Receptor-mediated endocytosis in the Caenorhabditis elegans oocyte

Abstract

The Caenorhabditis elegans oocyte is a highly amenable system for forward and reverse genetic analysis of receptor-mediated endocytosis. We describe the use of transgenic strains expressing a vitellogenin::green fluorescent protein (YP170::GFP) fusion to monitor yolk endocytosis by the C. elegans oocyte in vivo. This YP170::GFP reporter was used to assay the functions of C. elegans predicted proteins homologous to vertebrate endocytosis factors using RNA-mediated interference. We show that the basic components and pathways of endocytic trafficking are conserved between C. elegans and vertebrates, and that this system can be used to test the endocytic functions of any new gene. We also used the YP170::GFP assay to identify rme (receptor-mediated endocytosis) mutants. We describe a new member of the low-density lipoprotein receptor superfamily, RME-2, identified in our screens for endocytosis defective mutants. We show that RME-2 is the C. elegans yolk receptor.

Figure 1

YP170:: GFP accumulates in vesicles of the oocyte. (A) Whole adult hermaphrodite expressing YP170::GFP. YP170::GFP fluorescence is found in the intestine, late-stage oocytes, and embryos. YP170::GFP localization appeared very similar to endogenous YP170 by fluorescence microscopy (this work) and immuno-electron microscopy (Hall et al., 1999). Immuno-electron microscopic experiments did reveal accumulation of YP170::GFP in enlarged vesicles of the intestine, unlike endogenous YP170 in wild-type worms (Hall et al., 1999). Such accumulation may result from overexpression of YP170 in transgenic strains or properties acquired from the GFP tag. OO, late-stage oocytes; V, vulva. (B) Confocal micrograph of YP170::GFP vesicles within two nearly full-grown oocytes of one gonad arm. (C) Confocal micrograph of YP170::GFP vesicles within a two-cell embryo. (D) Stylized drawing of one gonad arm connected to the spermatheca and the uterus.

Figure 2

Ce-rab5(RNAi) blocks endocytosis of YP170::GFP. (A) Fluorescence micrograph of late-stage wild-type oocytes (arrows). The most full-grown oocyte (far left) is heavily labeled with YP170::GFP vesicles, whereas the two earlier-stage oocytes show less labeling. SP, spermatheca. (B) Fluorescence micrograph of full-grown or nearly full-grown Ce-rab5(RNAi) oocytes (arrows). All YP170::GFP fluorescence is found between oocytes, in the spermatheca, pseudocoelom, or intestine. Extremely high levels of extracellular YP170::GFP are seen in the spermatheca (SP). No uptake of YP170::GFP was observed.

Figure 3

rme-2 mutants are completely deficient in yolk endocytosis. (A and B) Normarski (A) and fluorescence (B) micrographs of wild-type hermaphrodites expressing YP170::GFP. The three latest-stage oocytes, each with progressively more internalized YP170::GFP, are marked with arrows. (C and D) Nomarski (C) and fluorescence (D) micrographs of rme-2(b1005) hermaphrodites expressing YP170::GFP. Note the absence of YP170::GFP in the three latest stage oocytes (arrows) despite the nearly normal morphology of the rme-2 germ line. All YP170::GFP fluorescence is in the pseudocoelom or in the intestine.

Figure 4

RME-2 is a member of the LDLR superfamily of lipoprotein receptors (GenBank accession number AF185706). Repetitive motifs shown include acidic ligand-binding sequences with six cysteines each, EGF-like repeats with six cysteines each (B.1 and B.2 type; Herz et al., 1988), YWTD repeats, which are thought to form a β-propeller structure (Springer, 1998), and the consensus internalization signal FDNPXY. VLDLR/OVR, very low-density lipoprotein receptor (Gafvels et al., 1993)/chicken yolk receptor (Shen et al., 1993); Dm Yolkless, Drosophila yolk receptor (Schonbaum et al., 1995); Ce LRP-1/Megalin, C. elegans megalin gene product (Yochem and Greenwald, 1993). Only a partial map of lrp-1 motifs is shown because of its large size. For a complete description, see Yochem and Greenwald (1993).

Figure 5

RME-2 Western blot. Total C. elegans proteins were size separated by SDS-PAGE, transferred to a polyvinylidene difluoride membrane, and probed with anti-RME-2(EXT) affinity-purified antibodies. Identical results were observed with anti-RME-2(INT) affinity-purified antibodies (our unpublished findings). Equal protein loading was confirmed by Coomassie blue staining of duplicate gels (our unpublished findings).

Figure 6

Localization of RME-2 in the gonad and embryo of a YP170::GFP-expressing strain. (A–C) Confocal micrographs of a dissected gonad, showing YP170::GFP (green) and anti-RME-2 (red). Embryo and intestine YP170::GFP fluorescence is evident at the bottom of A and C. Embryos in B do not stain with anti-RME-2 antibodies, because their eggshells are not permeablized by the procedure used to fix and stain the dissected gonad. High-magnification (3000×) confocal micrographs of full-grown oocytes in a middle focal plane (D–F) or top focal plane (G–I) show abundant YP170::GFP and anti-RME-2 immunofluorescence. RME-2 and YP170::GFP signals do not appear to colocalize at steady state. RME-2 immunoreactivity patterns are very different in embryos. All RME-2 appears intracellular in the four-cell embryo (J) and declines progressively during embryogenesis.

Figure 7

Anti-RME-2 immunostaining of rme-2 mutants. (A and B) Confocal micrographs of wild-type dissected gonads stained with antibodies to the RME-2 intracellular and extracellular domains, respectively. Micrographs were produced using identical scan settings, allowing direct comparison of fluorescent intensities. Oocytes are indicated by brackets. Differences in staining in A and B result from differences in gonad age rather than differences in immunofluoresence pattern between antisera. The gonad in B is from a young adult, whereas that in A is from an older adult. (C and D) rme-2(b1005) mutant oocytes failed to stain with anti-RME-2(INT) antibodies (C) and showed very little reactivity to anti-RME-2(EXT) antibodies (D). rme-2(b1008) mutant oocytes gave very similar results (our unpublished findings). (E and F) rme-2(b1026) mutant oocytes showed strong anti-RME-2 immunofluorescence throughout the cell, possibly indicating ER retention.

Figure 8

Ectopic expression of RME-2 induces ectopic binding of YP170::GFP to RME-2-expressing cells. Shown is anti-GFP (A) and anti-RME-2 (B) immunolabeling of an adult hermaphrodite body wall muscle cell in a strain of the genotype lin-15(n765); bIs1[vit-2::gfp, rol-6(d)]; Ex[myo-3::rme-2, lin-15(+)].

Figure 9

YP170::GFP and RME-2 localization after RNAi of certain endocytosis factors. Confocal micrographs of oocytes in dissected gonads from RNAi animals show YP170::GFP fluorescence (green) and anti-RME-2 immunostaining (red). Mid, middle focal plane; Top, top focal plane. Ce-clathrin heavy chain (Ce-chc) RNAi (A–C) resulted in accumulation of YP170::GFP and RME-2 at the cell surface and more diffuse surface anti-RME-2 staining (E) than wild type (Figure 6). Ce-rab11(RNAi) oocytes showed reduced internalized YP170::GFP (G) and more punctate, less diffuse RME-2 staining (H and K) than wild-type. Ce-rab7(RNAi) oocytes showed YP170::GFP accumulation in large peripheral vesicles (M) and RME-2 accumulation (N) in these same large vesicles. Some of the vesicles containing both YP170::GFP and RME-2 are marked with arrows.