The type II integral ER membrane protein VAP-B homolog in C. elegans is cleaved to release the N-terminal MSP domain to signal non-cell-autonomously

Abstract

VAMP/synaptobrevin-associated protein B (VAP-B) is a type II ER membrane protein, but its N-terminal MSP domain (MSPd) can be cleaved and secreted. Mutations preventing the cleavage and secretion of MSPd have been implicated in cases of human neurodegenerative diseases. The site of VAP cleavage and the tissues capable in releasing the processed MSPd are not understood. In this study, we analyze the C. elegans VAP-B homolog, VPR-1, for its processing and secretion from the intestine. We show that intestine-specific expression of an N-terminally FLAG-tagged VPR-1 rescues underdeveloped gonad and sterility defects in vpr-1 null hermaphrodites. Immunofluorescence studies reveal that the tagged intestinal expressed VPR-1 is present at the distal gonad. Mass spectrometry analysis of a smaller product of the N-terminally tagged VPR-1 identifies a specific cleavage site at Leu156. Mutation of the leucine results in loss of gonadal MSPd signal and reduced activity of the mutant VPR-1. Thus, we report for the first time the cleavage site of VPR-1 and provide direct evidence that intestinally expressed VPR-1 can be released and signal in the distal gonad. These results establish the foundation for further exploration of VAP cleavage, MSPd secretion, and non-cell-autonomous signaling in development and diseases.

Introduction

VAMP/synaptobrevin-associated proteins (VAPs) are endoplasmic reticulum (ER) integral proteins that are evolutionarily conserved from yeast to mammals (1). Two homologous genes, VAP-A and VAP-B, exist in mammalian genomes (2). VAP proteins contain three conserved domains: an N-terminal major sperm protein domain (MSPd), a central coiled-coil domain (CCD), and a C-terminal transmembrane domain (TMD) (1–3). The MSP domain is named after the nematode major sperm protein (MSP), a cytosolic protein that can polymerize to form the filamentous cytoskeleton required for sperm motility in C. elegans (1, 4). MSP also has a hormone-like activity to stimulate oocyte maturation and sheath cell contraction, but it is not understood how MSP is released from sperm to act non-cell-autonomously (5–7). In VAP, the amino acid residues responsible for MSP polymerization are not conserved in MSPd, though the domain might contribute to dimerization of VAP (1, 4). The MSPd folds into an immunoglobin-like, seven-stranded β sandwich and is involved in interaction with its protein partners (4, 7, 8). The CCD also participates in binding to other proteins, whereas the TMD anchors the VAP protein into the ER membrane (1). VAP proteins assume a type II membrane protein topology with the N-terminal MSPd extending into the cytosol rather than the ER lumen (9). This allows VAP proteins to function in various intracellular processes such as lipid transport and metabolism, membrane trafficking, the ER unfolded protein response, and microtubule organization (8, 10).

Although most studies focus on intracellular activities of VAP, recent work has suggested that VAP MSPd can function as a secreted non-cell-autonomous signal in C. elegans, Drosophila, and mammals (11–14). In Drosophila, the VAP homolog has been shown to regulate boutons at neuromuscular junctions (15). In human, a short, processed form of VAP, but not full length VAP, is detected in serum samples of healthy individuals (11). In C. elegans, the MSPd of the VAPB homolog, VPR-1, functions as a permissive signal for gonadogenesis (14). C. elegans null for vpr-1 are maternal effect sterile due to the underdevelopment of their gonad. The defects in gonadogenesis can be rescued by transgenically expressing vpr-1 in the germ line or the nervous system, and to a lesser extent other tissues, indicating a non-cell-autonomous role for VPR-1 in gonad development (14). VPR-1 is also shown to modulate mitochondrial localization and fission/fusion in striated muscles by binding to SAX-3 and Lar-like surface receptors (12, 16). Genetic mosaic analyses and tissue-specific transgenic expression of vpr-1 demonstrate that VPR-1 from the nervous system and the germ line, but not the muscles, maintains muscle mitochondrial localization and metabolism (12, 16). The results again support the notion that VPR-1 acts in a non-cell-autonomous fashion to regulate development of other tissues.

The relevance of VAP processing and signaling in normal physiology is inferred from studies of VAP-B mutants identified in human patients with amyotrophic lateral sclerosis (ALS) or Lou Geghrig’s disease (17–20). While most ALS cases are sporadic, a small percentage of cases have been traced genetically through large extended families (21). Studies of these extensive pedigrees identified several genes, including VAP-B, that segregate with cases of ALS (17–19, 22–26). Among the point mutations in MSPd of VAP-B that are associated with ALS, a proline to serine change at the amino acid 56, known as P56S, is best studied. P56S is a dominant negative mutation that disrupts hydrogen bonding between two β-sheets of MSPd (17). Unlike the wild type VAP, a mutant that mimics P56S substitution in Drosophila prevents secretion of MSPd and leads to protein aggregation (11). Moreover, though VAP MSPd is detected in cerebral spinal fluid of healthy individuals, it is largely absent in sporadic ALS patients with bulbar onset (20). These data imply that MSPd of VAP-B might function as a circulating hormone to signal to and regulate other cells. However, despite this evidence, a direct demonstration of the role of VAP cleavage has not been provided by any of the studies.

In this study, we present in vivo evidence for cleavage and secretion of the N-terminal MSPd of VPR-1 from intestinal cells in C. elegans. We identify the cleavage site of VPR-1 and show that mutation of this site reduces the activity of VPR-1 to rescue the sterility defects of vpr-1 mutant. We demonstrate that MSPd released from intestinal cells can reach the distal gonad in C. elegans and might regulate gonadal cell development. Our results show for the first time the biochemical evidence of VPR-1 processing site and the importance of VPR-1 cleavage for its long distance signaling and function.

Materials and Methods

C. elegans genetics and strains

C. elegans strains were maintained at 20°C and fed NA22 E. coli bacteria (27, 28). N2 Bristol (wild type) and VC1478 vpr-1 (tm1411)/hT2 [bli-4 (e937) let-? (q782) qIs48] (I; III) were the two strains used in this study. A list of the strains made for this study can be found in Supplemental Table 1. Phenotypes of vpr-1 (tm1411) mutants and transgenic lines were evaluated in vpr-1 (tm1411) homozygous F2 progeny from vpr-1 (tm1411)/hT2 heterozygotes (F0) due to the maternal effect of vpr-1 (14).

Molecular cloning

All transgenic expression plasmids were made in a TOPO vector backbone with an insert that included ges-1p intestinal specific promoter (29, 30), genomic vpr-1, and vpr-1 UTR (sequences listed in Supplemental Table 2). PCR was used to amplify insert sequences from genomic DNA with primers designed for Gibson Assembly. The sequences of these primers are included in Supplemental Table 2. All constructs were ligated by Gibson Assembly (New England Biolabs). FLAG and HA tags were inserted by designing overlapping primers that included the desired tag sequence and by adding corresponding forward and reverse oligos in the Gibson Assembly mix. Constructs were transformed into TOP10 cells for amplification. All constructs were confirmed by sequencing.

The Cas9 DNA repair templates were designed as previous described (14). The repair template to insert a FLAG tag at the N-terminus of the endogenous vpr-1 loci was a construct that contained a 2 kb left homology arm of vpr-1::double-FLAG tag sequence:: 2 kb right homology arm of vpr-1. The Cas9 targeting guide sequence was 5’- CTACCCACTAAGCACTGGCC −3’. The Cas9 DNA repair template to insert an HA tag at the C- terminus of the endogenous vpr-1 loci contained an insert of 2 kb left homology arm of vpr-1:: double-HA tag sequence:: 2 kb right homology arm of vpr-1. The Cas9 targeting guide sequence was 5’- CTCTCCTCATCGGGCTTATT −3’. The single guide RNA (sgRNA) plasmid was derived from Addgene plasmid 46169. PCR was used to amplify the entire sgRNA backbone except for the 20 base pairs guide sequence. The 20 base pairs guide sequence was incorporated into the sgRNA backbone by overlapping primers designed for Gibson Assembly. Gibson Assembly was used to ligate all constructs. Constructs were transformed into TOP10 cells for amplification. All constructs were confirmed by sequencing.

CRISPR/Cas9

CRISPR/Cas 9 methods were performed as previously described (31). Cas9 plasmid (75 ng/μl), vpr-1 sgRNA plasmid (50 ng/μl), vpr-1 repair template plasmid (100 ng/μl), co-CRISPR unc-119 sgRNA plasmid (50 ng/μl), co-CRISPR unc119 repair template (50 ng/μl), and myo-3p::mitoGFP co-injection marker (30 ng/μl) were injected into young adult unc119(ed3) hermaphrodite gonads. Progeny was screened for rescue of the unc-119 movement defect and loss of myo-3p::mitoGFP. Individual worms were isolated repeatedly to ensure 100% segregation. PCR and sequencing were used to confirm the insertion of FLAG and HA tags.

Transgenic Lines

Desired plasmids (50 μl) were injected into young adult vpr-1 (tm1411)/hT2 hermaphrodite gonads to generate C. elegans extrachromosomal lines. The myo-3p::mitoGFP transgene was co-injected (30 μl) as a marker for successful injections. C. elegans extrachromosomal strains that transmitted at ≥ 60% were used for assays. Multiple independent lines for each plasmid combination were analyzed.

Fertility and Brood size Assays

One L3-L4 staged C. elegans hermaphrodite was placed to an agar plate seeded with NA22 E. coli and transferred to a fresh plate with bacteria every 24 hours for 6 days. Live progenies were counted on preceding plate after 48 hours of incubation at 20°C. Brood counts of C. elegans hermaphrodites that died before day 5 were eliminated. C. elegans hermaphrodites that produced no progeny were considered sterile.

Statistical Tests

Two-tailed Student’s t-tests without the assumption of equal variance were conducted using Prism. Chi-Square tests were conducted using Microsoft Excel.

Immunofluorescences and Imaging

Adult C. elegans were washed in M9 buffer and decapitated with a syringe needle to expel the intestine. Tissues were fixed overnight at 4°C in 4% formalin (Sigma-Aldrich HT5011) then blocked for 1 hour at room temperature in PBS+ (1x PBS with 0.1% Triton X-100, 1% bovine serum albumin, 1% donkey serum, and 0.02% sodium azide). Rabbit anti-FLAG primary antibody (Thermo Fisher Scientific, Cat. No. 740001; 1:1,000) and mouse anti-HDEL primary antibody (previously used by (32) to mark the ER in C. elegans; Santa Cruz Biotechnology, Santa Cruz, CA; 1:250) were diluted in PBS+ and tissues were incubated occurred overnight at 4°C. Secondary antibody incubation with Alexa Fluor 555 goat anti-rabbit IgG (Thermo Fisher Cat no. A21428; 1:8,000 dilution) was for 1 hour at room temperature. Tissues were washed three times for 5 minutes each with PBS with 0.1% Triton X-100 in between antibody incubation periods. The final wash contained Hoechst nuclear stain (Thermo Fisher 33258) at 1:1,000 dilution to visualize nuclei and phalloidin stain (Invitrogen Alexa Fluor 488 Phalloidin) at 1:300. Confocal images were taken with a Nikon Ti2 spinning disk confocal with a Yokohama X1 disk and an Orca Flash4.0 sCMOS (Hamamatsu). Images were acquired in Nikon Elements AR 5.0.

Immunoprecipitation

Unsynchronized C. elegans were collected to prepare the worm lysate for immunoprecipitation. C. elegans were washed with M9 buffer to remove external traces of E. coli. A 5 mL pellet of C. elegans carcasses was frozen down at −80°C for up to one week. An equal amount of worm lysis buffer containing 25 mM Hepes-NaOH, 150 mM NaCl, 0.2 mM DTT, 10% glycerol, and 1% Triton X-100 supplemented with protease inhibitor cocktail (Roche Applied Sciences, Germany) was added to the frozen carcass pellet. To homogenize the worm carcasses into a lysate, the worm carcasses and worm lysis buffer mixture was blended using 0.5 mm zirconium oxide beads and a Bullet Blender 5 homogenizer (Next Advance Inc., NY, USA) at speed 10 for 5 minutes. The homogenate was then transferred to a 15 mL Falcon tube and centrifuged at 4,500 × g for 10 minutes at 4 °C. The clarified supernatant (10 mL) was incubated with either EZview Red Anti-FLAG M2 Affinity Gel (Sigma-Aldrich F2425) or EZview Red Anti-HA Affinity Gel (Sigma-Aldrich E6779) beads (500 μl) for 2 hours at 4°C. Beads with bound proteins were washed three times with 1x TBST buffer. Proteins bound to the anti-FLAG gel beads were eluted by 4 mg/mL 3x FLAG peptide (Sigma Aldrich F4799). Proteins bound to the anti-HA gel beads were eluted by 2x SDS sample buffer.

SDS-PAGE and western blotting

Protein samples were boiled in 2x SDS sample buffer for 5 minutes and loaded into 4–20% Mini-PROTEAN TGX Precast Gels (Bio-Rad 456). If used for Western Blots, SDS gels where transferred to immobilon-P PVDF membranes (Milipore). Immunoblots were blocked for 1 hour with 1% non-fat dry milk in 1x TBS buffer containing 0.1% Tween-20 (TBST). The membranes were then probed overnight at 4°C with rabbit anti-FLAG antibody (Thermo Fisher Scientific 740001; 1:8000) diluted in 1% non-fat dry milk. Membranes were then washed three times with TBST and incubated with IRDye 800CW Donkey anti-Rabbit secondary antibodies (Li-cor 925–32213) for 1 hour at room temperature. The signal was visualized by Licor Odyssey CLx. If SDS gel was submitted for in-gel digestion and mass spectrometry (below), then gel was incubated in Coomassie blue stain overnight at room temperature and destained with DI water.

In-gel Digestion

SDS gel bands were excised and excess stain was removed by an overnight wash of 50% 100 mM ammonium bicarbonate/50% acetonitrile. After destaining, disulfide bonds were reduced by 25 mM dithiothreitol at 50°C for 30 minutes. Alkylation of the free thiol groups was carried out with 55 mM iodoacetamide for 30 minutes in the dark. The excess alkylating agent was removed, and the gel pieces were washed twice with a 100 mM ammonium bicarbonate for 30 minutes. The gel pieces were evaporated to dryness in a SpeedVac (Savant) before enzymatic digestion. A 12.5 ng/ml concentration of trypsin (Promega Gold Mass Spectrometry Grade, which is modified to be autolytic resistant and includes TPCK to inactivate any chymotrypsin activity) was added to each gel sample and incubated overnight at 37°C. Peptides were extracted from the gel pieces using a 1:1 mixture of 1% formic acid and acetonitrile twice for 15 minutes. Extracts were pooled and evaporated to dryness. The samples were then resuspended in 30 μL of a 0.1% formic acid prior to mass spectrometry analysis.

NanocHiPLC-tandem mass spectrometry

An aliquot (5 μL) of each digest was loaded onto a Nano cHiPLC 200 μm x 0.5 mm ChromXP C18-CL 3 μm 120 Å reverse-phase trap cartridge (Eksigent, Dublin,CA) at 2 μL/min using an Eksigent autosampler. After washing the cartridge for 4 minutes with 0.1% formic acid in ddH₂O, the bound peptides were flushed onto a Nano cHiPLC column 200 μm x 15 cm ChromXP C18-CL 3 μm 120 Å (Eksigent, Dublin,CA) with a 45 minutes linear (5–50%) acetonitrile gradient in 0.1% formic acid at 1000 nL/min using an Eksigent Nano1D+ LC. (Dublin, CA). The column was washed with 90% acetonitrile-0.1% formic acid for 10 minutes and then re-equilibrated with 5% acetonitrile-0.1% formic acid for 10 minutes. The SCIEX 5600 Triple-Tof mass spectrometer (AB-Sciex, Toronto, Canada) was used to analyze the protein digest. The IonSpray voltage was 2300 V and the declustering potential was 80 V. Ionspray and curtain gases were set at 10 psi and 25 psi, respectively. The interface heater temperature was 120°C.

Eluted peptides were subjected to a time-of-flight survey scan from 400–1250 m/z to determine the top twenty most intense ions for MSMS analysis. Product ion time-of-flight scans at 50 milliseconds were carried out to obtain the tandem mass spectra of the selected parent ions over the range from m/z 100–1500. Spectra are centroided and de-isotoped by Analyst software, version TF (Applied Biosystems). A beta-galactosidase trypsin digest was used to establish and confirm the mass accuracy of the mass spectrometer.

Protein Pilot 4.5 Search Queries

The tandem mass spectrometry data were processed to provide protein identifications using an in-house Protein Pilot 4.5 search engine (SCIEX) using the Homo sapiens UniProt protein database and using a trypsin plus missed cleavage digestion parameter. Potential novel cleavage site peptide spectra were evaluated via de novo sequencing to verify the validity of the search results.

Results

Expression of vpr-1 in intestinal cells rescues vpr-1 null sterile phenotype

C. elegans null for vpr-1, or vpr-1(tm1411), are maternal effect sterile (Figure 1A–C). The vpr-1 (tm1411) line is maintained as heterozygous hermaphrodites containing a hT2 balancer chromosome, a translocation between chromosomes I and III that includes a GFP pharyngeal marker (Figure 1A) (33, 34). Homozygous vpr-1 null progenies are identified by the loss of the GFP pharyngeal marker during self-fertilization. These F1 vpr-1 null worms are viable and fertile due to material deposit of VPR-1. However, the second generation of C. elegans from these homozygous vpr-1 null hermaphrodites produce no progeny due to maternal effect sterility (Figure 1A–C), and the gonad of these F2 worms is underdeveloped (Figure 1E) when compared with age-matched wild type animals (N2, Figure 1D). We have previously shown that expression of vpr-1 in the intestinal cells is sufficient to rescue the vpr-1 defects in gonadogenesis (14). Here we further characterized the level of rescue with intestine-specific transgenic expression of vpr-1. We showed that expression of vpr-1 in intestinal cells by the ges-1 promoter results in a partial rescue in the F2 population with fertility detected in 60% of the worms (Figure 1B), and a significantly larger brood size is also increased significantly (Figure 1C). The gonad morphology of these adult fertile hermaphrodites is comparable to that of the wild type (Figure 1F).

Figure 1: vpr-1(tm1411) maternal effect sterility in C. elegans can be rescued by intestinal expression of vpr-1.

A) The vpr-1(tm1411) deletion allele is maintained by the hT2 balancer (F0). Homozygous null hermaphrodites, vpr-1(tm1411)/vpr-1(tm1411), are identified by selecting against the green pharyngeal marker (F1 and F2). Homozygous null lines that express an injected extrachromosomal transgene express the co-injection marker myo-3p::mitoGFP (green body wall muscles) and lack the pharyngeal GFP marker associated with the hT2 balancer (F1 and F2). Fertility rates, brood size assays, and gonad morphology of these C. elegans were assessed in F2 vpr-1(tm1411) hermaphrodites due to the maternal effect of vpr-1. B) Graph represents fertility assay. N2 wild type C. elegans are 100% fertile (25/25) while 0% of vpr-1(tm1411) hermaphrodites are fertile (0/30). Majority of vpr-1(tm1411); Ex (ges-1p::vpr-1) (25/42) and vpr-1(tm1411); Ex (ges-1p:: FLAG:vpr-1) (28/39) hermaphrodites are fertile and brood sizes are not significantly different. C. elegans expressing Ex (ges-1p::FLAG:vpr-1:HA) in the vpr-1(tm1411) null background are completely sterile (0/37) (all bars indicate a p <0.01, Chi-Square Test between the connected pair; no bar indicates no significant difference between the pair) C) Graph represents brood size count. Brood size of transgenic worms was compared to wild type. vpr-1(tm1411) hermaphrodites produce no progeny while intestinal expression of untagged or FLAG-tagged vpr-1 rescues this sterility phenotype (all bars indicate a p <0.0001, Student’s t-test between the connected pairs; n.s stands for no significant difference) D-G) C. elegans adult gonad images of the various genotypes assayed. The intestine is outline and pseudo-colored blue. The gonads are outlined and pseudo-colored yellow. Images captured with a 40x Plan Fluor 1.3NA objective. Scale bar: 50 μm.

VPR-1 from intestinal cells travels to the surface of the distal gonad

Different mechanisms might account for the non-cell-autonomous function of VPR-1. The protein might be release from intestine to signal in the gonad directly, or it might regulate a secondary extracellular signal to modulate worm sterility. To further explore the mechanism of VPR-1 function, we proceeded to examine the localization of VPR-1. For this purpose, we tagged vpr-1 with a double-FLAG epitope at the N-terminus and expressed it as a transgene using the intestinal ges-1 promoter (referred to as ges-1p::FLAG vpr-1) (30). This N-terminal tag sits adjacent to the MSPd-coding region of vpr-1. A fertility assay revealed that this construct does not show a significant difference in rescuing C. elegans fertility or brood size when compared with the untagged ges-1p::vpr-1 transgenic worm lines (Figure 1B–C). Moreover, no noticeable defects were detected in the gonad morphology of vpr-1 null adults rescued by the tagged versus untagged VPR-1 (Figure 1F–G).

Immunofluorescent staining of C. elegans intestinal cells with an anti-FLAG antibody revealed that the N-terminal FLAG-tagged VPR-1 localizes to the basolateral membrane of the intestinal cells (Figure 2; Supplemental Video 1). No localization was found around the apical membrane that lines the intestinal lumen. This basolateral localization implies that FLAG-VPR-1 is concentrated at subcellular positions that are close to the pseudocoelom or the circulatory system of the worm. This localization reveals that if VPR-1 is secreted from intestinal cells, it will likely enter the pseudocoelom, which is adjacent to the gonad, rather than the intestinal lumen.

Figure 2: FLAG VPR-1 localizes away from the intestinal lumen in intestinal cells.

VPR-1 localization is visualized by its N-terminal FLAG tag (red). In the intestine, VPR-1 localizes along the basolateral membrane of intestinal cells. Intense phalloidin staining (green) marks the apical intestinal brush boarder that lines the intestinal lumen. Images captured with a 60x APO Tirf 1.49NA objective. Scale bar: 50 μm

To further investigate possible secretion of FLAG-VPR-1 from C. elegans intestinal cells, we examined its localization around other cell types in ges-1p::FLAG vpr-1 transgenic worms. Previous work showed that the VPR-1 N-terminal MSPd binds to an unknown receptor on the somatic gonad to signal for proper gonad development (14). This suggest that the FLAG-tagged N-terminus of VPR-1 might reach the gonad if it is indeed released from the intestinal cells.

Immunofluorescent imaging revealed FLAG signal on the distal gonad of dissected ges-1p::FLAG vpr-1 transgenic worms (Figure 3). Z-stack imaging showed that the N-terminal FLAG signal of VPR-1 is restricted to the surface of the somatic gonad (Supplemental Video 2). No FLAG staining was detected inside the syncytium, on oocytes, fertilized embryos, or in the proximal gonad. Taken together, the results provide strong evidence that N-terminal FLAG-tagged VPR-1 is secreted from intestinal cells and travels to the distal gonad to modulate gonadal development.

Figure 3: VPR-1 is secreted from the intestine and binds to the distal gonad.

VPR-1 localization is visualized by its N-terminal FLAG tag (red). ges-1p specifically drives FLAG-VPR-1 In the intestinal cells. VPR-1 was also detected along the surface of the distal gonad indicating protein secretion from the intestinal cells. Images captured with a 60x APO Tirf 1.49NA objective. Scale bar: 50 μm

The N-terminal VPR-1 is cleaved at residue leucine 156

Previous work from our lab and others suggest that the VAP N-terminal MSP domain is cleaved and secreted whereas the full length VAP remains as an intracellular protein (11, 12). Therefore, we expected that the FLAG staining that we detected on the distal gonad of transgenic ges-1p:: FLAG vpr-1 hermaphrodites might reflect a cleaved product of VPR-1 that was expressed in the intestine. However, as a type II ER membrane protein, it is not clear where the cleavage might occur. We therefore set out to investigate the issue.

To distinguish the full length and the cleaved protein products, we applied CRISPR/Cas9 technology to tag the endogenous VPR-1 locus with a double-FLAG tag at its N-terminus and a double-HA tag at its C-terminus (vpr-1(xm17); Figure 4A). Tagging of the endogenous vpr-1 allele did not result in obvious defects in gonad morphology (Figure 4B–C), though the resulting hermaphroditic C. elegans did show slightly decreased brood size (Figure 4D). Nevertheless, as the worms are fertile and appear healthy, we believe that these terminal tags did not drastically impair endogenous function of VPR-1.

Figure 4: VPR-1 is cleaved into a < 25kDa peptide.

A) Diagram of CRISRP/Cas9 schematic to make the vpr-1(xm17) line by endogenously tagging the N-terminus of vpr-1 with double-FLAG and the C-terminus with double-HA tag. B-C) Endogenously tagged line, vpr-1(xm17), develops a normal gonad like wild type (N2) C. elegans. The intestine is outline and pseudo-colored blue. The gonads are outlined and pseudo-colored yellow Images captured with a 40x Plan Fluor 1.3NA objective. Scale bar: 50 μm. D) vpr-1(xm17) C. elegans have a reduced brood size, but are still significantly more fertile than the null. (all bars indicate a p <0.0001, Student’s t-test between the connected pairs) E) Immunoprecipitation with anti-HA beads pulls down full size VPR-1 protein at 37kD. F) Immunoprecipitation with anti-FLAG beads pulls down the full size VPR-1 protein at 37kD and the cleaved N-terminal peptide at <25kD.

Using the endogenously tagged vpr-1 line, we cultured the worms and made protein extract from unsynchronized adults. We then performed immunoprecipitation (IP) assay using either α-HA or α-FLAG antibody-conjugated agarose beads and analyzed the protein containing the N-terminal FLAG tag from the pulldown by Western Blot analysis. A protein band corresponding to the full sized VPR-1 protein at 37kDa was detected from the elutes using either antibody (Figure 4E–F). With α-FLAG beads, we also observed a smaller protein band at less than 25kDa that contained the N-terminal FLAG tag (Figure 4F). This band was not detected in the precipitate from α-HA IP (Figure 4E). These result indicate that the smaller band is the cleaved N-terminal product. To biochemically characterize this product, we excised this smaller band from the protein gel and subjected it to mass spectrometry analysis using conventional trypsin digestion (Figure 5). We recovered several peptides that matched the sequences within the N-terminal MSPd, but none that matched to the CCD or the TMD (Supplemental Figure 1). These peptides contained the characteristic C-terminal arginine or lysine typical of trypsin digestion. However, the most C-terminal peptide that we recovered with a 99% confidence level using the Protein Pilot 4.5 search engine (35) has the amino acid sequence of NEDSFASSGQAQEL (Supplemental Figure 1). This sequence ends at leucine 156 (Leu156), which is an amino acid not recognized by trypsin for cleavage. This result suggests that VPR-1 is cleaved after Leu156 to produce the N-terminal peptide that contains the MSP domain.

Figure 5:

The amino acid sequence NEDSFASSGQAQEL was identified as the end the VPR-1 25kDa band of by mass spectrometry.

Cleavage at Leu156 is required for VPR-1 function

To test the functionality of VPR-1 cleavage at Leu156, we made a point mutation changing leucine 156 to an alanine (L156A) in the ges-1p::FLAG vpr-1 construct. Two independently derived transgenic C. elegans lines were obtained. The fertility of these worms, the gonad morphology, and the localization of the N-terminal FLAG tag of the VPR-1 protein were then analyzed. Unlike the wild type FLAG-VPR-1 transgenic hermaphrodites, all of the hermaphrodites from the first line and the majority of the hermaphrodites from the second line were sterile (Figure 6A–B). An underdeveloped gonad, similar to that seen in vpr-1 null C. elegans, was observed and likely underlay the sterility phenotype (Figure 6D). The small percentage of fertile hermaphrodites from the second line developed a U-shaped somatic gonad but presented slight defects in the germ line shown by the lack of a continuous stream of developing oocytes (Figure 6E). Protein expression of FLAG-VPR-1 with L156A substitution was confirmed by immunofluorescent staining with anti-FLAG antibodies. Similar to the wild type FLAG-VPR-1 protein, the FLAG-VPR-1(L156A) localized to the basolateral membrane of intestinal cells (Figure 6F). However, FLAG-VPR-1(L156) was also detected throughout the intestinal cells at the luminal plane (Figure 6F). The defective gonad from the first line prevented us from further examination of gonadal FLAG signal. However, inspection of the distal gonad of the fertile hermaphrodites from the second line presented no FLAG signal (Figure 6G). These results suggest that cleavage at leucine 156 is required for the release of VPR-1 from intestinal cells to travel to and function in the gonad in order to regulate gonadal development and worm fertility.

Figure 6: L156A point mutation in vpr-1 causes sterility in C. elegans.

A) Graph represents fertility assay. N2 wild type C. elegans are 100% fertile (25/25) while 0% of vpr-1(tm1411) hermaphrodites are fertile (0/30). Majority of vpr-1(tm1411); Ex (ges-1p:: FLAG:vpr-1) (28/39) hermaphrodites are fertile. Two lines of C. elegans expressing Ex (ges-1p::FLAG:vpr-1L156A) were developed. All line 1 (0/30) and most of the line 2 hermaphrodites are sterile (10/27) and there is no significant difference between both lines and vpr-1(tm1411). (all bars indicate a p <0.01, Chi-Square Test between the connected pairs; no bar indicates no significant difference between the pair) B) C. elegans expressing ges-1p::FLAG:vpr-1 with a L156A point mutation are mostly infertile, but some hermaphrodites present weak levels of fertility, as in Line 2 (*p<0.05, **p<0.01, ***p<0.001, ****p<0.0001, Student’s t-test) C-E) Most C. elegans that transgenically express ges-1p::FLAG:vpr-1 with a L156A point mutation fail to develop a gonad like wild type (C). A few, however, do develop a somatic U-shaped gonad and are fertile (E). Asterisk indicates vulva. The intestine is outline and pseudo-colored blue. The gonads are outlined and pseudo-colored yellow Images captured with a 20x multi-immersion Plan Fluor 0.8NA objective. Scale bar: 100 μm. F) VPR-1 localization is visualized by immunofluorescence of its N-terminal FLAG tag (red). Antibody staining against HDEL marks the ER (magenta). Phalloidin staining marks the intestinal lumen (green). FLAG-VPR-1 protein with a L156A point mutation (red) localizes throughout the intestinal cells with slight concentration along the basolateral membrane. Images captured with a 60x APO Tirf 1.49NA objective. Scale bar: 50 μm. G) FLAG-VPR-1 protein with a L156A point mutation (red) driven by ges-1p in the intestinal cells cannot be detected on the distal gonad. Asterisk indicates intestinal tissue where FLAG-VPR-1(L156A) is expressed. Images captured with a 60x APO Tirf 1.49NA objective. Scale bar: 50 μm.

Discussion

As type II ER membrane proteins, VAPs regulate diverse intracellular processes, such as lipid transport, membrane trafficking, and microtubule organization. However, recent studies from our lab and several other groups also reveal an unexpected function of VAP in extracellular signaling (11, 12, 14, 16). This non-cell-autonomous activity seems to be mediated by the conserved N-terminal MSP domain, which is proposed to be cleaved and secreted to signal to other cells. However, VAP proteins lack a signal peptide characteristic of proteins secreted through the ER-Golgi pathway, and the topology of VAP proteins posits the MSPd in the cytosolic side rather than facing the ER lumen (9). This indicates that VAPs do not utilize the conventional protein secretion pathway through the vesicular network, but instead adopt an unconventional mechanism to release its N-terminal domain after cleavage.

The evidence for secreted, non-cell-autonomous function of VPR-1 comes from genetic studies in C. elegans. VPR-1 functions as a permissive signal for gonad development (14). Using a tissue-specific transgene to rescue vpr-1 null mutants, our lab has shown previously that expression of vpr-1 in either the nervous system or the germ line strongly rescues deficiencies in gonadogenesis. However, expression of vpr-1 in muscle or hypoderm does not lead to phenotypic rescue (14). This suggests that secretion of VPR-1 might be regulated in a cell type-specific fashion. This notion is supported by work performed in Drosophila where a VAP homolog is also processed and secreted in specific populations of cells (11). In both Drosophila and C. elegans, the MSPd of VAP homologs can be released from the neurons (11, 12). However, no study has been reported to examine whether intestinally expressed vpr-1 also rescues the gonad defects in the vpr-1 null mutants via direct cleavage and secretion of the MSPd. Our immunofluorescent staining experiment shown in this study reveals that the intestinal FLAG-tagged VPR-1 is indeed secreted and travels to the distal gonad (Figure 3). No staining is observed in the proximal gonad, implying that the retention of tagged VPR-1 at the distal gonad is likely mediated by specific binding of VPR-1 N-terminal peptide to surface receptor(s) expressed at distal gonad. Though our interpretation of the observed FLAG signal is that it represents the MSPd of VPR-1 based on our data on VPR-1 cleavage, it is possible, though unlikely, that the staining represents a larger fragment or the full length VPR-1. A congruent C-terminal tag in the transgenic ges-1p::FLAG:vpr-1 line could help distinguish the cleaved N-terminal peptide from the whole VPR-1 protein in intestinal cells. Unfortunately, vpr-1 tagged simultaneously with double-FLAG on the N-terminus and double-HA on the C-terminus and transgenically expressed in either the intestine using the ges-1 promoter (Figure 1C) or neurons using the glr-5 promoter did not rescue the sterile phenotype of vpr-1 null worms. Our preliminary observations of these transgenic worms suggest that adding the C-terminal HA tag negatively affects the functionality of extrachromosomally expressed vpr-1. This result differs from when we tagged the endogenous protein in a similar way using CRISPR/Cas9-mediated knock-in of the epitopes, as the hermaphrodites expressing the tagged protein are fertile (Figure 4D). The difference in VPR-1 protein behaviors might be due to distinct expression levels of the protein from the endogenous locus versus from the regulatory elements of an intestinal gene. As VAP proteins can form homo- and hetero-dimers, high level expression might drive the protein to form aggregates. This notion agrees with previous reports on gain-of-function point mutations in the N-terminus of mammalian VAPB which cause protein aggregation and prevent secretion (11, 36–40). The ubiquitous expression of endogenously tagged VPR-1 protein prevents us from distinguishing between cleaved or full length VPR-1 in mediating non-cell-autonomous function of VPR-1. Although our N-terminally FLAG-tagged VPR-1 that is expressed in the intestine partially rescues gonadogenesis, the effect on the brood size of these vpr-1 null mutants bearing the transgene is quite modest compared to when the endogenous gene is tagged in vivo. This result may suggest that expression of vpr-1 in multiple tissue types might be required to achieve optimal extracellular VPR-1 levels to rescue the brood size to a greater extent. Future work involving the development of transgenic lines that simultaneously express differentially tagged vpr-1 under both intestinal and neuronal promoters can provide additional insight into regulation of MSPd signaling from different tissues.

In this work, we identify for the first time the cleavage site of VPR-1 in C. elegans. By mass spectrometry, we show that VPR-1 is cleaved at leucine 156 to release an N-terminal peptide that consists of the entire MSPd sequence and a short linker between MSPd and CCD. The alignment of VAP sequences across multiple species reveals that amino acids around L156 seem to be conserved (Supplemental Figure 2). The mammalian homologs of VPR-1 in human and mouse harbor a conserved leucine residue at the amino acid position 157 (Supplemental Figure 2). Highlighted in this figure, the residues surrounding the cleavage site are similar in the other species (Supplemental Figure 2). Future work involving isolation and mass spectrometry of the short VAP products from these organisms will provide additional insight into possible conservation of VAP cleavage during evolution. Although we attempted to provide biochemical evidence to show an absence or dramatic decrease of MSPd 25 kDa protein due to the L156A substitution mutation, this was technically challenging given the maternal effect of vpr-1 and the small brood size of the rescued ges-1p::FLAG:vpr-1(L156A) animals.

The identification of the cleavage site of VPR-1 in C. elegans offers a great opportunity to gain further understanding of possible enzymes involved in VAP/VPR-1 processing. Unlike ligand processing in the secretory pathway where Furin family of proteases are often involved to recognize di-basic residues, the cleavage site at leucine does not conform to any consensus protease recognition motifs. This implies that the enzyme(s) involved might recognize certain amino acid features, such as bulky hydrophobic amino acids, rather than particular amino acid residues. Among the cytosolic proteases, chymotrypsin seems to have such characteristics. It is possible that two processing events occur to generate the N-terminal peptide ending at L156. C. elegans have 9 identified serine-type proteases known as the TRY family of proteins. The best studied of these proteases is TRY-5, which was identified as a secreted serine-protease that is required for the activation of male derived sperm in C. elegans (41). One of the TRY proteins could be acting in the cytosol to cleave the N-terminal MSPd of VPR-1 in the intestine. The potential cleavage by the protease might persist in VPR-1(L156A) though with reduced efficiency, hence the mutant hermaphrodites might have impaired, but not null, cleavage of VPR-1 in their intestinal cells. Therefore, though the released N-terminal MSPd cannot be detected in our immunofluorescence studies (Figure 6G), it might be cleaved in sufficient amounts to rescue gonadogenesis and fertility in a small percentage of animals (Line 2, Figure 6A,B).

Our data also bring up the issue of cell-specific regulation of VPR-1 cleavage and secretion. In theory, both steps can be regulated in a cell type specific fashion. Chymotrypsin-like enzymes exist in many cell types in C. elegans, but non-cell-autonomous rescue of vpr-1 null phenotypes occurs only when the protein is expressed in particular tissues. This suggests that either tissue-specific proteases participate in the cleavage event, the activity of the protease is regulated by a tissue-specific factor, or secretion of cleaved MSPd is controlled in a tissue-specific manner. It is interesting to note that the MSP protein after which the domain is named can also have non-cell-autonomous hormonal function to stimulate oocyte maturation, even though it is a cytosolic protein with no signaling peptide. The mechanism for MSP secretion is not completely understood, but it is tempting to speculate that MSP and MSPd might utilize a similar process to exit the cell and enter the extracellular space. If this is true, we may expect that cleavage of VAP is the rate limiting step, which might then be followed by constitutive secretion of the MSPd. This model can be tested by expression of a synthetic construct containing the N-terminal fragment only and followed by examination of its secretion and function. Alternatively, VAP cleavage and secretion might be linked intimately and both require tissue-specific regulation. This study was unable to discern between whether cleavage occurs at the ER or if the full VPR-1 protein is first trafficked to the plasma membrane before being cleaved to release the N-terminus extracellularly. Further studies will help to resolve the different models.

In summary, we present in vivo results in this work to show that the N-terminus of C. elegans VPR-1 is cleaved at a specific site at leucine 156 to release an N-terminal peptide that encompasses the MSPd (Figure 7). The MSPd-containing product is secreted from the intestinal cells into the pseudocoelom and bind the distal gonad to function as a non-cell-autonomous signal to regulate gonad development and hermaphrodite fertility (Figure 7). We demonstrate that mutation of the cleavage site results in absence of the gonadal MSPd signal and a much reduced ability of the mutant VPR-1 protein to rescue the vpr-1 null phenotype. Our identification of the novel cleavage sequence in VPR-1 in C. elegans paves the way for investigation of the cleavage enzyme(s) responsible for the unconventional processing and secretion of VAP. These results, together with our previously published work, open the opportunity to further explore the role of MSPd cleavage and secretion in neurodegenerative diseases such as ALS.

Figure 7: VPR-1 MSPd secretion and non-cell autonomous signaling model.

In the intestinal cells of C. elegans, MSPd is cleaved off VPR-1 at amino acid L156. The cleaving enzyme is not known. This cleaved MSPd is secreted out of the intestinal cells and into the pseudocoelom to travel and binds to the cell membrane of the distal gonad (dashed arrows).

Highlights:

• VPR-1 expression in the C. elegans intestinal cells partially rescues fertility.

• VPR-1 is secreted from the C. elegans intestinal cells and binds the distal gonad.

• VPR-1 N-terminal peptide is cleaved after leucine 156.