alleles at larval L4.3 stage, normalized to WT. Each dot represents one worm. ****P < 0.0001; n.s., not significant. Tested using an ANOVA with a Tukey HSD. d) DIC images of stage-matched WT and lon-1(jj284) null worms. Scale bar represents 40 μm. e) Western blot showing that SP::V5::LON-1 protein is detectable using anti-V5 (top). Actin was used as a loading control (bottom). *Marks a nonspecific protein band present in all samples. Each lane contains lysates from 100 young adult worms.

Fig. 2.

LON-1 is a secreted protein in the CAP superfamily. a) Predicted structure of LON-1 protein by Alphafold (Model AF-Q09566-F1), visualized using UCSF ChimeraX (Meng et al. 2023). The signal peptide is colored in green, the CAP domain in light blue, the hinge region in gray, and the CTD in yellow. b) Schematic of the LON-1::V5 protein expressed in Drosophila S2 cells. c) Western blot showing that LON-1::V5 protein is secreted into the media when expressed in Drosophila S2 cells. Cells were separated from their corresponding media 7 days after transfection. The “+” samples from cells transfected with pAc5::LON-1::V5. “−”: samples from un-transfected cells. d) Alignment of the CAP domains of LON-1 and two other CAP family members. From top to bottom: C. elegansLON-1 (Gene ID: 175753), Saccharomyces cerevisiae Pry-1 (Gene ID: 853366), and Homo sapiens CRISP2 (Gene ID: 7180). Protein sequence alignment was generated using Clustal Omega, accessed online (https://www.ebi.ac.UK/jdispatcher/msa/clustalo). This figure was generated with ENDscript 2 (Robert and Gouet 2014), using the predicted structure of LON-1 protein by Alphafold (Model AF-Q09566-F1). Color shaded boxes mark conserved amino acids that are mutated in various C. elegans point mutants. Mutations in purple (E153A and H171A) target a putative catalytic site, those in green (C185Y and C207S) target a cysteine bridge, and a mutation in blue (V175A) targets a putative fatty-acid binding region. e–g) Zoomed regions of predicted LON-1 structure showing the locations of amino acids of interest.

Fig. 3.

LON-1 CTD is not crucial for LON-1 function in vivo. a) Schematics of SP::V5::LON-1 present in lon-1(jj373) and SP::V5::LON-1ΔCTD present in lon-1(jj373 jj536) animals. b) Western blot showing that SP::V5::LON-1(ΔCTD) protein is detectable using anti-V5 (top). Actin was used as a loading control (bottom). Each lane contains lysates from 150 GA worms. The SP::V5::LON-1(ΔCTD) protein is of similar molecular weight to a nonspecific protein band (marked with an asterisk in Fig. 1d) found in lon-1(jj373[SP::V5::LON-1]) worms. c) Relative body lengths of lon-1 alleles at larval L4.1 stage, normalized to WT. Each dot represents one worm. ****P < 0.0001; n.s., not significant. Tested using an ANOVA with a Tukey HSD.

Fig. 4.

LON-1 binds sterols in vitro and mutations of conserved cysteine residues decrease the binding affinity. a) The binding of cholesterol sulfate to wild-type and mutant LON-1 proteins measured using MST (see materials and methods). Cholesterol sulfate was titrated into 50 ng of fluorescently labeled proteins. K_D values are indicated and the values represent mean ± SD of three independent measurements. b) Binding affinity (K_D) as calculated in a) were plotted for comparison. Asterisks indicate statistical significance (one-way ANOVA with Tukey's post hoc test; **P < 0.01; ***P < 0.001; n.s., not significant).

Fig. 5.

LON-1 binds palmitic acid in vitro at high micromolar concentrations. a) MST dose response curves of fluorescently labeled proteins and fatty acids. Wild-type and mutant LON-1 proteins were tested for palmitic acid binding. The values represent mean ± SD of three independent measurements, and the dissociation constant (K_D) are indicated. b) K_D values obtained in a) were plotted for comparison.

Fig. 6.

Point mutations in conserved residues in the CAP domain affect LON-1 function in vivo. a) Relative body lengths of various strains at larval L4.1 stage, normalized to WT. Each dot represents one worm. ****P < 0.0001; n.s., not significant. Tested using an ANOVA with a Tukey HSD. b) Table showing the Susm penetrance of various strains. Susm penetrance is measured as the percentage of 6 CC animals as scored by the arIs37(secreted CC::GFP) reporter. Statistical analysis was conducted by comparing double mutant lines with the sma-9(cc604) single mutants. ***P < 0.001; *P < 0.05 (unpaired two-tailed Student's t-test). c) Western blot showing that mutant LON-1 proteins are detectable using anti-V5 antibodies (top). Each lane contains lysate from 100 GA worms. Actin (bottom) was used as a loading control.