RNAi screen in Drosophila cells reveals the involvement of the Tom complex in Chlamydia infection

Abstract

Chlamydia spp. are intracellular obligate bacterial pathogens that infect a wide range of host cells. Here, we show that C. caviae enters, replicates, and performs a complete developmental cycle in Drosophila SL2 cells. Using this model system, we have performed a genome-wide RNA interference screen and identified 54 factors that, when depleted, inhibit C. caviae infection. By testing the effect of each candidate's knock down on L. monocytogenes infection, we have identified 31 candidates presumably specific of C. caviae infection. We found factors expected to have an effect on Chlamydia infection, such as heparansulfate glycosaminoglycans and actin and microtubule remodeling factors. We also identified factors that were not previously described as involved in Chlamydia infection. For instance, we identified members of the Tim-Tom complex, a multiprotein complex involved in the recognition and import of nuclear-encoded proteins to the mitochondria, as required for C. caviae infection of Drosophila cells. Finally, we confirmed that depletion of either Tom40 or Tom22 also reduced C. caviae infection in mammalian cells. However, C. trachomatis infection was not affected, suggesting that the mechanism involved is C. caviae specific.

Figure 1. C. caviae Replicate in Drosophila SL2 Cells

Immunofluorescence images of Drosophila SL2 cells incubated for 1, 24, 48, 72, and 96 h with C. caviae. After fixation, the samples were stained with the DNA dye Hoechst (DNA, blue) and a FITC-conjugated monoclonal C5+C8 antibody directed against MOMP and LPS (C. caviae, green). Merge: overlay of the two images.

Figure 2. C. caviae Undergo a Full Developmental Cycle in Drosophila SL2 Cells

(A) Electron micrograph of semi-thin sections (∼70 nm) of Drosophila SL2 cells 45 h post infection with C. caviae. Black square: Area magnified and shown in (B). Bar: 2 μm. (B) Higher magnification of the bacteria of the C. caviae inclusion shown in (A). Bar: 500 nm. (C) C. caviae were isolated from Drosophila SL2 cells 3, 24, 48, 72, 96, 120, and 144 h post infection and were used to infect HeLa cells for 24 h. The percentage of HeLa cells that present an inclusion is shown.

Figure 3. C. caviae TTSS Is Functional in Drosophila SL2 Cells

Drosophila SL2 cells infected for 48 h with C. caviae, were stained with the DNA dye Hoechst (DNA, blue) and polyclonal antibodies directed against the IncA proteins (IncA, red). Merge: overlay of the two images. N: nucleus; Inc: inclusion. Bars: 10 μm.

Figure 4. Genome-Wide RNAi Screen to Identify Host Factors Involved in C. caviae Infection

(A) Schematic representation of the screening procedure. dsRNA were incubated with Drosophila SL2 cells for 3.5 d before incubation with C. caviae for 48 h. After fixation and staining, images were acquired using an automated microscope. (B) Illustration of C. caviae growth inhibition phenotype that was selected in the primary screen for further analysis. Drosophila SL2 cells were incubated with Tom40 dsRNA (Tom40 RNAi) or with buffer alone (No RNAi) for 3.5 d and were subsequently incubated with C. caviae for 48 h. The samples were fixed and stained with the DNA dye Hoechst (DNA, blue) and a FITC-conjugated monoclonal C5+C8 antibody directed against MOMP and LPS (C. caviae, green). Merge: overlay of the two images.

Figure 5. Functional Categories of the 162 Candidates Identified in the Primary Screen

Based on their predicted biological functions, the 162 candidates identified in the primary screen were categorized in 14 functional groups. The number surrounding the chart indicates the number of candidates in each category. See Table S1 for the list of candidates.

Figure 6. Tom40 and Tom22 Depletion Reduce C. caviae Infection of Mammalian Cells

(A) HeLa 229 cells were transfected for 3 d with a CDH1 control siRNA (Ctrl), or four siRNA duplexes against Tom40 or Tom22 either pooled (Mix) or individually (1, 2, 3, 4) and analyzed by immunoblotting using antibodies directed against Tom40, Tom22, or actin. (B and C) Immunofluorescence images of HeLa 229 cells transfected for 3 d with CDH1, Tom40, or Tom22 siRNA and subsequently incubated for 24 h with C. caviae. After fixation the cells were stained with the DNA dye Hoechst (DNA, blue) and polyclonal antibodies against C. caviae (C. caviae, green). Merge: overlay of the two images. (B) Low magnification images (10×). (C) High magnification images (100×). (D) Quantification of the percentage of large C. caviae inclusions upon CDH1, Tom40, or Tom22 siRNA. Inclusions were defined as objects whose size ranged from 10 to 150 μm². Objects whose size ranged from 30 to 150 μm² were defined as large inclusions. Mix: pool of four siRNAs; 1, 2, 3, 4: individual siRNA.

Figure 7. Tom40 Depletion Affects C. caviae Inclusion Size and Differentiation into EBs

Electron micrographs of HeLa229 cells depleted for CDH1 (top panels, CDH1 siRNA) or Tom40 (bottom panels, Tom40 siRNA) 24 h post C. caviae infection. Representative images of small (left panels) and large (right panels) inclusions are shown. Bar: 2 μM.

Figure 8. Tom40 and Tom22 Depletion Reduces the Production of C. caviae, but Not C. trachomatis, Infectious Progeny

Quantification of the infectivity of C. caviae (A) or C. trachomatis (B) progeny isolated from CDH1 or Tom40- or Tom22-depleted cells. The progeny was isolated 48 h post infection and dilutions were incubated with fresh HeLa 229 cells for 24 h. After fixation and staining, the IFUs/ml was determined by assessment of the number of infected cells. Cc: C. caviae; CtL2: C. trachomatis.