Fig 1.
Actin patches on the CCV membrane are enriched for late endocytic vesicles and fusion regulatory proteins.
(A and B) Colocalization of endosomal components with CCV actin patches. Vero cells fixed at 3 dpi and stained for F-actin, early endosomes (EEA1+), late endosomes (CD63+), or lysosomes (LAMP1+). CCV actin patches cluster with late endocytic markers. Clustering with large, individual vesicles is also seen (white arrows). (C and D) Colocalization of fusion regulatory proteins with actin patches. Histograms depict the means ± SD of ≥ 60 cells for at least 3 independent experiments. Statistical significance was determined using Student’s t-test (****P <0.0001). Colocalization was determined using Pearson’s correlation coefficient. NMII, C. burnetii Nine Mile phase II strain. Scale bar, 5 μm.
Fig 2.
CCV actin patches facilitate vesicle fusion via docking and clustering of late endocytic vesicles to the CCV membrane.
(A) LAMP1+ vesicles fuse with the CCV at actin patches. Live cell images of a Vero cell at 3 dpi expressing lysosome-GFP (LAMP1) and actin-RFP. (B) Latrunculin A (LatA) treatment redistributes LAMP1+ clusters at actin patches around the CCV to the juxta-nuclear region (white arrows). Live cell images of a Vero cell at 3 dpi expressing lysosome-GFP (LAMP1) and actin-RFP. (C and D) LatA treatment redistributes VAMP7+ clusters at actin patches to the juxta-nuclear region (white arrows). Vero cells at 3 dpi were treated with LatA for 10 min prior to fixation and immunostaining for F-actin, VAMP7, and NMII. Histogram depicts the mean intensity ± SD of ≥ 60 cells for at least 3 independent experiments. Statistical significance was determined using Student’s t-test (****P <0.0001). NMII, C. burnetii Nine Mile phase II strain. Scale bar, 2.5 μm.
Fig 3.
Formation of CCV actin patches depend on secretion of effectors by the C. burnetii Dot/Icm type 4B secretion system.
(A, B, and C) Chloramphenicol reversibly causes CCV shrinkage and elimination of actin patches and associated clusters of CD63+ vesicles. Vero cells at 2 dpi were treated with chloramphenicol for 24 hr (+chlor) then fixed along with the corresponding 3 dpi untreated control (-chlor). Treated cells at 3 dpi were also washed to remove chloramphenicol and allowed an additional 24 hr recovery before fixation (recovery). Cells were fluorescently stained for F-actin and CD63. (D, E, and F) CCV actin patches and colocalization with CD63 are lost in cells infected with a C. burnetii dotA mutant. Vero cells infected for 24 hr with wild type C. burnetii or a dotA mutant were fixed and stained for F-actin and CD63. Asterisks mark the C. burnetii vacuole in the actin panels. Histograms depict the mean intensity of CCV actin or area ± SD of ≥ 60 cells for at least 3 independent experiments. Statistical significance was determined using Student’s t-test (****P <0.0001). Colocalization was determined using Pearson’s correlation coefficient. NMII, C. burnetii Nine Mile phase II strain. Scale bar, 5 μm.
Fig 4.
Retromer-associated actin regulators colocalize with CCV actin patches.
(A and B) CCV actin patches colocalize with Arp3, cortactin, WASH, and FAM21. The retromer protein VPS35 also colocalizes with patches whereas N-WASP does not. Vero cells at 3 dpi were immunostained for the indicated proteins. Histogram depicts the means ± SD of ≥ 60 cells for at least 3 independent experiments. Statistical significance was determined using Student’s t-test (****P <0.0001). Colocalization was determined using Pearson’s correlation coefficient. NMII, C. burnetii Nine Mile phase II strain. Scale bar, 5 μm.
Fig 5.
Retromer colocalizes with CD63+ vesicles that cluster with CCV actin patches.
(A and C) Vero cells were immunostained for retromer (VPS35+), early endosomes (EEA1+), or late endosomes/lysosomes (CD63+, LAMP1+). EEA1+ and CD63+ vesicles colocalize similarly with VPS35, while LAMP1+ vesicles are much less colocalized. (B and C) Vero cells 3 dpi, stained as in (A). Clustered CD63+ vesicles on the CCV membrane have high colocalization with VPS35 compared to EEA1+ vesicles. LAMP1+ vesicles on the CCV membrane also show increased colocalization with VPS35. Colocalization analysis of vesicles and CCVs was determined using Pearson’s correlation coefficient. Graphs represent the means ± SD of ≥ 60 cells from at least 3 independent experiments. Statistical significance determined by Student’s t-test (****P <0.0001). NMII, C. burnetii Nine Mile phase II strain. Scale bar, 5 μm.
Fig 6.
Knockdown of VPS35 reduces CCV actin patches and redistributes Rab7 with no effect on C. burnetii growth.
(A) Immunoblot of HEK 293 cells with knockdown of VPS35 by siRNA. A non-targeting pool (NT) siRNA was included. Actin was used as a loading control. (B, C, D, and E) Growth of C. burnetii in siVPS35 or NT-treated HEK 293 cells using qPCR to determine genome equivalents (GE). Three dpi VPS35 knockdown or NT-treated HEK 293 cells fluorescently stained for F-actin and Rab7. Histograms depict the mean intensity of CCV actin or area ± SD of ≥ 60 cells for at least 3 independent experiments. Statistical significance was determined by the Student’s t-test (****P <0.0001). NMII, C. burnetii Nine Mile phase II strain. Scale bar, 5 μm.
Fig 7.
WASH deficiency prevents CCV actin patch formation without affecting C. burnetii growth or CCV size.
(A, B, and C) WASHout MEFs (WASH-negative) do not have CCV actin patches and show uniform distribution of Rab7 on the CCV membrane. WASHf/f MEFs untreated or treated with 4-hydroxy-tamoxifen (WASHout) were fixed at 3 dpi and fluorescently stained for F-actin and Rab7. (D) Growth of C. burnetii in WASHf/f or WASHout MEFs using qPCR for determining genome equivalents (GE). Histograms depict the mean intensity of CCV actin or area ± SD of ≥ 60 cells for at least 3 independent experiments. Statistical significance was determined by the Student’s t-test (****P <0.0001). (E) WASHout MEFs (WASH-negative) do not have CCV actin patches and show uniform distribution of VPS35. Same treatment regime as (A) but fixed and fluorescently stained for VPS35 and actin. NMII, C. burnetii Nine Mile phase II strain. Scale bar, 5 μm.
Fig 8.
Arp2/3 is needed for CCV biogenesis.
(A and B) CK-666 treatment prevents CCV expansion. Vero cells at 24 hr post-infection were treated with CK-666 or chloramphenicol for 2 days. Cells were fixed and fluorescently stained for CD63 and F-actin. (C and D) CK-666 treatment causes CCV collapse. Vero cells at 2 dpi were treated with CK-666 for 24 hr, then fixed along with the corresponding 3 dpi DMSO-treated control or washed to remove CK-666 and allowed an additional 24 hr recovery before fixation. Cells were fluorescently stained for F-actin and CD63. Histograms depict the means of the percentage of cells with normal CCV expansion or CCV area ± SD of ≥ 60 cells for at least 3 independent experiments. Statistical significance was determined by the Student’s t-test (****P <0.0001). NMII, C. burnetii Nine Mile phase II strain. Scale bar, 5 μm.
Fig 9.
Arp2/3 is required for endocytic trafficking of transferrin in C. burnetii infected cells.
(A-C) Two dpi Vero cells treated with CK-666 or DMSO for 24 hr were incubated with 488-transferrin (488Tfr) for 1 hr, washed with PBS, then fixed and immunostained for CD63 and C. burnetii. CK-666 inhibits trafficking of 488Tfr to CCVs, decreasing the amount of 488Tfr on the CCV and colocalization with CCV CD63 clusters. Graphs represent the means ± SD of ≥ 60 cells from 3 independent experiments. Statistical significance was determined by the Student’s t-test (****P <0.0001). NMII, C. burnetii Nine Mile phase II strain. Scale bar, 5 μm.