Fig 1.
ASFV uses both clathrin-mediated endocytosis and macropinocytosis to enter swine macrophages.
A) ASFV-infected swine macrophages (MOI 10) were pulsed for 10 min at 37°C transferrin conjugated to AF488 and 10-kDa dextran conjugated to AF555. After fixation, ASFV particles were labeled with an antibody against capsid protein p72 followed by an Alexa 647-conjugated secondary antibody. Note that incoming ASFV particles colocalized with either dextran, used as a macropinocytosis tracer (red arrows), or CME marker transferrin (green arrows). Bar, 5 μm. B) Macrophages pre-treated for 15 min with a number of inhibitors for clathrin-dependent endocytosis (chlorpromazine (CPZ), pitstop 2 (PTS2) and dynasore (DYN)) and macropinocytosis (EIPA, IPA-3 and cytochalasin D (CYT.D)) were incubated with DiD-labeled fluorescent ASFV particles (MOI 5) for 30 min. Then, the cells were incubated for an additional 30 min period in the presence of inhibitors and analyzed for virus uptake by flow cytometry (dark red colums). In a second set of experiments, macrophages were treated as above but infection was extended to 2.5 hpi to allow detection of the expression of early viral protein p32 by immunoblotting (light red columns). Data are expressed as relative fluorescence or p32 expression compared to a control infection (mean of three independent experiment ± SD). C) The expression of clathrin heavy chain (Chc) was silenced in macrophages by using two different siRNAs (Ch1, Ch2). Then, macrophages were infected with ASFV (MOI 5) for 2.5 h and analyzed by immunoblotting for Chc, ASFV protein p32 and ß-tubulin (Tub). No siRNA-treated infected cells (Asfv) and a scrambled (Scr) siRNA were used as controls. Numbers under the blot image indicate normalized expression levels for Chc and p32.
Fig 2.
ASFV endocytosis at the ultrastructural level.
Mock (A) and ASFV-infected (B and inset C) (MOI 100) swine macrophages were fixed at 20 mpi and analyzed by field emission scanning EM. The characteristic cell membrane protrusions of monocyte/macrophage lineage can be visualized in both conditions. ASFV particles (depicted in blue in inset C) are detected on flat membrane areas and also in the interstices of the membrane evaginations. D-E, I-J, L-M) ASFV-infected macrophages (MOI 100) were examined at 10 mpi by transmission EM after fixation. Panel D shows a side view of an infected macrophage. Note that ASFV particles seem to be endocytosed by macropinocytic-like membrane protrusions (red arrows) as well as by membrane invaginations characteristic of clathrin-dependent endocytosis (green arrows). Inset D1 shows a detail of panel D at higher magnification. E shows a virion apparently engulfed by macropinocytosis. F-G) Macrophages were incubated with both ASFV and dextran-coated 30-nm gold particles for 7 min at 37°C. F shows macropinocytic-like uptake of both ASFV and dextran-gold particles and G shows colocalization at a putative macropinosome. H) Size distribution histogram of endosomes containing either dextran-coated 30-nm gold particles (upper) or ASFV particles (lower) after 7 min of incubation. I-J, L-M) ASFV uptake by clathrin-coated pits (I, J and K) and coated vesicles (M and N). The clathrin coats are indicated by arrows. L) ASFV-infected macrophages were incubated for 15 min with 15 μM Pitstop 2. Arrows indicate virions at coated pits. Bars, 200 nm except for panels A and B (5 μm), D (1 μm) and L (500 nm).
Fig 3.
ASFV does not induce macropinocytosis.
A) Mock and ASFV-infected (MOI 100) macrophages were fixed and processed by transmission EM at 15, 30 and 60 mpi. Representative micrographs of Mock and ASFV-infected macrophages (left panels) and Vero cells (upper right panels) at 15 mpi are shown. ASFV particles are indicated by red arrowheads. Bars, 1 μm. Cells treated as before were quantified for the membrane protrusion density (number of protrusions > 200 nm per μm of plasma membrane from 50 cell profiles). Data display mean values and SD from a representative experiment. B) Macrophages (MAC) or Vero cells were infected (MOI 5; E70 or BA71V strains, respectively) for the indicated times and the phosphorylation levels of Pak1 (Thr423) were determined by immunoblotting. As a control of Pak activation, VACV-infected cells (MOI 5) were analyzed at 30 mpi. Levels of total Pak1 are shown below. C) Mock- and DiD-labeled ASFV-infected Vero cells (MOI 10) were incubated at 37°C for 15 min, fixed, stained for actin (green) and cell nucleus (Hoechst 33258, blue), and analyzed by wide-field fluorescence and DIC microscopy. As control, cells treated with 200 nM PMA, or infected with VACV (MOI 10) are shown. Bar, 5 μm. Incoming virus particles are shown in red. The yellow arrowheads indicate spike-like protrusions in PMA-treated cells while the white arroheads indicate blebbing in VACV-infected cells. Bar, 5 μm. In the bottom panel, the percentage of cells displaying blebbing in control cells (MOCK) or after infection with ASFV or VACV for 15 min is shown. Data indicate mean ± SD from two independent experiments. D) Mock- and ASFV-infected Vero cells (MOI 10) were pulsed for 30 and 60 min at 37°C with 10-kDa dextran-AF-488. As a control, cells treated with 200 nM PMA (30 and 60 min) or infected with VACV (60 min) were analyzed in parallel. Dextran uptake was quantified by flow cytometry and normalized to unstimulated cells (mean percentage ± SD of three independent experiments).
Fig 4.
Transport of incoming ASFV along the endocytic pathway.
A-B) ASFV-infected Vero cells (MOI 10) were analyzed at the indicated times by immunofluorescence detection of viral protein p17, early endosome/macropinosome marker EEA1 and late endosome/lysosome marker CD63. A) Colocalization of incoming viruses (red) with EEA1+ vesicles (green) at 10 mpi (left) or with CD63+ vesicles (green) at 30 mpi (right). B) Quantification of ASFV particles colocalizing with either EEA1, CD63 or both markers at the indicated times. C) Immunoelectron microscopy of incoming ASFV particles within endocytic vesicles labeled for early endosomal marker TFR, late endosomal/lysosomal markers CD63 and LAMP1, and lysosomal cathepsin L (CATH L). Immunogold labeling was performed on thawed cryosections of ASFV-infected cells using the indicated antibodies and protein A-gold (15 nm) conjugates. Bars, 200 nm. D) Transport of incoming ASFV in swine macrophages. Infected macrophages (MOI 100) were processed by EM at the indicated times. For each time, intracellular particles were classified and quantified according to their presence within early endosomes and macropinosomes (EE&MP), late endosomes and lysosomes (LE&LYS) or at the cytosol (CORES). E) ASFV-infected macrophages (E70, MOI 5) were treated for 2.5 h with inhibitors of endosome maturation (Baf A1, nocodazole and wortmannin) and analyzed for expression of viral protein p32 by immunoblotting (upper panel). Graphic shows percentage of p32 expression (mean ± SD from triplicate samples) referred to control, non-treated cells. F) COS-1 cells were transfected with two different siRNAs (1 and 2) targeted to Rab7, a scrambled siRNA or not treated (no siRNA). Then, cells were infected with ASFV (MOI 5) for 5 h. Expression of proteins p32, Rab7 and ß-actin was analyzed by immunoblotting. Numbers under the blot image indicated normalized expression levels for Rab7 and p32.
Fig 5.
ASFV disassembly in swine macrophages.
Macrophages were infected with ASFV (MOI 100) for 10, 15, 30, 45, 60 and 90 min and processed by EM. A-D) Selected EM images of the endosomal transport and the disassembly process undergone by ASFV. After endocytosis, incoming extracellular particles (A) are first detected (10–15 mpi) inside early endosomes (B) or macropinosomes keeping their structure nearly intact. Later (30–45 mpi), particles are predominantly found at multivesicular late endosomes (C), where a significant proportion loses their protein capsid (depicted in green) and the outer envelope (purple). Finally (45 mpi onwards), a fraction of particles reach lysosome-like structures (D) where most of them appear as a dense cores enwrapped by the inner envelope (red). Bars, 50 nm. E) Quantification of the ASFV disassembly. Incoming virus particles were classified according to their endocytic compartment (EE&MP: early endosomes and macropinosomes; LE: late multivesicular endosomes, LYS: lysosomes) and then, according to their layer content (ie: inner envelope, ca: capsid, oe: outer envelope). Data are expressed as percentages (mean ± deviation of duplicate experiments) of particles with a given domain inside each compartment. As a reference, extracellular virions (EV) attached to the cell surface were also quantified. More than 100 particles per compartment and experiment were analyzed. Note that the loss of the capsid and the outer envelope occurs essentially at multivesicular late endosomes and lysosome-like vesicles.
Fig 6.
ASFV fuses through the inner envelope at late multivesicular endosomes.
ASFV-infected swine macrophages (MOI 100) were analyzed by EM at 90 mpi. A-D) Representative EM images of the fusion process. After the loss of protein capsid and outer viral membrane, the inner viral envelope becomes exposed, interacting with the luminal face of the limiting membrane of multivesicular late endosomes (MVE) (A). Then, the inner envelope fuses with the endosomal membrane (B) and delivers naked cores into the cytosol (C). As a result of the disassembly and fusion events, extracellular particles lose the three domains surrounding the virus core before reaching the cytosol (D). Insets A1, B1 and C1 show lower magnification images of panel A, B and C, respectively. To facilitate the interpretation, the inner viral envelope is depicted in red, the virus core in brown and the endosomal membrane in purple. Bars, 100 nm.
Fig 7.
ASFV disassembly and uncoating depends on acidic pH.
Swine macrophages were infected with ASFV for 90 min in the presence or absence of 100 nM Baf A1, an inhibitor of endosomal acidification, and then processed by EM. A) Size distribution of virus-containing endosomes (n>100) in the presence or absence of Baf A1. Note that after drug treatment, the virus-containing endosomes are significantly smaller as a consequence of the impairment of endosome maturation. B) Under these conditions, most ASFV particles appear essentially intact (lower panels). By contrast, most ASFV particles in non-treated cells undergo significant disruption at late endosomes (upper panel). Outer envelope (purple), capsid (green) and inner envelope (red) are indicated. Bars, 100 nm. C) Quantification of Baf A1 effect on virus disruption. Internalized particles from treated (+BAF A1) and nontreated (-BAF A1) cells were quantified according to their layer content (i.e: inner envelope, ca: capsid, oe: outer envelope, co: cytosolic core). Data are expressed as percentages of particles (mean ± SD of triplicate samples) with a given domain for each condition. More than 100 particles per experiment were analyzed. Note that Baf A1 significantly prevents ASFV disassembly and core release. Bars, 100 nm.
Fig 8.
Low pH induces ASFV disruption.
(A) In vitro effect of pH acidification on ASFV structure. Purified virions were exposed to pH 6.5 or 5.0 for 60 min at 37°C and then ultracentrifuged and analyzed by EM after ultrathin sectioning. Note that, at pH 6.5, particles look virtually intact while the particles exposed to acidic pH (5.0) are severely disrupted, displaying evident signals of capsid disassembly as well as disruption and detachment of the outer membrane. Outer envelope (purple), capsid (green) and inner envelope (red) are indicated. (B) Purified virions adsorbed to EM grids were exposed to pH 6.5 or 5.0 for 15 min at 37°C and then immunogold labeled for major capsid protein p72. Note that, at pH 6.5, particles look virtually intact remaining non-accessible for p72 labeling. In contrast, particles exposed to acidic pH (5.0) are severely disrupted. Under these conditions, gold-labeled anti-p72 antibodies (black arrowheads) penetrate the particles through the disrupted outer envelope (red arrowheads), decorating discrete regions where the capsid is still present. Also, p72 labeling is detected on the support film around the particles, as a consequence of capsid disassembly. (C) Purified virions adsorbed to EM grids were exposed to pH 6.5 or 5.0 for 15 min at 37°C and then labeled for major inner envelope protein p17. Note that particles disrupted at pH 5.0 become accessible to anti-p17 labeling. Bars, 100 nm.
Fig 9.
ASFV fusion depends on virus protein pE248R.
A) Percoll-purified ASFV particles were obtained from cells infected with parental (wt) virus or recombinant virus vE248Ri under permissive (+IPTG) and non-permissive (-IPTG) conditions. Virus samples were analyzed by immunoblotting for the presence of capsid protein p72 and pE248R. (B) Vero cells were infected for 2 h with purified recombinant vE248Ri (~ 2 μg for 50,000 cells) grown with (+) or without (-) IPTG and then examined by EM. Disassembly of control vE248R+ (upper panel) and defective vE248R- (lower panels) was similar but core delivery into the cytosol was severely impaired for defective E248R- particles, which accumulated within lysosomes (LYS). To facilitate the interpretation, the inner viral envelope is depicted in red and the cytosolic viral core in brown. MVE (multivesicular endosomes). Bars, 200 nm. C) Quantification of disruption (lower panel) and core delivery (upper panel) of control E248R+ and defective E248R- particles. Endocytosed particles (END. VIRUS) and cytosolic cores (CYT. CORES) of the above experiment were quantified for both conditions and expressed as a percentage of intracellular virus. Also, the endocytosed particles of each condition were classified (lower panel) according to their layer content (i.e: inner envelope, ca: capsid, oe: outer envelope). One representative of two independent experiments is shown.
Fig 10.
Model for ASFV internalization and uncoating.
ASFV enters swine macrophages by clathrin-mediated endocytosis (left) and constitutive macropinocytosis (right). After the uptake, incoming particles are transported from early endosomes or macropinosomes to late endosomes, where they undergo the uncoating process, which would involve pH-dependent capsid disassembly and disruption of the outer viral membrane. Then, the exposed inner viral envelope fuses with the endosomal membrane to deliver genome-containing naked cores into the cytosol. A fraction of disrupted particles reach lysosomes, from where they might fuse or be further degraded. It cannot be excluded that lysosomal hydrolases (i.e. lipases and proteases) may also contribute to virus disruption or uncoating.