Nuclear pre-mRNA 3'-end processing regulates synapse and axon development in C. elegans

Abstract

Nuclear pre-mRNA 3'-end processing is vital for the production of mature mRNA and the generation of the 3' untranslated region (UTR). However, the roles and regulation of this event in cellular development remain poorly understood. Here, we report the function of a nuclear pre-mRNA 3'-end processing pathway in synapse and axon formation in C. elegans. In a genetic enhancer screen for synaptogenesis mutants, we identified a novel polyproline-rich protein, Synaptic defective enhancer-1 (SYDN-1). Loss of function of sydn-1 causes abnormal synapse and axon development, and displays striking synergistic interactions with several genes that regulate specific aspects of synapses. SYDN-1 is required in neurons and localizes to distinct regions of the nucleus. Through a genetic suppressor screen, we found that the neuronal defects of sydn-1 mutants are suppressed by loss of function in Polyadenylation factor subunit-2 (PFS-2), a conserved WD40-repeat protein that interacts with multiple subcomplexes of the pre-mRNA 3'-end processing machinery. PFS-2 partially colocalizes with SYDN-1, and SYDN-1 influences the nuclear abundance of PFS-2. Inactivation of several members of the nuclear 3'-end processing complex suppresses sydn-1 mutants. Furthermore, lack of sydn-1 can increase the activity of 3'-end processing. Our studies provide in vivo evidence for pre-mRNA 3'-end processing in synapse and axon development and identify SYDN-1 as a negative regulator of this cellular event in neurons.

Fig. 1.

C. elegans sydn-1 mutants exhibit abnormal synapses and axons. (A) rpm-1 animals exhibit grossly normal movement. sydn-1;rpm-1 animals are small and exhibit a severe Unc phenotype. sydn-1 animals appear normal in size and are mildly Unc. (B) SNB-1::GFP in the dorsal nerve cord of GABAergic motoneurons. WT puncta appear round and evenly spaced (arrowheads). In sydn-1 animals, puncta are unevenly shaped (arrowhead) and fewer in number. Quantification is shown to the right. N indicates the number of animals. Error bars indicate s.e.m., here and in all subsequent figures. ^***, P<0.001; Student's t-test. (C) Soma expression of P_unc-25SNB-1::GFP in GABAergic motoneurons is unaltered in sydn-1 animals. Quantification is shown to the right. N indicates the number of cell bodies. ns, P>0.3; Student's t-test. (D) SNB-1::GFP in the dorsal nerve cord of cholinergic motoneurons. In sydn-1, puncta are fewer and unevenly distributed. (E) GABAergic neuron morphology is grossly normal in sydn-1 mutants; however, axonal GFP brightness is noticeably reduced in some regions. The double-headed arrow indicates a small ectopic branch. (F) sydn-1 mutant axons extend ectopic branches. Images in B-F are shown with anterior to the left, ventral down. CB, cell body. Scale bars: 5 μm in B,E; 2 μm in C; 10 μm in D; 1 μm in F.

Fig. 2.

Endogenous synaptic protein expression is altered in sydn-1, sydn-1;syd-2 and sydn-1;rpm-1 mutants. (A,B) Confocal projections of nerve ring (top row) and dorsal nerve cord (bottom row) stained with anti-SNT-1 (A) or anti-ELKS-1 (B). Staining intensity was noticeably decreased in sydn-1 and was highly exacerbated in sydn-1;syd-2 double mutants. Scale bars: 100 μm. (C) The number of ELKS-1-positive puncta in the dorsal cords. N indicates the number of animals. ns, not significant; ^***, P<0.0001; Student's t-test. (D) Quantification of the size of ELKS-1-stained puncta in the dorsal cords. Animals are the same as those in C. N indicates the number of puncta analyzed. Box indicates s.e.m.; the top and bottom lines indicate the minimum and maximum puncta sizes, respectively. ^***, P<0.0001; ^**, P<0.01; ns, not significant; non-parametric test (GraphPad Prism).

Fig. 3.

sydn-1 mutants show increased numbers of small axonal profiles. (A) Electron micrographs of ventral nerve cord cross-sections in WT and sydn-1 animals. Right bundle is indicated by the white outline, the left bundle by the black outline. Small axons are marked in yellow. (B) Magnified view of the red-boxed regions in A. Markedly small axons are marked with an asterisk. (C) Reduced axon diameters in sydn-1 animals. N indicates the number of axons. ^***, P<0.001; Student's t-test. (D) Increased number of neurite profiles in sydn-1 mutants. N indicates the number of animals. P=0.07; Fisher's exact test. Scale bars: 400 nm in A; 200 nm in B.

Fig. 4.

sydn-1 synergizes with mutations in synapse development genes. (A) SNB-1::GFP in the dorsal cord of GABAergic motoneurons. syd-2, rpm-1 and sydn-1 exhibit abnormally shaped puncta (left). sydn-1;syd-2 and sydn-1;rpm-1 double mutants exhibit reduced numbers of puncta (right, with quantitation in bar chart). N indicates the number of animals. ^***, P<0.001; Student's t-test. (B) GFP-labeled GABAergic neuron axons (ventral down and anterior left). sydn-1;syd-2 double mutants display ectopic and excessive axon branches. A magnified view is shown on the right. (C) Quantification of the percentage of neurons per animal that display excess GABAergic neuron axon branches. N indicates the number of animals. ^*, P<0.01; ^***, P<0.001; ns, P>0.2; Student's t-test. Scale bars: 5 μm.

Fig. 5.

SYDN-1 is a novel nuclear protein that acts in the nervous system. (A) The C. elegans sydn-1(ju541) locus is at chromosome I: –8.96. (B) The sydn-1 gene structure. Left straight line, promoter; black boxes, exons; bent lines, introns. ju541 contains a 1 kb deletion (purple box). (C) The sydn-1 synapse phenotype is rescued by transgenic expression of genomic sydn-1 (P_sydn-1), pan-neuronal-driven sydn-1 (P_rgef-1) and GABAergic neuron-driven sydn-1 (P_unc-25) cDNA. (D) Synaptic puncta in transgenically rescued sydn-1 animals are similar to those of WT (see Fig. 1) in shape, intensity and spacing. Quantitation in the central dorsal cord of transgenically rescued lines is shown to the right. N indicates the number of animals. ^***, P<0.001; Student's t-test. (E) Epifluorescence images of the ventral nerve cord of sydn-1(ju541);Ex(P_rgef-1FLAG::SYDN-1) animals immunostained for FLAG and co-stained with DAPI. Arrows indicate regions of intense SYDN-1 and weak DAPI staining. (F) Confocal stacked images of sydn-1;Ex(P_rgef-1FLAG::SYDN-1) animals stained for FLAG show speckled (left) or ring (right) patterns. Scale bars: 5 μm in D; 2 μm in E,F.

Fig. 6.

Loss of function of pfs-2 suppresses sydn-1. (A) C. elegans pfs-2(ju608) was mapped to chromosome II: +22.9. (B) The pfs-2 gene structure (key as Fig. 5). ju608 contains a G-to-A nucleotide transition in the third exon, causing an arginine-to-glutamine mutation. (C) In sydn-1;pfs-2, SNB-1::GFP puncta in the dorsal nerve cord of GABAergic neurons appear normal. Scale bar: 5 μm. Quantification of SNB-1::GFP is shown to the right. Overexpression of ju608 (pfs-2OE) in sydn-1 does not significantly change the number or characteristics of mutant puncta. N indicates the number of animals. ^***, P<0.0001; ns, P>0.3; Student's t-test. (D) In sydn-1;syd-2;pfs-2, axons are normal in morphology and GFP intensity. Expression of pfs-2(WT) rescues the suppression activity of pfs-2 (right-hand panel). (E) Illustrations of transgenic constructs (right) and a summary of transgenic rescue (left). (F) Quantitation of pfs-2(ju608) suppression activities on ectopic branches. Suppression activity of pfs-2(ju608) is not rescued by expression of only the PFS-2 WD-repeat domain. N indicates the number of animals. ^***, P<0.001; ^*, P<0.01; Student's t-test. (G) Alignment of PFS-2 with homologs. S. cerevisiae Pfs2p, NM_001018284.1; C. elegans PFS-2, Z83120.1; A. thaliana FY, NM_121351.4; D. melanogaster CG1109, NM_169091.1; H. sapiens WDR33, NM_018383.3. Gray bar denotes a region that is homologous to C. elegans PFS-2 (∼57% to ∼70% similarity from yeast to human). The ju608 mutation occurs in a completely conserved arginine in the first WD repeat. A phylogram of PFS-2 homologs is shown at the bottom. (H) RNAi knockdown of pfs-2 suppresses ectopic axon branches in sydn-1;eri-1;syd-2;P_unc-25GFP. N indicates the number of animals. ^***, P<0.001; Student's t-test.

Fig. 7.

SYDN-1 regulates PFS-2 abundance. (A) Single-plane confocal images of the head region of WT animals expressing a P_pfs-2 PFS-2::GFP; P_rgef-1 FLAG::SYDN-1 transgene, immunostained for anti-GFP, anti-FLAG and stained for DAPI. PFS-2 and SYDN-1 are colocalized to the nucleus in neurons (arrow). PFS-2 is seen in non-neuronal nuclei (asterisk) due to expression from its endogenous promoter. (B) Confocal single-plane images of WT animals expressing a P_pfs-2 PFS-2::GFP; P_rgef-1 FLAG::SYDN-1 transgene show partial colocalization between PFS-2::GFP and FLAG::SYDN-1. PFS-2 and SYDN-1 together appear speckled (left) or as a smooth ring (right). (C) Confocal images of P_pfs-2 PFS-2::GFP immunostained with anti-GFP in WT or sydn-1 (top). P_rgef-FLAG::SYDN-1 immunostained with anti-FLAG is not changed in pfs-2 mutants (below). Quantitation of fluorescence intensity is shown on the right. N indicates the number of nuclei. ^***, P<0.001; ns, P>0.5; Student's t-test. Scale bars: 5 μm in A; 2 μm in B,C.

Fig. 8.

NpolyA activities are elevated in sydn-1 mutants. (A) RNAi knockdown of several polyadenylation components suppresses ectopic branching in sydn-1;eri-1;syd-2;P_unc-25GFP animals. The percentage of ectopic branches is the percentage of neurons per animal that display excess axon protrusions. Vector control is the L4440 RNAi vector. N indicates the number of animals. ^***, P<0.0001; ^**, P<0.001; ns, P>0.07; Student's t-test. (B) The NpolyA site insertion in lin-15B in lin-15(n765ts) animals (Cui et al., 2008). (C) pfs-2 suppresses the Muv phenotype of lin-15(n765ts) animals. N indicates the number of animals. ^***, P<0.001; Student's t-test. (D) sydn-1;lin-15(n765ts) animals display statistically increased numbers of vulval protrusions. N indicates the number of animals. Average number of vulval protrusions: lin-15(n765), 2.5±0.07; sydn-1(ju541);lin-15(n765), 3.1±0.07; average ± s.e.m.; P<0.00001, Student's t-test. (E) Examples of Muv phenotypes. Vulval protrusions (asterisk). Scale bar: 50 μm. (F) qRT-PCR of lin-15A transcript levels normalized to ama-1 transcript levels and displayed relative to WT. ^**, P<0.01; ^*, P<0.05; Student's t-test.