Fig 1.
The ESCRT machinery assembles on LdLPVs.
(A).The scheme for expressing recombinant fluorophore-tagged components of the ESCRT machinery in RAW264.7 macrophages is shown. Transfected cells on coverslips were infected for 24, 48 or 72 hrs with metacyclic L. donovani parasites. B) Representative images of transfected and infected cells are shown. Parasite and host cell nuclei are labeled with DAPI (blue). C) The proportion of Leishmania parasitophorous vacuoles with colocalization of the fluorophore-tagged iESCRT molecule was plotted. At least 100 parasite vacuoles were counted per coverslip time point. Counts were performed in duplicate on coverslips for at least three independent experiments. One-way ANOVA with multiple comparisons followed by post hoc Tukey’s honest significant difference test to determine statistical significance was performed. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. N. S, not significant.
Fig 2.
Assessing the effect of VSP4-E228Q on the retention of ESCRT on LPVs.
(A). The scheme for assessing the effect of expressing the VPS4a dominant negative on ESCRT III distribution in transfected and infected cells. Macrophages were transfected with pEGFP-VSP4-E228Q and pLNCX2-mCherry-CHMP4B, plated on coverslips, and infected with L. donovani. Cells on coverslips were fixed at 24, 48, and 72 hours. B) Representative images of uninfected and infected cells are shown. (C) LPVs harboring parasites in VPS4-E228Q-transfected cells were scored for CHMP4B mCherry reactivity. (D) The course of infection in transfected cells was assessed at 24, 48, and 72 hours post-infection. Counts were obtained in duplicate on coverslips for at least three independent experiments. One-way ANOVA with multiple comparisons followed by post hoc Tukey’s honest significant difference test to determine statistical significance was performed. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. N. S, not significant. Graphs were generated in GraphPad Prism*8. Illustration was generated with Biorender.
Fig 3.
LLOMe-induced damage of LdLPV suggests that damage-dependent ESCRT recruitment is mechanistically different from the ‘normal’ scheme of ESCRT recruitment to LdLPVs.
(A). The experimental scheme is shown. Cells were transfected with either pLNCX2-mCherry-CHMP2B or pLNCX2-mCherry-CHMP4B and plated on coverslips. They were then infected for 48 hrs, after which they were incubated with LLOMe (1mM or 5mM) or vehicle for 30 min. Cover slips were recovered when the drug was washed out (time 0) and then subsequently after 10-minute intervals. (B). Representative images of LLOMe-treated CHMP4B-transfected cells without infection are shown (red circles are due to CHMP4B in damaged lysosomes). Images from CHMP4B transfected and infected cells treated with either the vehicle or the drug are shown. LAMP1 labeling was used to delineate the contours of LPVs. (C) The number of LPVs that were positive for CHMP4B after the pulse and chase of LLOMe is shown. (D) A graph of the proportion of positive cells at the 0 time is shown. (E) The number of LPVs that were positive for CHMP2B after the pulse and chase of LLOMe is shown. (F) A graph of the proportion of CHMP2B-positive cells at the 0 time is shown. These graphs were compiled from two experiments, each with three coverslips for each point. Graphs were generated in GraphPad Prism*8 and statistical analysis was performed. A one-way ANOVA with multiple comparisons, followed by a post hoc Tukey’s honest significant difference test, was performed to determine statistical significance. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. N. S, not significant. Illustration was generated with Biorender.
Fig 4.
PI(4)P and PI(3,4)P2 are displayed on LdLPVs.
RAW264.7 macrophages were transfected with pMRXIP-GFP-P4M-SidMx2 (PI4P biosensor), NES-eGFP-cPHx3 (PI(3,4)P2 biosensor), NES-EGFP-PH-ARNO2G-I303Ex2 (PI(3,4,5)P3 biosensor). Cells on coverslips were infected with L. donovani. Representative images of 48-hour-infected macrophages labeled for the detection of the biosensors and LAMP1 are shown. White arrows indicate the locations of the biosensor label. These images are representative of two experiments.
Fig 5.
Generation of cells expressing limiting levels of ESCRT components.
(A). The scheme for generating stable lines with knocked-down levels of TSG101 or ALIX is shown. Alix shRNA plasmid, or TSG101 shRNA plasmid, was transfected into macrophages and then plated. Twenty-four hrs after plating, the selection of cells by growth in puromycin commenced. Oligoclonal cell lines were identified, and the testing scheme for these cell lines is presented. The control line was transfected with a control short hairpin RNA (shRNA) plasmid. (B). Oligoclonal lines were tested for the expression of TSG101 or ALIX by Western blotting. Representative blots and densitometric analysis of blots to test for the expression of TSG101 or ALIX are shown. Plots were generated from densitometry of blots from at least 2 experiments. Graphs were generated in GraphPad Prism*8, where unpaired student T test statistical analysis was performed. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. N. S, not significant. Illustration was generated with Biorender.
Fig 6.
Knockdown of TSG101 or ALIX in RAW264.7 macrophages reveals the critical role of ALIX.
(A). Selected oligo-clonal lines of RAW264.7 ALIXKD, TSG101KD, and shCTRL cells were transfected with the pLNCX2-mCherry-CHMP4B. They were then infected with L. donovani parasites. Coverslips were recovered after 24, 48, or 72 h of infection and then processed to detect CHMP4B recruitment to LdLPVs. (B). Representative images from 48 hrs infected shCTRL, TSG101KD, or ALIXKD are shown. Contours of the LdLPV membrane were detected by LAMP1 labeling. The parasite and host cell nuclei were labeled with DAPI. ALIXKD clone 2 was not transfected with CHMP4B-mCherry. (C) CHMP4B recruitment to LdLPVs in KD cell lines was quantified as described in the Materials and Methods section. Colocalization of LAMP1 with CHMP4B was determined. At least 100 LdLPVs were measured per coverslip, per treatment, and time point. (D) The number of parasites per macrophage in infected KD cell lines was enumerated after 24, 48, or 72 hours of infection. At least 100 macrophages were counted per coverslip, treatment, and time point. (E) The LdLPV sizes in KD lines after 24, 48, and 72 hours of infection were measured and plotted. The diameter of LdLPVs was measured using the “Insert” scale tool of the BZ-X800 analyzer. At least 100 LdLPV sizes were measured per coverslip, time point, and treatment. Data was compiled and graphed using GraphPad Prism 8. One-way ANOVA with multiple comparisons followed by post hoc Tukey’s honest significant difference test to determine statistical significance was performed. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. N. S, not significant. Illustration was generated with Biorender.
Fig 7.
The ESCRT machinery is recruited to LdLPVs in THP-1 cells.
After the differentiation of THP-1 monocytes into macrophages, they were infected with metacyclic promastigote forms of L. donovani. At 24, 48, and 72 hrs post-infection, cells on coverslips were fixed and processed by immunofluorescence labeling for the detection of representative components of the ESCRT I complex (TSG101) and the ESCRT III complex (CHMP4B). A) shows a representative cell labeled for the detection of TSG101 (red label), LAMP1 (green label), and parasite and host nuclei detected by DAPI labeling (blue label). An LdLPV that is dividing (yellow arrow and inset) is shown. TSG101 is in the plane of scission that bridges the emerging daughter LdLPVs. B) shows a representative cell labeled for the detection of CHMP4B. An LdLPV is highlighted (yellow arrow and inset) wherein nucleated CHMP4B is at the center of an LdLPV between two parasites. C) The plot shows the proportion of LdLPVs with the recruitment of TSG101 and CHMP4B. Counts were performed in duplicate on coverslips for at least three independent experiments. Illustration was generated with Biorender.
Fig 8.
BrdU incorporation into intracellular parasites shows that parasite replication is impaired in ALIXKD.
(A). The experimental scheme is shown. After 6 hrs of infection, BrdU is added to the culture medium. After 24 hrs of infection, the culture medium was washed out and replaced with BrdU-free medium. Parasites that initiate replication are expected to incorporate BrdU, which is lost upon successive rounds of replication. (B) Representative images of microscopic evaluation of BrdU incorporation (red label) at 24 and 72 hrs in shCTRL, TSG101KD, and ALIXKD are shown. White arrows in the images point to representative examples of labeled parasites. C) The percentage of parasites that were labeled positively for BrdU at the indicated times was plotted. At least 100 parasites were counted per coverslip, treatment, and time point. A one-way ANOVA was performed, and statistical significance was determined by post hoc Tukey’s honest significant difference test. Data was compiled and graphed using GraphPad Prism 8. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. N.S, not significant. Illustration was generated with Biorender.
Fig 9.
Graphical summary: LdLPVs display phosphoinositides that are the membrane anchors for components of the ESCRT machinery.
ALIX is the critical component that is recruited. Members of the ESCRT I complex are recruited, but their recruitment is dispensable. Components of the ESCRT III complex and VPS4 are recruited subsequently to complete the activation of the ESCRT machinery. The scission of LdLPVs completes the replication of parasites and LdLPVs. When ALIX expression is limited, LdLPV fission is defective, and so too is parasite replication. This results in a reduction in the parasite burden. Illustration was generated with Biorender.