The PAPS transporter PST-1 is required for heparan sulfation and is essential for viability and neural development in C. elegans

Abstract

Sulfations of sugars, such as heparan sulfates (HS), or tyrosines require the universal sulfate donor 3'-phospho-adenosine-5'-phosphosulfate (PAPS) to be transported from the cytosol into the Golgi. Metazoan genomes encode two putative PAPS transporters (PAPST1 and PAPST2), which have been shown in vitro to preferentially transport PAPS across membranes. We have identified the C. elegans orthologs of PAPST1 and PAPST2 and named them pst-1 and pst-2, respectively. We show that pst-1 is essential for viability in C. elegans, functions non-redundantly with pst-2, and can act non-autonomously to mediate essential functions. Additionally, pst-1 is required for specific aspects of nervous system development rather than for formation of the major neuronal ganglia or fascicles. Neuronal defects correlate with reduced complexity of HS modification patterns, as measured by direct biochemical analysis. Our results suggest that pst-1 functions in metazoans to establish the complex HS modification patterns that are required for the development of neuronal connectivity.

Fig. 1.

Molecular characterization of the pst-1/let-462 gene. (A) pst-1(ot20) was mapped between the two polymorphisms otP9 and otP10 on the proximal left arm of chromosome V. Transgenic rescue (solid line) or failure to rescue (dashed line) the cold-sensitive lethality of pst-1(ot20) with genomic pieces of DNA is indicated (see text). A PCR product spanning the predicted gene M03F8.2 can completely rescue the lethality. M03F8.2 is alternatively spliced and predicted to be the downstream gene in an operon with M03F8.3 (CEOP5092, WB195). (B) AIY interneurons (upper left panel) display an axon branching phenotype (white arrowhead) if kal-1 is overexpressed in AIY neurons (upper right panel). This phenotype is suppressed in pst-1(ot20) mutants (lower left panel) and reappears upon transgenic expression of wild-type pst-1 (lower right panel) in a pst-1(ot20) mutant background. Magnification, 400×. (C) Ability of various genomic fragments to rescue pst-1(ot20)-dependent suppression of the kal-1-induced overexpression phenotype in AIY interneurons. Genomic fragments are as indicated in A. (D) RNA-mediated interference (RNAi) against candidate genes in a transgenic strain (otIs76 mgIs18) overexpressing kal-1 under an AIY-specific promoter. The effect of the pst-1(ot20) genomic mutation is shown for comparison. Only RNAi against M03F8.2 (pst-1), but not other genes in the genomic region phenocopy the pst-1(ot20)-dependent suppression. RNAi against kal-1 serves as a positive control. n=50-100. Error bars denote the standard error of proportion. (E) Model of the predicted topology of the PAPS transporter PST-1. The different alleles and the predicted molecular changes are indicated. TM, transmembrane domain; G, glycine; E, glutamic acid; S, serine; P, proline; op, opal.

Fig. 2.

pst-1/let-462 is essential for viability and organogenesis. (A) Embryonic lethality of pst-1(s1956) null mutant animals. Animals lacking zygotic but not maternal pst-1 were identified by the absence of a fluorescent balancer (oyIs14V). Embryos lacking both maternal and zygotic pst-1 were identified as non-transgenic progeny from a transgenically rescued homozygous pst-1(s1956) mother. Multicopy transgenic arrays are not expressed in the C. elegans germline. Thus, non-transgenic animals from a transgenic homozygous mutant mother are devoid of all maternal products. All strains in this paper carrying the s1956 null allele have dpy-18(e364); unc-46(e177) in the background. The dpy-18(e364); unc-46(e177) background was used as the isogenic control. (B) Embryonic lethality of the pst-1(s1956) animals at different temperatures. Data for 20°C is identical to that for A and shown for comparison only. (C) Percentage of pst-1(s1956) animals arrested at different embryonic stages as indicated. Animals were scored 15 hours after an egg preparation, when 90% of control animals have hatched. (D) Control and s1956 mutant embryos at the comma stage and the threefold stage. Left panels show differential interference contrast pictures and right panels show hypodermal junctions for each genotype and embryonic stage. Hypodermal junctions are visualized with an integrated transgene of the junctional protein ajm-1. Magnification, 630×. (E) Differential interference contrast pictures of pst-1(s1956) null animals with molting defects (red arrowheads) at the first molt. Magnification, 630×. (F) pst-1(s1956) null mutant animals display a pharynx detachment phenotype. White arrowheads point at the attachment site of the pharynx to the tip of the nose and a red arrowhead points to the detached pharynx. Magnification, 630×. (G) Embryonic lethality of pst-1(ot20) mutants or wild-type animals was determined in the progeny of animals either shifted to the non-permissive temperature at the L3 larval stage or grown at 20°C. (H) Temperature-shift experiments with the pst-1(ot20) temperature-sensitive allele (gray line, n=100-163) in comparison with wild-type N2 (black line, n=50-102). The graph shows the percentage of animals alive after 4 days when shifted to the non-permissive temperature at the indicated developmental stages.

Fig. 3.

pst-1/let-462 mutants display neuronal guidance defects. (A) The nervous system in pst-1(s1956) null mutants lacking maternal and zygotic products displays no gross abnormalities compared with a control. Magnification, 400×. (B) The AIY interneurons have essentially wild-type axonal morphologies in pst-1(s1956) null mutants. A cell-positioning defect in pst-1 mutants was not quantified. Magnification, 400×. (C) Migration of the canal-associated neurons (CANs) appears largely independent of pst-1 (11.4% defects, n=96; compared to control 2.4% defects, n=41) whereas the migration of the hermaphrodite-specific neurons (HSNs) requires pst-1. A red arrow indicates a migration defect in HSN. All schematics are ventral views. Magnification, 160×. (D) Midline crossover defects in PVP and PVQ ventral nerve cord interneurons in pst-1(ot20) mutant animals are indicated by red arrows. Note that the defects in PVP (odr-2::gfp) and PVQ (sra-6::rfp) correlate in the majority of cases. Magnification, 160×. (E) PLM axon extension defects are indicated in pst-1(s1956) null mutants in a sublateral view (red arrow). Magnification, 400×. (F) pst-1(s1956) null mutants display defects in synaptic branch formation. The synaptic branch forms during the L1 stage in wild-type animals. Sufficient animals of the pst-1(s1956) null mutant are alive to allow scoring at the late L2 stage (Fig. 2). Magnification, 400×. (G) Localization of GFP-tagged synaptobrevin in the ventral nerve cord of D-type motor is defective in pst-1(s1956) null mutants compared with isogenic controls. Confocal images of ventral nerve cords are shown in lateral views. Anterior is to the right or left for s1956 and the control, respectively. Magnification, 630×. (H) Quantification of ventral nerve cord fluorescence (see Materials and Methods). pst-1 null mutant animals (n=7) have significantly reduced fluorescence compared with isogenic controls (n=7) (^*P<0.05, Student's t-test).

Fig. 4.

Cellular focus of pst-1 function. (A) Transgenic rescue of the cold-sensitive lethality of pst-1(ot20) mutant animals. Graphs show the percentage of transgenic animals that gave viable progeny after being shifted to the non-permissive temperature (15°C) at the L4 stage. Non-transgenic siblings did not give viable progeny (not shown). Data for the genomic region adjacent to pst-1 (PCR C, compare Fig. 1A) is shown as a negative control. Similar results were obtained using a different injection marker (supplementary material Fig. S6). (B) Transgenic rescue of pst-1(ot20)-dependent suppression of the kal-1-induced overexpression phenotype in AIY interneurons. The percentage of animals with branches in AIY interneurons is shown. White bars indicate the branching phenotype in a wild-type background, and black bars in a pst-1(ot20) mutant background. Gray bars show percentage branching in AIY interneurons of transgenic animals, in comparison with their non-transgenic siblings (black bars). Error bars denote the standard error of proportion. Statistically significant differences between transgenic animals and non-transgenic siblings are indicated (ns, not significant; ^*P<0.05; ^**P<0.005; ^***P<0.0005). (C) Transgenic rescue of the cold-sensitive lethality of pst-1(ot20) mutant animals with the pst-1a or pst-2 cDNAs under control of the pst-1a promoter as indicated. Quantification as in A. (D) Transgenic rescue of pst-1(ot20)-dependent suppression of the kal-1 induced overexpression phenotype in AIY interneurons with the pst-1a or pst-2 cDNAs under control of the pst-1a promoter as indicated. Quantification, color coding and statistics as in B.

Fig. 5.

Expression of pst-1 and pst-2. (A) Schematic of expression constructs. The 20kb genomic environs of pst-1 and pst-2 (Wormbase, WB203) are shown. A yellow fluorescent protein (YFP) with a nuclear localization signal (NLS) under control of an intercistronic region directing SL2-splicing (red) is inserted after the stop codon for pst-1a in fosmid WRM063bF12. A cyan fluorescent protein (CFP) with NLS is inserted after the stop codon for pst-2 in fosmid WRM0612bH01. Black boxes denote coding exons of pst-1 and pst-2, gray boxes of the preceding genes within the operons (indicated as green bars), and white boxes denote coding exons of surrounding genes. Predicted start codons and direction of transcription are indicated by a rectangular arrow and stop codons by an asterisk. Straight arrows indicate the direction of transcription of partially shown genes. Approximate extent of the fosmid clones in kilobases (kb) is indicated on the left and right, respectively. Gene names are as shown. SL2, splice leader 2; CEOP, C. elegans operon. (B) Expression of the pst-1 reporter at different developmental stages as shown. sc, seam cells; DIC, differential interference contrast; L1, first larval stage; phgc, pharyngeal gland cells. (C) Expression of the pst-2 reporter at different developmental stages as indicated. Note faint expression in the hypodermis (hypodermal ridge). (D) Expression of pst-1 (left panel) and pst-2 (middle panel) appears mutually exclusive during the L2 larval stage (right panel). in, intestine; L2, second larval stage. Magnification is 400× in all panels except for the adult in B, which is 160×.

Fig. 6.

Subcellular localization of PST-1 and PST-2. (A) Schematic of C-terminal fusions of PST-1A, PST-2 and of a mannosidase II fragment (MANS) that targets expression to the Golgi (Rolls et al., 2002) with the fluorescent proteins are indicated. All constructs are under control of the pst-1a promoter (supplementary material Fig. S7). Not drawn to scale. (B) PST-1A::YFP and PST-2::CFP display partial colocalization. Magnification, 630×. (C) The Golgi marker MANS::RFP displays partial colocalization with both PST-1A::YFP and PST-2::CFP. (D) Triple stain of PST-1A::YFP, PST-2::CFP and the MANS::RFP Golgi marker. Arrowheads denote lack of colocalization in an intestinal cell in all cases. All insets created with 2× digital zoom. Magnification, 630×.

Fig. 7.

pst-1/let-462 is required for complex HS modification patterns. (A) Suppression of the rol-6(e187) by RNAi-mediated knockdown against the genes indicated. RNAi experiments were conducted in the eri-1; lin-15b RNAi-hypersensitive background for maximum efficiency (Sieburth et al., 2005). (B) Suppression of rol-6(e187) by pst-1(ot20) at the non-permissive temperature. (C) Genetic interaction between pst-1(s1956) and the hst-6 hst-2 double-null mutant. The triple mutant is not more severe in the PLM-overextension phenotype than either the pst-1(s1956) or the hst-6 hst-2 double-mutant alone, indicating that all three genes act in the same genetic pathway. All strains carry dpy-18(e364); unc-46(e177) in the background. (D) Genetic interaction between pst-1(s1956) and the hst-6 hst-2 double-null mutant for the PLM-branching phenotype. (E) Chromatograms of HS disaccharides from control and pst-1(ot20) mutant worms. An asterisk denotes material that was not retained by the chromatographic column. D0A0, ΔUA-GlcNAc; D0S0, ΔUA-GlcNS; D0A6, ΔUA-GlcNAc6S; D2S0, ΔUA2S-GlcNS; D2S6, ΔUA2S-GlcNS6S. (F) Quantification of individual disaccharides from control (black bars) and pst-1(ot20) mutant worms (gray bars). In F and G, error bars denote the standard error (n=4 independent experiments for pst-1(ot20); n=5 independent experiments for control). (G) Sulfation extent is calculated as the mole fraction of O-sulfates divided by the mole fraction of sulfated disaccharides in wild-type (black bar) and pst-1(ot20) mutant worms (gray bar).