Fig 1.
let-653 acts in the excretory duct and pore cells, not the canal cell.
(A-D) let-653 mutants have morphological defects in the duct and pore. (A,B) Schematics of excretory system morphology in WT (A) and let-653 mutants (B) at the early L1 stage. G1 pore cell is shown in blue, duct cell in yellow, canal cell in red, and G2 and W epidermal cells in pink. White regions represent lumens. Heavy black lines represent apical junctions. Black arrow, pore cell autocellular junction (AJ); red arrow, missing AJ in let-653 mutant. Arrowheads, duct-canal intercellular junction (IJ), *, duct lumen dilation. Anterior is to the left and ventral is down in all images. (C, D) Duct (d) and pore (p) morphology in WT and let-653(cs178) mutants. Cell bodies (red in C”, D”) and all apical junctions (green in C”, D”) are marked as labeled. DIC, differential interference contrast of head; box shows region magnified in subsequent panels. c, canal cell nucleus. ph, pharynx. (E) LET-653b protein schematic and locations of mutant allele lesions. SP, signal peptide. PAN, PAN-Apple domains. ZP, zona pellucida domain. Conserved ZP domain cysteines are indicated above in black; additional cysteines are indicted in grey. PTS, Proline/Threonine/Serine-rich region is predicted by Net-O-Glyc [99] to be highly O-glycosylated. LET-653b is the shortest of three protein isoforms produced from the let-653 locus by alternative splicing; the other isoforms contain a larger central PTS domain (www.wormbase.org). CCS, consensus furin cleavage site (R-X-R/K-R) [100]. (F) Tissue-specific expression of let-653b cDNA in the duct and/or pore efficiently rescued let-653(s1733) lethality. Data from multiple independent transgenic lines per construct are shown. *, p<0.01, Fisher’s Exact test, compared to non-transgenic siblings.
Fig 2.
let-653 is required to maintain lumen integrity during duct tube elongation and narrowing.
(A) Time course of duct lumen dilation and G1 pore AJ disappearance in let-653(cs178) mutants. Lumen dilation was visualized by DIC as a small bubble at least 1 μm in diameter adjacent to the canal nucleus, and the AJ was visualized with AJM-1::GFP as in panel B. n≥30 for each timepoint. (B) Lumen dilation initiates near the duct-canal junction. DIC and inverted grayscale AJM-1::GFP images of a let-653(cs178) 3-fold embryo at early stage of lumen dilation. Note presence of pore AJ (arrow) and duct-canal junction (arrowhead). (C) Duct lumen measurements reveal two distinct phases of lumen growth–elongation and widening (I), followed by elongation with narrowing (II). Duct lumen length and width measurements in WT (black or green) and let-653(cs178) mutants (red). Each dot represents a single animal containing RDY-2::GFP and AJM-1::GFP markers as in D-I, except 1.5-fold measurements based on LET-653::SfGFP marker, and green dots based on WT TEM specimens. For let-653, only young 3-fold embryos with an intact duct lumen were scored. The appearance of lumen fragmentation and dilation prevented accurate measurements in older animals. (D-I) In let-653 mutants, the distal duct lumen fragments during elongation and narrowing. Bracket indicates duct lumen. Black arrow indicates pore AJ. Arrowhead indicates duct-canal junction. Red arrows point to lumen or junction discontinuities. Asterisks indicate duct lumen dilation. Inverted grayscale fluorescent images of staged WT (D-F) and let-653(cs178) (G-I) animals expressing the apical membrane marker RDY-2::GFP and junction marker AJM-1::GFP. RDY-2::GFP also marks the canal lumen, which shows up prominently near the duct-canal junction in some images. F and I also include the cytoplasmic duct marker lin-48pro::mRFP. (J-N) The pore loses its AJ and intercellular junctions in let-653(cs178) L1 larvae. (J,K) F-actin visualization using confocal microscopy of VAB-10(ABD)::GFP. F-actin localizes near apical membranes and junctions in both WT (J) and let-653 mutants (K). (L-N) Pore AJ loss and duct-pore separation after hatch. Inverted grayscale images of let-653 mutants expressing the cytoplasmic marker dct-5pro::mCherry and junction marker AJM-1::GFP. Scale bars, 5 μm.
Fig 3.
TEM analysis of let-653 mutant excretory system.
TEM sections of the excretory duct or pore region in WT (A-C) or let-653 mutants (D-G), pseudocolored to indicate the duct (d, yellow), G1 pore (p, blue), G2 (pink), canal cell (c, red) and excretory gland cell (gl, green). Dotted lines indicate basement membrane of the pharynx (ph). (A) Lateral section from a normal, late 3-fold stage embryo, showing the duct cell body and the luminal connection between the duct and canal cells. Black arrowheads indicate intercellular junctions, and lines indicate lumen. Multiple luminal segments appear in a single cross-section due to the winding path of the wild-type duct lumen. (B) Transverse section through the duct process in a normal L1 larva. White arrowhead indicates the cuticle lining of the duct lumen. (C) Excretory pore opening from the same animal as in A. (D, D’) Skewed lateral section through a let-653(cs178) late 3-fold stage embryo. A portion of the duct lumen (asterisk) and the canal cell lumen are greatly dilated behind a region devoid of lumen. Most of the duct cell body and the duct nucleus are outside the plane of this section; the central nucleus belongs to a different cell. Boxed region in D is magnified in D’; this region corresponds to the beginning of the duct process (compare to B). This region lacks a continuous lumen, but instead contains multiple smaller membrane-bound structures (black arrows). (E-F) Transverse sections from the duct processes of two let-653(s1733) L1 larvae. The lumen is fragmented (E) or contains abnormally small (E’) or large (F) rings of cuticle-like material (compare to B). (G) The G1 pore is intact and has a normal lumen diameter (compare to C) in this mutant L1. Scale bars, 500 nm.
Fig 4.
let-653 is required for proper alae formation and vulva lumen shaping.
(A,B) Comparison of body shape in WT and let-653(cs178) L4 larvae rescued to viability with a duct-specific lin-48pro::LET-653::SfGFP transgene. The mutants are slightly shorter and fatter than WT. (C,D) TEM images of alae in WT and let-653(s1733) L1 larvae. Transverse sections through the head region at the posterior bulb of the pharynx are shown. Boxed regions are magnified in adjacent panels. Arrowed bars indicate width of the alae. (C’, D’) let-653 mutants have alae and cuticle organization defects. White dotted line indicates edge of the epidermal cell layer. In let-653 mutants, the striated layer of the cuticle (black bracket) appears thin or absent, and the space between this layer and the underlying epidermis is expanded and filled with diffuse, lightly staining material. (E) The permeability barrier is intact in let-653 mutants. L1 larvae were incubated in 2 ug/ml Hoechst dye in M9 for 15 minutes and then rinsed and scored for nuclear fluorescence. let-4(mn105) was used as a positive control for permeability [38]. Data from three replicates shown, n>15 per genotype for each experiment. No significant difference found between let-653 and WT permeability by Student’s t-test, two-tailed. (F-I) Comparison of L4 vulva lumen morphology in WT (F), duct-rescued let-653 mutants (G,H) and a hypomorphic chondroitin biosynthesis mutant sqv-5(k175) [101]. Animals were staged as mid-L4 based on fusion of the anchor cell and horizontal orientation of the ventral vulva “fingers”. Modest vulval expansion defects or asymmetries were observed in 11/31 of mutant larvae.
Fig 5.
LET-653 is a component of a transient pre-cuticular apical ECM.
(A-B) Translational fusions of LET-653 to SfGFP rescue let-653(cs178) mutant lethality and show a cyclic pattern of accumulation in the duct lumen that precedes cuticle deposition. Expression data are from fusions visualized in an otherwise wild-type background. Protein domains are schematized as in Fig 1E. Green triangle represents SfGFP and pink triangle represents signal sequence (ss). n≥15 animals for each timepoint. AEL, after egg lay. Grey shading indicates phases I and II of duct lumen elongation as in Fig 2C. (C) ssSfGFP alone does not accumulate in the duct lumen. (D-F) At 1.5-fold, LET-653 fusions accumulate between the embryo and the eggshell, and in duct (lines), buccal (‘b’) and rectal luminal (‘r’) regions, while ssSfGFP accumulates only between the embryo and eggshell. Epifluorescent and corresponding DIC images are shown. (G-I) LET-653 fusions, but not ssSfGFP alone, re-accumulate in the late L1 larval duct lumen. Confocal slices. (G’,H’,I’) lin-48pro::mRFP marks the duct cell. (J-L) In the L4 vulva lumen, LET-653 accumulates near the apical membrane (lines) and along fibrils within the luminal core (arrow). ssSfGFP alone has no such preferential localization. Confocal slices. D’,E’,F’,J’,K’,L’ show DIC images for comparison. (C-L) WT background. Scale bars, 5 μm.
Fig 6.
LET-653 is cleaved at its C-terminus.
Anti-GFP Western blot performed on 1.5-fold transgenic embryos. Full-length LET-653 fusions were expected to be found at 100kD, or higher if post-translationally modified. These were not detected, but fusions lacking the PAN domains were readily observed. C-terminally tagged LET-653 yields a band just below 40kD (arrowhead), indicating that SfGFP and about 80 amino acids of LET-653’s C-terminus have been cleaved from the full-length protein, consistent with cleavage at the predicted CCS. The loading control antigen is UNC-15/paramyosin. Blot shown is representative of n = 11.
Fig 7.
TEM visualization of the vulval aECM.
A) TEM image of a wild-type mid-L4 vulva, showing two discrete aECM compartments. Boxed regions are magnified in panels i and ii. Red dot in panel ii is an artifact of original annotation. Fuschia lines outline cells in the vulA-F rings. B, B’) Confocal slices focused to the dorsal vulva, showing ssSfGFP::LET-653 localization near the apical membrane. Asterisks indicate intracellular accumulation of the fusion protein. C) Schematic representation of LET-653 localization to distinct luminal matrix pools.
Fig 8.
LET-653 PAN and ZP domains confer distinct localization patterns and mediate duct lumen cycling and protection, respectively.
(A-C) LET-653 fusions lacking the C-terminal, ZP or PAN domains showed differential functions and localization in the duct. Only LET-653(ZP) rescued mutant defects. (D,G,J) LET-653(∆C) accumulated intracellularly rather than extracellularly, suggesting a failure in secretion. (E-F) LET-653(PAN) and LET-653(ZP) fusions localized normally to the duct in 1.5 fold embryos. (H-I) LET-653(PAN) accumulated in the late L1 larval duct lumen, but LET-653(ZP) did not. Confocal slices. (G’,H’,I,) lin-48pro::mRFP marks the duct cell. (K) LET-653(PAN) associated with fibrous material in the center of the lumen (arrow) (L) LET-653(ZP) associated with the dorsal apical membrane region. D’,E’,J’,K’,L’ show DIC images for comparison. Scale bars, 5 μm.
Fig 9.
LET-653 is sufficient to expand the gut lumen.
(A-D) Representative 1.5-fold embryos after heat shock overexpression of full-length LET-653 (A), LET-653(PAN) (B), LET-653(PAN+ZP-N) (C) or LET-653(PAN+ZP-N+mucin) (D) fusions under the hsp16.41 promoter. Gut lumen expansion is visible by DIC in panel A. (A’-D’) GFP expression. Tagged protein was expressed and secreted to similar degrees for the first three genotypes, but secreted more variably for the final genotype. White brackets flank lumen edges at the widest point along the gut length. Scale bar, 5 μm. (A”-D”) Schematic outlines of the entire embryo and gut (lumen shaded dark gray). Narrow regions alternate with wide “bulbs” (red arrows), suggesting a local expansion effect. (E) Individual plotted measurements of gut lumen width in microns. n = 13 for full-length LET-653 and n≥21 for the others. Median, 25th percentile, and 75th percentile are shown for each. *** p<0.001, Student’s t-test, two-tailed.
Fig 10.
Dynamic PAN- and stable ZP-mediated interactions in the duct lumen.
Fluorescence recovery after photobleaching (FRAP) of LET-653 translational fusions in the duct lumen. (A-C) Pre-bleach, bleach, and post-bleach frames taken from FRAP experiments on the indicated fusions in 1.5-fold embryos (A-A”) or L1 larvae (B-B”,C-C”). For each experiment, the post-bleach frame chosen corresponds to a time point after recovery plateau had been reached. Mean fluorescence intensity measurements were taken within the bleach ROI (white box) and, in L1 experiments, an adjacent ROI (purple box). Scale bars, 5 μm. (D-F) Fluorescence recovery curves plotting mean and SE for n≥5 replicates per stage for ssSfGFP::LET-653 and n = 2 replicates for ssSfGFP::(PAN). t = 0s represents the first post-bleach frame; raw values for each experiment were normalized to the highest measurement taken. (G, H) Comparisons of mobile fractions and recovery half-times. Full-length LET-653 (D,E) exhibited a significantly lower average mobile fraction than LET-653(PAN) (F). However, the half-time of recovery (t1/2) was not significantly different between the two fusion proteins. ** p<0.01, * p<0.05, Student’s t-test, two-tailed.
Fig 11.
The LET-653 C-terminus remains stably associated with the ZP-containing membrane-proximal matrix compartment.
(A-C) FRAP on two spatially distinct pools of full-length LET-653 in the L4 larva vulval lumen. (A) Schematic of ssSfGFP::LET-653 localization in the vulva. Boxes represent bleach ROIs shown as dotted white boxes in (B-B”,C-C”). Pre-bleach, bleach, and post-bleach frames taken from experiments on edge-associated (B) and central (C) compartments. Recovery of the edge region is minimal. (D-F) FRAP on ssSfGFP::LET-653(PAN) (D-D”), LET-653(ZP)::SfGFP (E-E”), or ssSfGFP alone (F-F”). Scale bars, 5 μm. (G) Fluorescence recovery curves plotting mean and SE for n≥2 replicates per genotype and region. (H, I) Comparisons of mobile fractions and recovery half-times. ZP-containing fusions in the edge domain are significantly less mobile than fusions in the central domain, irrespective of genotype or tag placement. Faster recovery in the edge experiments may be attributed to fast-moving protein in the adjacent diffuse pool, some of which was included in the ROIs chosen (B, E’). *** p<0.001, * p<0.05, Student’s t-test, two-tailed.
Fig 12.
let-653 functions in parallel to lpr-1.
(A) Comparative timecourse of lumen dilation and pore AJ loss in let-4(mn105), lpr-1(cs73), and let-653(cs178) mutants. let-653 single mutant data are reproduced from Fig 2A. n at least 17 animals per genotype for each timepoint. (B) let-4(mn105) and (C) lpr-1(cs207) embryos bearing the LET-653::SfGFP transgene. LET-653::SfGFP is visible in the duct/pore lumen (lines). Scale bars, 10 μm. (D) Double mutant analysis. ** p<0.002, *p<0.01, Fisher’s Exact test, compared to single mutants at same timepoint.
Fig 13.
Summary and model for the role of LET-653 in tube morphogenesis.
(A) Model for LET-653 aECM assembly. Upon LET-653 secretion, PAN-mediated interactions with the luminal matrix (purple) capture LET-653 and prevent its rapid outflow from the lumen. The ZP-C and/or C-terminal regions then interact with a membrane-associated factor (pink rectangle) that anchors LET-653 more tightly. A membrane-associated protease (scissors) cleaves LET-653 at the CCS, allowing ZP-N dependent polymerization and/or recruitment of additional matrix factors; however, the C-terminal peptide (black squiggle) remains tightly affiliated with the matrix. Arrangement of LET-653 polymers is hypothetical; see text for alternative models. (B) Role of LET-653 in duct tube maintenance. During phase I of duct tube morphogenesis, membrane-associated LET-653(ZP) (blue) resists (T-bars) stretching and/or constrictive forces (arrows) to maintain a uniform lumen diameter and allow fluid passage. During phase II, LET-653 is gradually cleared from the lumen, and a LET-653-dependent cuticle structure begins to form and take over the lumen shaping role. Finally, once tube morphogenesis is complete, the cuticle matures. In let-653 mutants, the lumen becomes too narrow during elongation and fragments into separate compartments (arrow). Upstream of the fragmentation, the lumen dilates (asterisk) as a secondary consequence of fluid accumulation.