Fig 1.
CCL2 exposure to C. elegans results in delayed development and enlargement of the intestinal lumen.
(A) Wild-type C. elegans fed on control (empty vector) E. coli reached the L4 larval stage 44 h after hatching, whereas worms fed on CCL2-expressing E. coli only reached larval stage L2 (stereo microscopy). (B,C) Wild-type L4 larvae kept on control or CCL2-expressing E. coli for 24 h. The detectable developmental delay was evident already after 24 h (B; stereo microscopy). CCL2 also caused an enlarged intestinal lumen in these larvae (C; DIC microscopy). Scale bar: 0.5 mm (A-B), 20 μm (C); inset: 2x magnification of the lumenal section of the intestine.
Fig 2.
CCL2 binds to and alters the brush border of the intestine without being internalized.
(A-D) C. elegans L4 larvae expressing PGP-1::GFP were fed with a non-toxic (100 μg/ml; A) or toxic (500 μg/ml; B-D) concentration of CCL2-TAMRA (tetramethylrhodamine: red) for the indicated hours and observed using confocal microscopy. (A) CCL2-TAMRA colocalized with PGP-1::GFP at the apical intestinal plasma membrane. (B-D) Toxic concentrations of CCL2-TAMRA led to a time-dependent alteration of the brush border without any detectable uptake into intestinal cells. Scale bar: 10 μm; inset: 2x magnification of the lumenal section of the intestine.
Fig 3.
CCL2 leads to microvilli loss, intestinal plasma membrane invaginations, and terminal web gaps.
Wild-type C. elegans L4 larvae were fed on control (empty vector) (A,C,E) or CCL2-expressing E. coli (B,D,F-H) for the indicated hours and then observed under a transmission (A-D, G,H) or focused ion beam scanning electron microscope (E,F). (A,C) The intestinal apical surface of control animals showed the typical dense brush border made of microvilli (asterisk) and glycocalyx (arrow), and an intact terminal web (arrowhead) underlying the intestinal apical plasma membrane. (B,D) Exposure to CCL2 prompts missing microvilli (asterisks); invaginations of intestinal apical plasma membrane filled with lumenal content (arrows); and disruption of the terminal web (arrowheads). (E,F) 3D reconstruction of selected microvilli in control (E) and CCL2-treated (F) animals showing the disorganized brush border following CCL2 exposure; fused microvilli (yellow; filled arrow); cell border (open arrow); adherens junction (arrowhead). (G,H) Morphological changes are already visible at the ultrastructural level after 3 h (G) and 6 h (H) of exposure to CCL2; invaginations of plasma membrane (filled arrows); terminal web (filled arrowheads); dark cap at the tip of microvilli (open arrows); intact actin filament bundles (open arrowheads); depolymerized actin filament bundles (asterisks). Scale bar: 500 nm.
Fig 4.
CCL2 alters the structure of the intestinal brush border and terminal web in living animals.
(A-E) C. elegans L4 larvae of the indicated genotype were fed for 24 h on control (empty vector) or CCL2-expressing E. coli. Various fluorescent transgenic reporters were used to detect changes in intestinal cell architecture by confocal microscopy. (A) PGP-1::GFP is expressed at the intestinal apical plasma membrane. (B) RAB-8::GFP labels the intestinal cytoplasm. (C) ERM-1::GFP links the intestinal apical plasma membrane to the actin filaments within the microvilli. (D) ACT-5::GFP labels actin filaments within the microvilli and the terminal web underlying the intestinal apical plasma membrane. (E) IFB-2::CFP labels intermediate filaments in the terminal web. Scale bar: 10 μm; inset: 2x magnification of the lumenal section of the intestine.
Fig 5.
CCL2 does not induce pore formation in the intestinal apical plasma membrane.
(A-C) C. elegans L4 larvae expressing PGP-1::GFP were fed on control (empty vector pQE30) (A), Cry21A- (B) or CCL2- (C) expressing E. coli for 24 h, transferred into wells containing propidium iodide (PI; red) for 2 h and observed using confocal microscopy. Cry21A is a pore-forming toxin of B. thuringiensis [31]. PI entered the intestinal cytoplasm of Cry21A-fed (B), but not of control- (A) or CCL2- (C) fed animals. Scale bar: 10 μm.
Fig 6.
CCL2 is not toxic in the absence of bacteria.
(A-D) C. elegans L4 larvae expressing PGP-1::GFP were exposed to the indicated TAMRA-labeled protein (red; BSA (bovine serum albumin); B-D) for 24 h in the presence (A,B) or absence (C,D) of B. subtilis and observed using confocal microscopy. Only the combination of CCL2-TAMRA and B. subtilis led to toxicity (B). Scale bar: 10 μm; inset: 2x magnification of the lumenal section of the intestine; n/a = not applicable.
Fig 7.
Intestinal expression of α1,3-fucosylated N-glycan cores is sufficient for toxicity of CCL2.
(A) Structure of the C. elegans CCL2 core glycotarget [10,25]. (B) Alleles conferring resistance to CCL2 recovered after transposon (Mos1)- or EMS mutagenesis. (Chr.: chromosome; Mutation: base pair change and its position (position 1: “A” of start codon ATG) in the genomic sequence as well as the resulting amino acid change. (C) Intestinal expression of wild-type bre-1, ger-1, or fut-1 in the respective C. elegans mutant was sufficient to restore sensitivity to CCL2. All constructs were under the control of the intestinal promoter ges-1 and C-terminally tagged with mCherry (red). Transgenic and control animals were grown on CCL2-expressing E. coli for 3 days. Data shown are average ± standard error of the mean of three experiments. Asterisks (*): After 3 days, all worms were L2 larvae.