The fusogen EFF-1 controls sculpting of mechanosensory dendrites

Abstract

The mechanisms controlling the formation and maintenance of neuronal trees are poorly understood. We examined the dynamic development of two arborized mechanoreceptor neurons (PVDs) required for reception of strong mechanical stimuli in Caenorhabditis elegans. The PVDs elaborated dendritic trees comprising structural units we call "menorahs." We studied how the number, structure, and function of menorahs were maintained. EFF-1, an essential protein mediating cell fusion, acted autonomously in the PVDs to trim developing menorahs. eff-1 mutants displayed hyperbranched, disorganized menorahs. Overexpression of EFF-1 in the PVD reduced branching. Neuronal pruning appeared to involve EFF-1-dependent branch retraction and neurite-neurite autofusion. Thus, EFF-1 activities may act as a quality control mechanism during the sculpting of dendritic trees.

Fig. 1

PVD menorah sculpting by EFF-1. “c” denotes cell body, small droplets are autofluorescent gut granules, anterior is left, and dorsal is up. (A) Full stereotypic arborization pattern in young adult. (B) PVD in eff-1(hy21) showing disorganized menorahs. (Insets) Wild-type menorah (left) and eff-1(hy21) mutant (right) with excess branching from secondary stem (arrow) and abnormal retrograde migration of quaternary branch (asterisk). Scale bars represent 50 μm [(A) and (B)] and 10 μm [(B) insets]. (C) Menorah pattern map [primary to senary (1° to 6°) are blue, purple, red, green, orange, and yellow, respectively]. (D) Excess bifurcations from secondary stems in eff-1(hy21ts). (E) Increased retrograde branches in eff-1(hy21ts). In (D) and (E), *P < 0.0001, two-tailed t test. Data are mean ± SE. All experiments were compared with wild type at 20°C, n = 20, 260 menorahs. hy21 at 25°C, n = 6, 355 menorahs. hy21 at 15°C, n = 10, 322 menorahs. n, number of animals analyzed. We also performed a one-way analysis of variance (ANOVA) followed by Dunnett’s multiple comparison test (11).

Fig. 2

EFF-1 autonomously rescues PVD arborization and restricts branching. (A) eff-1(hy21) at 25°C had twice as many secondary and tertiary branches as the wild type. Scale bars represent 10 μm. (B) Menorahs in wild-type animals. (C) Expression of EFF-1 (des-2p::EFF-1) in the PVD in eff-1(hy21) mutants at 25°C partially rescued the menorah pattern and reduced the number of branches. (D) Overexpression of des-2p::EFF-1 in wild-type animals caused a menorah gradient starting from the PVD cell body until dendrites disappeared toward the posterior (arrows). (E) Apparent EFF-1 dosage–sensitive reduction in branching. *P < 0.0001, **P < 0.001, ***P < 0.05, two-tailed t test. We also performed ANOVA (11). Data are mean ± SE. eff-1(hy21), n = 6; number of wild-type animals, n = 6; eff-1(hy21);des-2p::EFF-1, 25°C, n = 5; des-2p::EFF-1, n = 6. (F) Quantification of the number of secondary to senary branches showed that secondary and tertiary branches were doubled in the mutant compared with wild type and that secondary to quinary branches were reduced when EFF-1 was expressed ectopically in the PVD. (G) A model for the maintenance of PVD branching in an eff-1 dosage–dependent manner. Low levels of eff-1 (A) increased the number of PVD branches, and elevated levels of eff-1 reduced branching (D).

Fig. 3

Dendrite autofusion and fission trim menorahs. Scale bars represent 5 μm, (A) and (B); 20 μm and insets 5 μm, (C); and 0.5 μm, (D). (A) Trimming of dendrites that break off (arrows) and fission of branches in des-2p::EFF-1 animal (arrowheads). m, body wall muscle. (B) Fission of a branch in des-2p::EFF-1 animal (arrowhead). (C) eff-1(hy21ts) animal grown at 25° downshifted to 15° for 4 hours and then upshifted to 25° for 2 hours. Loops can be seen (arrows and insets). (D) Four TEM prints from wild-type animal “N2U” (6) showing the route of a PVD fusion process at the edge of the ventral nerve cord. Prints 1572 and 1597 show quaternary dendrites (arrowheads, red pseudocolor) passing between a thin layer of hypodermis and body wall muscle (M). PVD branch emerged from one quaternary dendrite at 1572, joined the minor fascicle of the ventral cord between 1578 and 1592, and ran ventralward to autofuse with another quaternary dendrite at 1596 and 1597 (arrows). Whereas this fusion process of PVD spans about 3 μm along the anterior-posterior axis, the marked neuron processes (42, 51, 58, and 96) were traced for thousands of serial sections back to their cell bodies. Dorsal is down (fig. S9; www.wormimage.org). (E) Cartoon representing one menorah (red) with eight terminal dendrites autofused along the dorsal midline (arrows).

Fig. 4

EFF-1–dependent retraction of branches. (A, B, D, and E) Time-lapse confocal projections of L4 and young-adult animals. (A) Growth of a wild-type menorah (movie S3). Growing tertiary branches (arrows). (B) Retraction of branches in des-2p::EFF-1 animal (arrowheads). (C) Number of branches growing and retracting in des-2p::EFF-1 and wild type. Branches first showed dynamic movements but eventually either shortened or lengthened (dynamic branching ending with growth or retraction). **P < 0.01, ***P < 0.05, two-tailed t test. Data shown as mean ± SE. Wild type, n = 3; des-2p::EFF-1, n = 2. (D) L4 eff-1(hy21ts) at 25°C, branches were static. (E) Retracting branch in L4 eff-1(hy21ts) grown at 25°C and shifted to 15°C for 4 hours (arrowhead; movie S1). (F) Number of branches showing growth and retraction in eff-1(hy21) grown at 25°C and downshifted as in (E). *P < 0.001, **P < 0.01, two-tailed t test. Data are mean ± SE. eff-1(hy21) 25°C, n = 3; eff-1(hy21) 25° to 15°C, n = 4 animals. (G) Wild-type branches grew (0.8 nm/s; N = 5 branches) and retracted (0.9 nm/s; N = 3 branches). In eff-1(hy21) temperature-shifted animals, branches retracted (0.9 nm/s; N = 10 branches). Data are mean ± SE. Scale bars represent 5 μm.