Global Mapping of the Inc-Human Interactome Reveals that Retromer Restricts Chlamydia Infection

Abstract

Chlamydia trachomatis is a leading cause of genital and ocular infections for which no vaccine exists. Upon entry into host cells, C. trachomatis resides within a membrane-bound compartment—the inclusion—and secretes inclusion membrane proteins (Incs) that are thought to modulate the host-bacterium interface. To expand our understanding of Inc function(s), we subjected putative C. trachomatis Incs to affinity purification-mass spectroscopy (AP-MS). We identified Inc-human interactions for 38/58 Incs with enrichment in host processes consistent with Chlamydia's intracellular life cycle. There is significant overlap between Inc targets and viral proteins, suggesting common pathogenic mechanisms among obligate intracellular microbes. IncE binds to sorting nexins (SNXs) 5/6, components of the retromer, which relocalizes SNX5/6 to the inclusion membrane and augments inclusion membrane tubulation. Depletion of retromer components enhances progeny production, revealing that retromer restricts Chlamydia infection. This study demonstrates the value of proteomics in unveiling host-pathogen interactions in genetically challenging microbes.

Figure 1. Constructing the Chlamydia Inc-Human Interactome

(A) Predicted topology of Inc proteins in the Chlamydia inclusion membrane. (B) Workflow summary. (C) Overlap of Chlamydia prey with previously published AP-MS interactomes of HIV (Jager et al., 2011), HCV (Ramage et al., 2015), and KSHV (Davis et al., 2014). p-values determined by the hyper-geometric test. (D) Inc-Virus prey overlap. Parentheses contain the number of prey shared between Chlamydia Incs and HIV, KSHV and/or HCV. (E) Bar graph depicting the percentage of Inc prey. For the 354 prey identified for 38 Incs (“All”), 54% of prey were affinity purified by core Incs while 46% of prey were affinity purified by noncore Incs. Of the 98 prey shared by Chlamydia Incs and one or more virus (“with viral overlap”), there was a statistically significant increase in core Incs that overlap with viral prey (68%) versus noncore Incs (32%). There was no significant change in the percentage of prey bound to core versus noncore Incs for the 294 prey not shared with viruses (“without viral overlap”). p< 0.001 determined by the hyper-geometric test. See also Figure S1 and Table S1 and S3.

Figure 2. Network Representation of the Inc-Host Interactome

The high-confidence Inc-host network contains 38 Inc proteins (core Incs, dark grey; non-core Incs, light blue) and 335 unique human prey (light grey). Inc-human interactions (blue lines) were identified by AP-MS. Interactions between human proteins (dark grey lines) were curated from CORUM and STRING databases. Inc human prey in common with HIV, KSHV, or HCV prey (Davis et al., 2014; Jager et al., 2011; Ramage et al., 2015) are outlined in red. A subset of identified host complexes or proteins with similar functions are labeled. See also Table S2.

Figure 3. Predicted Functions for Specific Incs

Schematic diagram of the Chlamydia developmental cycle, indicating predicted functions of putative Incs derived from GO, KEGG, and PFAM enrichment terms from our entire PPI dataset (see Table S2). Core Incs, black; Noncore Incs, blue. We note that some predicted Incs have not been observed on the inclusion membrane at 24 hpi(Li et al., 2008, Dehoux, 2011 #3312). See also Table S1 and Figures S2 and S3.

Figure 4. IncE Interacts with Retromer SNX-BARs in vivo and in vitro

(A) Affinity purifications of HEK293T cells transiently expressing Strep-tagged Incs and immunblotted with the indicated antibodies. (B and C) IncE interacts with endogenous SNX6 (B) and transiently expressed SNX5-FLAG (C) in vivo. Immunoblot analysis of immunoprecipitations with anti-SNX6, -FLAG or -goat IgG in HeLa cells uninfected (−) or infected (+) with C. trachomatis (C.t.) for 24 hrs. Input represents 1% of lysates used for immunoprecipitation. Immunoblots are representative of 3 independent experiments. Untransfected, UT. (D) Schematic of IncE deletion constructs. The predicted transmembrane (TM) domains are shaded in black, the predicted N- and C-terminal cytosolic domains are shaded in grey. The numbers refer to amino acids. Constructs that co-affinity purify with SNX-BARs are indicated. (E) Affinity purifications of Strep-tagged IncE deletion constructs transiently expressed in HEK293T cells. (F) IncE_101-132 binds the PX domains of SNX5 and SNX6 in vitro. Purified 6xHis-MBP-Inc-Strep protein was immobilized to Strep-Tactin beads and incubated with the indicated purified 8xHis-SNX_PX, subjected to SDS-PAGE and visualized by Coomassie Blue stain. Arrow, IncE or IncD. Arrowhead, SNX_PX. *, Strep-Tactin. Molecular weight markers are indicated. See also Figure S4.

Figure 5. IncE Colocalizes with Retromer SNX-BARs on the Inclusion

(A–C) HeLa cells were infected with C. trachomatis (C.t.) for 24 hrs and analyzed by confocal microscopy for the localization of (A) endogenous SNX1/2/6 or transfected SNX5-FLAG and IncE, (B) SNX27-EGFP (pseudocolored red) and IncE or (C) endogenous SNX1 and VPS35. Panels are (A) single Z-slices or (B and C) maximum intensity projections of 0.3 μm Z-slices. I, inclusion. N, nucleus. UI, uninfected. C.t., C. trachomatis-infected. Arrowheads point to tubules. Scale bar = 10 μm. See also Figure S5.

Figure 6. IncE is Sufficient to Enhance SNX-BAR Recruitment and Inclusion Tubulation

(A–C) HeLa cells were infected with C. trachomatis serovar L2 transformed with pTet-IncE-FLAG or pTet-IncG-FLAG. As indicated, Tc was added at 1 hpi. At 20 hpi, cells were analyzed by confocal microscopy for (A) FLAG and (B and C), SNX6. (A and B) Maximal intensity projections of 0.2 μm Z-slices. Panel (C) shows maximal intensity projections and single XY slices from the boxed region in panel (B). Scale bar = 10 μm. (D and E) HeLa cells were infected with C. trachomatis serovar L2 transformed with empty vector or pIncE. At 15 hpi, cells were fixed and stained with anti-SNX6, IncE, and DAPI. Panel (E) shows enlargement of boxed region of Panel (D). All images are maximum intensity projections. Scale bar = 10 μm. (F) Quantitation of inclusion tubules per cell (dark grey bars) and length of the longest tubule per cell (light grey bars) (μm) in infected HeLa cells depleted of the indicated retromer components by siRNA. Data are mean +/− SEM from ≥3 independent experiments. *p < 0.001 compared to corresponding control, unpaired (two-tailed) t-test. See also Figure S6.

Figure 7. IncE_101-132-EGFP Disrupts Retromer Trafficking and Retromer Depletion Restricts C. trachomatis Infection

(A) Localization of endogenous CI-MPR is perturbed in HeLa cells transiently transfected with IncE_101-132-EGFP, but not IncD_95-141-EGFP, compared to untransfected (UT). Cells were fixed and stained with antibodies to TGN46, CI-MPR, and costained with DAPI. Shown are single Z-slices from confocal images. Scale bar = 10 μm. (B) Quantitation of TGN46 colocalization with CI-MPR in cells expressing IncE_101-132-EGFP or IncD_95-141-EGFP. Mean Pearson’s correlation coefficients from 2 two independent experiments (n~20 cells per experiment). *p < 0.05; one-way ANOVA, post hoc Tukey test. (C) CI-MPR remains localized to VPS35-positive compartments in cells expressing IncE_101-132-EGFP, but not IncD_95-141-EGFP. HeLa cells were fixed and stained antibodies to VPS35, CI-MPR, and costained with DAPI. Shown are single Z-slices from confocal images. White arrows, VPS35-positive compartment; Yellow arrows, CI-MPR-positive compartment. Left panel (Merge), scale bar = 5 μm. Right panels show enlargements of boxed areas. Scale bar = 2 μm. (D and E) Quantitation of infectious progeny (D) or primary infection (inclusion formation) (E) in HeLa cells infected with C. trachomatis serovar D following depletion of indicated retromer components. Data are mean +/− SEM from = 3 independent experiments. *p < 0.05 compared to control, unpaired (two-tailed) t-test.