Ectromelia Virus Affects Mitochondrial Network Morphology, Distribution, and Physiology in Murine Fibroblasts and Macrophage Cell Line

Mitochondria are multifunctional organelles that participate in numerous processes in response to viral infection, but they are also a target for viruses. The aim of this study was to define subcellular events leading to alterations in mitochondrial morphology and function during infection with ectromelia virus (ECTV). We used two different cell lines and a combination of immunofluorescence techniques, confocal and electron microscopy, and flow cytometry to address subcellular changes following infection. Early in infection of L929 fibroblasts and RAW 264.7 macrophages, mitochondria gathered around viral factories. Later, the mitochondrial network became fragmented, forming punctate mitochondria that co-localized with the progeny virions. ECTV-co-localized mitochondria associated with the cytoskeleton components. Mitochondrial membrane potential, mitochondrial fission–fusion, mitochondrial mass, and generation of reactive oxygen species (ROS) were severely altered later in ECTV infection leading to damage of mitochondria. These results suggest an important role of mitochondria in supplying energy for virus replication and morphogenesis. Presumably, mitochondria participate in transport of viral particles inside and outside of the cell and/or they are a source of membranes for viral envelope formation. We speculate that the observed changes in the mitochondrial network organization and physiology in ECTV-infected cells provide suitable conditions for viral replication and morphogenesis.


Plaque Assay
Vero cells were seeded in 24-well plates and infected with 10-fold serial dilutions of ECTV suspension obtained from RAW 264.7 macrophages at 4, 12, and 24 hpi. At 5 days post infection, Vero cell monolayers were fixed in 4% paraformaldehyde (PFA, Sigma-Aldrich), stained with 0.3% crystal violet, and air dried. Plaques were counted using an Olympus IX71 inverted microscope (Olympus, Tokyo, Japan).

Immunofluorescent Staining and Microscopy Analysis
Permissive cells were seeded on glass coverslips in a 24-well plate and mitochondria were labeled with 300 nM Mito Red (560/580 [Ex/Em]) for 20 min in the dark at 37 • C with 5% CO 2 in a humidified incubator. The interaction of Mito Red with mitochondria depends on the mitochondrial membrane potential. After incubation, cells were washed three times with warm (37 • C) culture medium for 5 min. To detect ECTV antigens, F-actin, or Drp1, cells were fixed with 4% PFA in PBS for 20 min; for α-tubulin labeling, cells were fixed in ice-cold methanol (Sigma-Aldrich) for 2 min. Fixed cells were then permeabilized with 0.5% Triton X-100 (Sigma-Aldrich) in PBS (15 min) and blocked with 3% bovine serum albumin (BSA, Sigma-Aldrich) in 0.1% Triton X-100-PBS solution (30 min) to prevent nonspecific binding. The cells were subsequently incubated with primary Abs anti-Drp1 or α-tubulin for 1 h. Next, secondary anti-mouse Abs conjugated with FITC or rhodamine Red-X were used (60 min in the dark). Viral antigens were stained with FITC-conjugated rabbit anti-ECTV Abs, as previously described [24]. For F-actin detection, cells were incubated with 0.5 µg/mL phalloidin-FITC or -TRITC for 20 min in the dark. DNA was stained with 1 µg/mL Hoechst 33342 for 5 min in the dark. Slides were mounted in ProLong Gold Antifade Reagent (ThermoFisher Scientific).
Autophagy was detected with an LC3B Antibody Kit for Autophagy (ThermoFisher Scientific) containing anti-LC3B antibodies and chloroquine diphosphate. Before staining with Mito Red, the uninfected cells were pretreated with 30 µM chloroquine disphosphate for 16 h and used as a positive control. After fixation and permeabilization, the cells were incubated with primary anti-LC3B Abs for 1 h. Next, secondary anti-rabbit antibodies conjugated with FITC were used (1 h in the dark).

Transmission Electron Microscopy (TEM) Analysis
At 18 hpi, control and ECTV-infected L929 and RAW 264.7 cells grown in 24-well culture plates were collected and fixed for 2 h with 2.5% glutaraldehyde in phosphate buffer. After fixation, cells were rinsed in phosphate buffer and embedded in 2% agarose to form agarose blocks and then cut into smaller pieces. Next, cells were post-fixed for 1 h with 1% osmium tetroxide. Cells were dehydrated for 10 min at room temperature (RT) in each solution of ethanol with increasing concentrations (70% and 95%), twice in absolute ethanol for 10 min at RT and once in propylene oxide for 20 min at 4 • C. In the next step, cells were embedded in epoxy resin and cut into sections using a Leica EM UC6 Viruses 2018, 10, 266 5 of 30 ultramicrotome (Leica). The samples were examined with transmission electron microscope Jeol JEM-1220 (JEOL Ltd., Tokyo, Japan) equipped with an SIS Morada 11 MP camera.

Scanning Electron Microscopy (SEM) Analysis
At 18 hpi, control and ECTV-infected L929 and RAW 264.7 cells grown on glass coverslips in a 24-well plate were fixed for 1 h with 2.5% glutaraldehyde in phosphate buffer. Next, cells were rinsed in phosphate buffer and post-fixed for 1 h with 1% osmium tetroxide in phosphate buffer. Fixed cells were dehydrated for 10 min at RT in each solution of ethanol with increasing concentrations (70% and 95%), twice in absolute ethanol for 10 min at RT, and twice in acetone for 10 min at RT. Next, samples were dried using a CPD 7501 critical point drier (Polaron, Hatfield, PA, USA) and coated with a gold layer in a JFC-1300 sputter-coater (JEOL Ltd.). For analysis of specimens, the FEI Quanta 200 environmental scanning electron microscope (ESEM) with EDAX energy dispersive spectroscopy (EDS) system (FEI, Tokyo, Japan) was used.

Determination of Mitochondrial Membrane Potential
Mitochondrial membrane potential (MMP) was measured with a flow cytometer (LSRFortessa (BD Biosciences)) by using the dual-emission potential-sensitive probe JC-1 as a fluorescent dye. At high concentrations (due to high MMP), the dye aggregates, yielding red fluorescence with emission at~590 nm, while at low concentration (due to low MMP), JC-1 is predominantly a monomer that yields green fluorescence with emission at~529 nm. The ratio of red to green fluorescence of JC-1 is dependent only on MMP, and is not influenced by mitochondrial size, shape, or density. Therefore, mitochondrial depolarization is indicated by a decrease in the red/green fluorescence intensity ratio.
At indicated time points, control and ECTV-infected L929 and RAW 264.7 cells grown in 12-well culture plates were collected and incubated with culture medium supplemented with 10 or 6 µM JC-1 dye, respectively, for 20 min in the dark at 37 • C with 5% CO 2 in a humidified incubator. Cells were then centrifuged, suspended in cold (4 • C) PBS, and analyzed by flow cytometry (λ ex = 488 nm). Data were presented as the ratio of red to green fluorescence of JC-1 dye.
Uninfected L929 and RAW 264.7 cells treated both with JC-1 and 200 or 300 µM CCCP (carbonyl cyanide 3-chlorophenylhydrazone; Sigma-Aldrich), respectively, were used as negative control to confirm that the JC-1 response is sensitive to changes in MMP. CCCP is a protonophore (H + ionophore) and uncoupler of oxidative phosphorylation in mitochondria that collapse MMP.

Measurement of Reactive Oxygen Species Level
Reactive oxygen species (ROS) levels were measured with the redox-sensitive dye CM-H 2 DCFDA by flow cytometry. Non-fluorescent CM-H 2 DCFDA is converted to highly fluorescent CM-DCF (5-(6)-chloromethyl-2 ,7 -dichlorofluorescein) upon cleavage of the acetate groups by intracellular esterases and oxidation; [Ex/Em]~492-495/517-527 nm. Control and ECTV-infected L929 and RAW 264.7 cells grown in 12-well culture plates were collected at indicated time points and incubated in PBS supplemented with 1 µM or 2 µM CM-H 2 DCFDA dye, respectively, for 10 min in the dark at 37 • C with 5% CO 2 in a humidified incubator. Next, cells were centrifuged, suspended in cold (4 • C) PBS, and analyzed in a flow cytometer (λ ex = 488 nm). Uninfected L929 and RAW 264.7 cells treated with both CM-H 2 DCFDA and 1 mM H 2 O 2 were used as a positive control.

Measurement of Mitochondrial Mass
Mitochondrial mass was measured by flow cytometery (BD LSRFortessa, BD Biosciences) by using MitoTracker Green FM (490/516 nm [Em/Ex]). MitoTracker Green FM is a non-fluorescent dye in aqueous solutions, but becomes fluorescent once it accumulates in the lipid environment of mitochondria, regardless of membrane potential. At indicated time points, control and ECTV-infected L929 and RAW 264.7 cells prepared as described in Section 2.3, were collected and incubated with culture medium supplemented with 200 nM MitoTracker Green FM for 20 min in the dark at 37 • C Viruses 2018, 10, 266 6 of 30 with 5% CO 2 in a humidified incubator. Cells were then centrifuged, suspended in cold (4 • C) PBS, and analyzed by flow cytometry (λ ex = 488 nm).

Determination of ATP Levels
ATP levels were measured spectrophotometrically using an ATP Assay Kit (Abcam, Cambridge, MA, USA) which is based on the phosphorylation of glycerol in order to generate a product that can be easily quantified colorimetrically. At 18 hpi, control and ECTV-infected L929 cells (see Section 2.3) were collected and treated according to specifications in the ATP Assay Kit instruction manual. The absorbance of samples was measured using an Epoch Microplate Spectrophotometer (BioTek, Winooski, VT, USA) at OD 570 nm and the results were computed from the standard curve.

Western Blot Analysis
Control and ECTV-infected L929 and RAW 264.7 cells were lysed using RIPA buffer (ThermoFisher Scientific) supplemented with 1% of protease and phosphatase inhibitor cocktail (ThermoFisher Scientific). The protein concentrations were measured with a bicinchoninic acid (BCA) assay (ThermoFisher Scientific) according to the manufacturer's instructions. Next, 10-20 µg of total lysate was separated with 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and transferred onto a polyvinylidene fluoride (PVDF) membrane. After blocking with 5% non-fat dry milk in 0.1% Tween 20-PBS solution, membranes were incubated with primary Abs overnight at 4 • C as follows: anti-Opa1, anti-Drp1, anti-LC3B-1:1000 dilution, anti-GAPDH-1:1000 dilution, or anti-Fis1-1:3000 dilution. Secondary Abs conjugated with HRP (1:10,000) were used. The membranes were developed using tetramethylbenzidine (TMB) or SuperSignal West Pico PLUS (Thermo Fisher Scientific) as substrates. Next, the results were visualized in the Molecular Imager VersaDoc system (Bio-Rad, Hercules, MA, USA) or using X-ray film. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was used as a loading control and for protein normalization during densitometry measurements. ImageJ software was used for densitometry analysis.

Apoptosis Measurement by Flow Cytometry
The apoptotic rate of L929 and RAW 264.7 cells was determined using an Annexin V-FITC detection kit (Sigma-Aldrich) according to the manufacturer's protocol. At indicated time points, control and ECTV-infected L929 and RAW 264.7 cells grown in 12-well culture plates were collected, washed twice with cold PBS, and resuspended in 1× binding buffer. In the next step, 1 × 10 5 cells contained in 100 µL of binding buffer were labeled with 5 µL Annexin V-FITC and 5 µL propidium iodide (PI) and incubated for 15 min at RT in the dark. Then, cells were resuspended in 400 µL of 1× binding buffer and analyzed by flow cytometry (λ ex = 488 nm).
The kit includes Annexin V conjugated with FITC to label phosphatidylserine sites on the membrane surface, and PI to label the cellular DNA in necrotic cells where the cell membrane has been totally compromised. It can differentiate among early apoptotic cells (annexin V positive, PI negative), late apoptotic or necrotic cells (annexin V positive, PI positive), and viable cells (annexin V negative, PI negative).

Statistical Analysis
The results were presented as means ± SD of at least three independent experiments. The significance of statistical differences was evaluated using Student's t-test. The results were accepted as significantly different when p ≤ 0.05 *, p ≤ 0.01 **, or p ≤ 0.001 ***.

Kinetics of ECTV Replication in Murine Cell Lines
The kinetics of ECTV replication in RAW 264.7 macrophages and L929 fibroblasts were determined by standard plaque assay and fluorescence microscopy visualization of ECTV antigens using rabbit polyclonal antibodies anti-ECTV conjugated with FITC. Viral and cellular (nuclear) DNA were labeled with Hoechst 33342. The first ECTV replication centers were observed at 4 hpi in RAW 264.7 cells, and this was determined as the early stage of infection ( Figure 1A). Our recent studies have shown that viral factories in L929 fibroblasts also appeared at 4 hpi, but only 17% of cells displayed the presence of replication centers [25] in contrast to RAW 264.7 macrophages where 90% of ECTV-infected cells had viral replication centers ( Figure 1B). Early viral factories were observed as regular shapes (round or oval), but at 8 hpi, 60% of RAW 264.7 cells presented irregular "bloated" viral factories with progeny virions, whereas L929 exhibited a similar stage not earlier than 12 hpi (Figure 1) [25]. Plaque assay confirmed that during ECTV infection in RAW 264.7 macrophages the number of infectious particles increased between 4 and 24 hpi ( Figure 1C). have shown that viral factories in L929 fibroblasts also appeared at 4 hpi, but only 17% of cells displayed the presence of replication centers [25] in contrast to RAW 264.7 macrophages where 90% of ECTV-infected cells had viral replication centers ( Figure 1B). Early viral factories were observed as regular shapes (round or oval), but at 8 hpi, 60% of RAW 264.7 cells presented irregular "bloated" viral factories with progeny virions, whereas L929 exhibited a similar stage not earlier than 12 hpi ( Figure 1) [25]. Plaque assay confirmed that during ECTV infection in RAW 264.7 macrophages the number of infectious particles increased between 4 and 24 hpi ( Figure 1C).

Cell Morphology during ECTV Infection
Visualization of morphological changes in ECTV-infected cells was performed using fluorescence staining of F-actin with phalloidin and subsequent fluorescent microscopy and SEM analysis. Infected RAW 264.7 macrophages (Figure 2A,B) strongly adhered to the slide surface

Mitochondrial Network Morphology and Distribution in Murine Cell Lines during ECTV Infection
As described in our previous study [24], the branching structure of the mitochondrial network in uninfected L929 fibroblasts collapsed and became disorganized followed by destruction of

Mitochondrial Network Morphology and Distribution in Murine Cell Lines during ECTV Infection
As described in our previous study [24], the branching structure of the mitochondrial network in uninfected L929 fibroblasts collapsed and became disorganized followed by destruction of mitochondrial tubules under the influence of ECTV at the late stages of infection (18 and 24 hpi). Mitochondria of uninfected RAW 264.7 cells were not branched, but had a regular distribution within the cytoplasm (Figure 3). The beginning of viral factory formation observed at 4 hpi was the earliest unequivocal evidence of the mitochondrial distribution changes within the cytoplasm of infected RAW 264.7 cells. At that time, small and round ECTV replication centers were observed in the perinuclear region with characteristic mitochondrial clustering in close proximity to these areas ( Figure 3). A similar observation was made in L929 fibroblasts at 8 hpi with ECTV, when the viral factories occurred in most of the infected cells [24,25]. The manifestation of expanded ("bloated") viral factories and the appearance of progeny virions within the cytoplasm of ECTV-infected RAW 264.7 and L929 cells at 8 and 12 hpi, respectively, began the progressive changes of mitochondrial network morphology during infection. Moreover, α-tubulin staining by immunofluorescence revealed the rearrangement of microtubules and disappearance of the microtubule organizing center (MTOC) during the late stages of ECTV infection in RAW 264.7 macrophages ( Figure 4B vs. Figure 4A), as we previously described in L929 cells [24,25]. Further, the MTOC region in uninfected control cells was characterized by the absence of mitochondria in L929 [24] and RAW 264.7 cells ( Figure 4A); however, the disappearance of this structure at later stages of ECTV infection led to the accumulation of mitochondria in this area [24].
Subsequent immunofluorescence staining using a marker of MTOC-γ-tubulin (which forms the centrosome and is responsible for microtubule nucleation)-confirmed the disappearance of the centrosome in~30% of ECTV-infected RAW 264.7 cells ( Figure 4B,C). The rest of the infected cells presented a weaker fluorescent signal from the centrosome or change of the centrosome location within the cytoplasm, in comparison to the control cells ( Figure 4B). The centrosome containing a pair of centrioles was observed in the perinuclear region in uninfected control RAW 264.7 cells. Disappearance of the centrosome was previously described in L929 cells at the late stage of ECTV infection [25].
Already from 12 hpi, the mitochondria of L929 cells were fragmented and disorganized with the loss of connection between themselves ( Figure 5) [24]. The majority of mitochondria were tightly clustered in close association with viral factories (between virus replication centers and the nucleus or inside these "bloated" structures). However, some mitochondria were irregularly dispersed within the cytoplasm and displayed various forms, such as (a) a loose mitochondrial network ("loose net"), when the branched network was relaxed; (b) short, non-connected mitochondrial tubules; (c) donut-like mitochondria; and (d) single, small, circular mitochondria (punctate mitochondria) ( Figure 5B), as was previously described at 18 hpi [24]. In contrast, mitochondrial disorganization and partial fragmentation in RAW 264.7 cells were observed starting from 8 hpi, but elongated mitochondrial tubules occurred as well ( Figure 3). Interestingly, such mitochondrial elongation was not observed in uninfected control cells. Western blot analysis showed that the level of mitochondrial fusion protein Opa1 at 4 hpi with ECTV was slightly, but significantly higher by 26.4 ± 7.0% (p = 0.0227) in comparison to the control ( Figure 6). The level of Opa1 at 18 hpi in RAW 264.7 cells was reduced by 15.1 ± 4.9% compared to the control. A similar decrease in Opa1 level by 23.4 ± 2.5% (p = 0.0039) was observed at 18 hpi in ECTV-infected L929 cells compared to in uninfected cells. Furthermore, the level of mitochondrial fission Fis1 protein increased 1.5-and 2.5-fold in L929 and RAW 264.7 cells, respectively, compared to the control ( Figure 6).
Protein expression of Drp1 remained unchanged in both cell lines during ECTV infection compared to control, but its distribution within the cytoplasm was altered. Drp1 was distributed regularly within the cytoplasm in control cells, whereas in ECTV-infected L929 and RAW 264.7 cells, Drp1 localized in close association with viral factories, which corresponded to the mitochondrial accumulation around such compartments during infection ( Figure 7). Viruses 2018, 10, x 10 of 30           Figure 9D,E), as was also described in our previous study [24]. Such discrete mitochondrial interaction with virions was observed on the top of ECTV-induced actin tails and filopodia, as well as on microtubules ( Figure 10).
To further scrutinize the mitochondria-ECTV interaction we used transmission electron microscopy (TEM). Figure 11A shows a mitochondrion (~470 × 300 nm) in L929 cells structurally connected to ECTV virion (~250 × 125 nm) via some form of membrane bridge. The structure of such mitochondria was altered with a clearly swollen matrix ( Figure 11A). Viral particles connected to mitochondria were characterized by the presence of only one membrane, typical for mature virions (MVs) of poxviruses ( Figure 11B).  Figure 9D,E), as was also described in our previous study [24]. Such discrete mitochondrial interaction with virions was observed on the top of ECTV-induced actin tails and filopodia, as well as on microtubules ( Figure 10).
To further scrutinize the mitochondria-ECTV interaction we used transmission electron microscopy (TEM). Figure 11A shows a mitochondrion (~470 × 300 nm) in L929 cells structurally connected to ECTV virion (~250 × 125 nm) via some form of membrane bridge. The structure of such mitochondria was altered with a clearly swollen matrix ( Figure 11A). Viral particles connected to mitochondria were characterized by the presence of only one membrane, typical for mature virions (MVs) of poxviruses ( Figure 11B).

Mitochondrial Physiology in Murine Cells during ECTV Infection
Because the mitochondrial networks in RAW 264.7 and L929 cells were disorganized and partially fragmented at later stages of infection, we hypothesized that infection with ECTV can negatively impact the mitochondrial physiology. Taking into account the divergences in the replication kinetics of ECTV between fibroblasts and macrophages, appropriate time points after infection of each cell line were selected for mitochondrial physiology analysis.
Due to the critical role of MMP in mitochondrial homeostasis, we evaluated MMP in uninfected and infected L929 and RAW 264.7 cells using JC-1 staining and subsequent flow cytometry analysis ( Figure 12A). At 4 hpi with ECTV, MMP was decreased by almost one third (p ≤ 0.05) in L929 fibroblasts, compared to the uninfected control; however, no statistically significant difference in

Mitochondrial Physiology in Murine Cells during ECTV Infection
Because the mitochondrial networks in RAW 264.7 and L929 cells were disorganized and partially fragmented at later stages of infection, we hypothesized that infection with ECTV can negatively impact the mitochondrial physiology. Taking into account the divergences in the replication kinetics of ECTV between fibroblasts and macrophages, appropriate time points after infection of each cell line were selected for mitochondrial physiology analysis.
Due to the critical role of MMP in mitochondrial homeostasis, we evaluated MMP in uninfected and infected L929 and RAW 264.7 cells using JC-1 staining and subsequent flow cytometry analysis ( Figure 12A). At 4 hpi with ECTV, MMP was decreased by almost one third (p ≤ 0.05) in L929 fibroblasts, compared to the uninfected control; however, no statistically significant difference in MMP was reported in infected RAW 264.7 macrophages at 4 hpi ( Figure 12B). The first statistically significant reduction of MMP in RAW 264.7 cells occurred at 8 hpi (by 18.3 ± 3.7%; p = 0.013), whereas in L929 cells, MMP declined by 45% (p ≤ 0.001) compared to control. At 18 hpi, MMP in RAW 264.7 cells was lower by 40% (p ≤ 0.01) compared to control, while in L929 cells MMP was reduced by more than 50% (p ≤ 0.05). At 24 hpi, MMP in L929 fibroblasts was decreased by 60%. Generally, a considerable percentage of cells with depolarized MMP was observed in both cell lines and the results showed increasing numbers of cells with low MMP as the infection progressed ( Figure 12B). MMP was reported in infected RAW 264.7 macrophages at 4 hpi ( Figure 12B). The first statistically significant reduction of MMP in RAW 264.7 cells occurred at 8 hpi (by 18.3 ± 3.7%; p = 0.013), whereas in L929 cells, MMP declined by 45% (p ≤ 0.001) compared to control. At 18 hpi, MMP in RAW 264.7 cells was lower by 40% (p ≤ 0.01) compared to control, while in L929 cells MMP was reduced by more than 50% (p ≤ 0.05). At 24 hpi, MMP in L929 fibroblasts was decreased by 60%. Generally, a considerable percentage of cells with depolarized MMP was observed in both cell lines and the results showed increasing numbers of cells with low MMP as the infection progressed ( Figure 12B). In order to verify whether the collapse of MMP in ECTV-infected cells is associated with cell death, we measured the apoptotic rate of L929 and RAW 264.7 cells. Our result showed that ECTV infection does not induce cell death in L929 fibroblasts; however, the apoptotic/necrotic rate in RAW 264.7 macrophages reached ~40% at the late stages of infection ( Figure 13). In order to verify whether the collapse of MMP in ECTV-infected cells is associated with cell death, we measured the apoptotic rate of L929 and RAW 264.7 cells. Our result showed that ECTV infection does not induce cell death in L929 fibroblasts; however, the apoptotic/necrotic rate in RAW 264.7 macrophages reached~40% at the late stages of infection ( Figure 13). Reduction in MMP is accompanied by ROS generation; therefore, in the next step we performed measurement of the ROS level in L929 and RAW 264.7 cells using CM-H2DCFDA labeling and flow cytometric analysis ( Figure 14A). At 4 hpi, there were no statistically significant changes in ROS production between infected and control cells in both cell lines ( Figure 14B). However, fibroblasts and macrophages presented an increased level of ROS at the later stages of infection, compared to the uninfected control (p ≤ 0.05). The production of ROS was higher by approximately 35% in RAW 264.7 at 18 hpi, and by ~50% and ~70% in L929 cells at 18 and 24 hpi, respectively ( Figure 14B).
Reduction in MMP results in decreased energy provision, so we next examined how ECTV infection influences ATP production in permissive cells. Interestingly, in ECTV-infected L929 fibroblasts at 18 hpi, the level of ATP remained unchanged compared to that in uninfected control cells ( Figure 14C).
Since ECTV-infected cells exhibited an imbalance in fusion-fission processes, we finally addressed mitochondrial mass in such cells. The mitochondrial mass measurement was performed using flow cytometric analysis of live L929 and RAW 264.7 cells labeled with MitoTracker Green FM. At 4 hpi with ECTV, there were no statistically significant changes in mitochondrial mass in both cell lines compared to the control (p ≤ 0.05). From 8 hpi, L929 and RAW 264.7 cells were characterized by decreased mitochondrial mass until the later stages of infection ( Figure 15). Reductions by ~30% at 8 hpi and ~20% at 18 hpi were noticed in RAW 264.7 cells, whereas decreases by ~20% at 8 hpi, ~30% at 18 hpi, and ~50% at 24 hpi were observed in L929 cells. Due to the loss of mitochondrial mass in Reduction in MMP is accompanied by ROS generation; therefore, in the next step we performed measurement of the ROS level in L929 and RAW 264.7 cells using CM-H 2 DCFDA labeling and flow cytometric analysis ( Figure 14A). At 4 hpi, there were no statistically significant changes in ROS production between infected and control cells in both cell lines ( Figure 14B). However, fibroblasts and macrophages presented an increased level of ROS at the later stages of infection, compared to the uninfected control (p ≤ 0.05). The production of ROS was higher by approximately 35% in RAW 264.7 at 18 hpi, and by~50% and~70% in L929 cells at 18 and 24 hpi, respectively ( Figure 14B).
Reduction in MMP results in decreased energy provision, so we next examined how ECTV infection influences ATP production in permissive cells. Interestingly, in ECTV-infected L929 fibroblasts at 18 hpi, the level of ATP remained unchanged compared to that in uninfected control cells ( Figure 14C).
Since ECTV-infected cells exhibited an imbalance in fusion-fission processes, we finally addressed mitochondrial mass in such cells. The mitochondrial mass measurement was performed using flow cytometric analysis of live L929 and RAW 264.7 cells labeled with MitoTracker Green FM. At 4 hpi with ECTV, there were no statistically significant changes in mitochondrial mass in both cell lines compared to the control (p ≤ 0.05). From 8 hpi, L929 and RAW 264.7 cells were characterized by decreased mitochondrial mass until the later stages of infection ( Figure 15). Reductions by~30% at 8 hpi and~20% at 18 hpi were noticed in RAW 264.7 cells, whereas decreases by~20% at 8 hpi,~30% at 18 hpi, and~50% at 24 hpi were observed in L929 cells. Due to the loss of mitochondrial mass in ECTV-infected cells, we additionally verified the presence of autophagy during infection. Fluorescence microscopy analysis showed intensified aggregates of LC3B in infected cells ( Figure 16A). Additionally, Western blot analysis indicated a significantly (p ≤ 0.01) higher level of LC3B II (active form of LC3B) in virus-infected L929 fibroblasts at 18 hpi, compared to the control ( Figure 16B,C). Both results strongly suggest the induction of autophagy.

ECTV-infected cells, we additionally verified the presence of autophagy during infection.
Fluorescence microscopy analysis showed intensified aggregates of LC3B in infected cells ( Figure  16A). Additionally, Western blot analysis indicated a significantly (p ≤ 0.01) higher level of LC3B II (active form of LC3B) in virus-infected L929 fibroblasts at 18 hpi, compared to the control ( Figure  16B,C). Both results strongly suggest the induction of autophagy.

Discussion
Structural and functional changes in the cell caused by viral infection are referred to as cytopathic effect (CPE). The nature of CPE in a cell culture infected with poxvirus depends on the type of virus, type of cells or tissue, and the MOI [28,29]. Therefore, in order to link CPE, including alterations in mitochondrial network morphology and physiology, to the particular stages of infection, we first evaluated the kinetics of ECTV replication in RAW 264.7 and L929 cells. Despite using the same MOI of 5, ECTV replicated faster in RAW 264.7 macrophages than in L929 fibroblasts, as evidenced by replication center formation. Our data confirmed that the duration of the poxvirus replication cycle is cell type dependent [30]. The differences between RAW 264.7 and L929 cells in the kinetics of viral replication may arise from the higher tropism of ECTV to macrophages or greater efficiency of virus entry into this cell type. Studies have shown that mature virions (MV) of VACV use a distinct form of macropinocytosis for host cell entry [31]. Moreover, MV correspond in size to apoptotic bodies, and, as such, may also be engulfed through macropinocytosis by specialized cells such as macrophages [31][32][33]. Others have hypothesized that VACV virions enter host cells by mimicking apoptotic bodies, leading not only to effective macropinocytosis, but also to the avoidance of immune recognition [31]. This is due to the fact that macropinocytosis of apoptotic debris suppresses the activation of innate immune responses [34]. It can be assumed that RAW 264.7, as phagocytic cells, may exhibit a greater ability to macropinocytose viral particles, unlike non-specialized L929 fibroblasts.
Alterations in cell shape and size are strictly related to rearrangement of the cytoskeleton, which is responsible for intracellular distribution of organelles, including mitochondria. Mitochondrial organization within the cytoplasm contributes to their function and is essential for cell survival [35].
Our results indicate that ECTV infection causes cytoskeleton rearrangement in RAW 264.7 macrophages, which is expressed by changes in cell shape and occurrence of cytoplasmic projections. Similar results were shown in our previous study on ECTV-infected L929 [25] and BALB 3T3, HeLa, and Vero cells [23]. Other studies also have shown similar observations during VACV, VARV, monkeypox virus (MPXV) [36], and myxoma virus (MYXV) [27] infection. The observed presence of ECTV particles on or in cytoplasmic extensions of RAW 264.7 in the current study suggests the role of these structures in cell-to-cell viral spreading that has been confirmed by numerous studies on different types of viruses belonging to distinct families, such as Poxviridae [23,25,27,37,38], Reoviridae [39], and Togaviridae [40]. We can speculate that the formation of "cytoplasmic corridors" by ECTV-infected cells allows for direct transportation of viral particles between infected and uninfected cells, avoiding contact of cell membranes. Thus, virus entry into host cells could be simplified, permitting earlier initiation of virus replication. Furthermore, the lack of contact of viral particles with the extracellular environment may provide protection against unfavorable conditions outside the cell, including the antiviral immune mechanisms in vivo.
Proper mitochondrial network morphology and distribution within the cytoplasm is essential for cell survival and contributes to organelle function. Immunofluorescence and Western blot analysis revealed that mitochondrial fusion-fission processes were altered in ECTV-infected RAW 264.7 and L929 cells, compared to in uninfected control cells. The mitochondrial network of these cells was partly fragmented and disorganized. Our present and previous studies indicate that formation of viral factories relates to the beginning of changes in localization of mitochondria within the cytoplasm in RAW 264.7 and L929 cells [24]. Mitochondria of infected cells were clustered in close proximity to the early ECTV replication centers. Our data show that the MTOC-a structure from which microtubules emerge-partly or completely disappears in ECTV-infected RAW 264.7 macrophages, as was previously described in L929 fibroblasts [24,25] and during VACV infection of Hela cells [41]. Viral factories form in the vicinity of the MTOC, suggesting the role of microtubules in the transport of poxvirus particles [24,25,41]. It is likely that localization of mitochondria near viroplasms (viral factories) in ECTV-infected cells may compensate for the loss of the MTOC.
Rojo et al. [42] have shown that mitochondria in Vero cells also aggregated near viral factories during infection with African swine fever virus (ASFV)-another large DNA virus predominantly replicating within the cytoplasm. Rojo et al. [42] suggested the role of mitochondria in supplying energy for the ASFV morphogenetic processes [42]. Other studies confirmed this fact by showing that ATP is required for the wrapping of ASFV progeny particles [43,44]. Taken together, the accumulation of mitochondria in close proximity to the ECTV replication centers suggests that mitochondria are active and probably supply ATP for replication of viral DNA and morphogenesis of progeny virions. According to Chang et al. [45] VACV-infected cells produced two times more ATP than control cells, and a high level of ATP is necessary to drive efficient virus replication. Moreover, an elevated oxygen consumption rate (ATP synthesis) during VACV infection was also confirmed in later studies [46]. Surprisingly, the level of ATP in ECTV-infected L929 cells did not decrease, suggesting that despite alteration in mitochondrial physiology, these organelles were still able to produce ATP, therefore delaying induction of apoptosis.
Recent studies have indicated that infection with DNA and RNA viruses such as hepatitis B virus (HBV) [47], human cytomegalovirus (HCMV) [48], Epstein-Barr virus (EBV) [49], hepatitis C virus (HCV) [50], influenza A virus (IAV) [51], pseudorabies virus (PRV) [52], and Dengue virus (DENV) [53] is able to affect the structure of the mitochondrial network, leading to its fragmentation. Kim et al. [47] found that HBV and its encoded HBx protein promoted perinuclear clustering of mitochondria and triggered mitochondrial translocation of Drp1, resulting in mitochondrial fission. Although the present study shows that the level of Drp1 protein was unchanged during ECTV infection, the localization of this protein was altered. Drp1 accumulated near mitochondria and viral factories, suggesting that a similar mechanism may exist in ECTV-infected cells. To the best of our knowledge, there is no report describing a poxviral protein responsible for mitochondrial dynamics alteration. However, poxviruses encode mitochondria-localized proteins, including EVM025 from ECTV [19], F1L from VACV [54], and M11L from MYXV [55] that localize to the mitochondria through the C-terminal transmembrane domain, where they can carry out their anti-apoptotic functions. Furthermore, mutation of some ECTV proteins appears to have a different role in in vivo and in vitro conditions. For example, A36 YdF ECTV, which has mutations at the two critical tyrosine residues Y112 and Y132 in the A36 protein, exhibits disrupted actin-based motility and shows impaired release in vitro, but relatively normal dissemination in vivo [56].
Apart from partial fragmentation of the mitochondrial network, the ECTV-infected cells displayed elongated mitochondria as well. A study by Castanier et al. [57] has shown that excessive mitochondrial fusion (elongated, branched mitochondria) increases the interaction of MAVS with other molecules involved in the RLR-dependent signaling pathway, which leads to production of anti-viral factors-type I IFNs and pro-inflammatory cytokines upon poly(I:C) stimulation or Sendai virus infection. Conversely, fragmented mitochondria cause a decrease in ability to transmit the signals for type I IFN production. Inhibition of mitochondrial fragmentation by silencing Drp1 during HCV infection enhanced type I IFN production, suggesting that HCV-induced mitochondrial fission contributes in part to the evasion of the innate immune system [50]. Based on our results, we speculate that excessive mitochondrial fission observed during the late stages of ECTV infection would serve to perturb the cellular antiviral defense. Because there is no data describing an ECTV-encoded protein that stimulates mitochondrial fragmentation, it is likely that mitochondrial fission is related to the progressive damage of mitochondria during infection, as a "side effect" of the virus activity. Nevertheless, even accidental activation of mitochondrial fission would presumably lead to a blockade of MAVS downstream signaling, resulting in reduced IFN synthesis and sustained viral infection. Recently, [58] it has been shown that ECTV infection of GM-CSF-derived bone marrow cells (composed of conventional dendritic cells and macrophages) induces down-regulation of genes involved in the RIG-I-MAVS signaling pathway.
The use of the cytoskeleton for movement of viral particles was previously described during infections with orthopoxviruses, such as ECTV [23], VACV [59], VARV, and MPXV [36], as well as other distinct poxviruses, e.g., Yaba-like disease virus (YLDV) [60], swinepox virus (SWPV) [61], and MYXV [27]. Intracellular transport of virions along microtubules and actin polymerization to form virus-induced actin comet tails are necessary for viral spreading and require the energy from ATP hydrolysis. The major cellular source of ATP is mitochondria; this would suggest that punctate mitochondria supply energy for the transportation of progeny virions inside and outside the cell through the cytoskeleton. Mitochondrial fragmentation facilitates transportation of mitochondria along microtubules, whereas motility of strongly elongated organelles is prevented [62,63]. We can speculate that the selective interaction of progeny ECTV virions with only punctate mitochondria in the infected cells provides efficient transport of viral particles along the cytoskeletal elements. Considerably lower numbers (statistically significant; p ≤ 0.05) of this type of mitochondria interacted with ECTV in L929 (25.1 ± 10.0), RAW 264.7 (20.8 ± 10.1) and MEFs WT (11.8 ± 3.5) compared to in MEFs Mfn1−/−/Mfn2−/− (99.5 ± 25.2) which are defective in the mitochondrial fusion process. This suggests that mitochondrial elongation prevents excessive mitochondrial fission and may constitute a protective mechanism against viral transport and spreading. However, this assumption requires further investigation. The presence of punctate mitochondria co-localized with ECTV virions on the top of actin tails suggests that the intact mitochondria are released outside the cell, a rather controversial issue. However, Maeda and Fadeel [64] have recently shown that mitochondria are released from necroptotic cells before the disruption of the plasma membrane, indicating that this is an active process. Mitochondria released in such a manner maintain their function with no mtDNA removal and can affect the cytokine production by macrophages and cell maturation of antigen-presenting dendritic cells [64].
TEM analysis showed that mitochondria in L929 cells formed a structural connection with ECTV virions via a membrane bridge. It is possible that the membrane bridge structure may have an origin in the mitochondrial outer membrane, inner membrane, or both. In any case, the pivotal issue here concerns definition of the role of this structure. Interestingly, the mitochondrion-connected viral particle had only one membrane, a characteristic typically associated with mature virions (MV) of poxviruses. Some MVs are wrapped in additional, cell-derived membranes to become triple-membrane particles called wrapped virions (WVs) [65]. As it currently stands, the modified trans-Golgi or endosomal membrane is the origin of the viral envelope [65,66]. Due to the fact that mitochondria contain two membranes-inner and outer-it is not excluded that ECTV virions could utilize mitochondrial membrane/s for viral envelope formation. Hawes et al. [67] have shown that the thickness of the African swine fever virus (ASFV) envelope is very similar to that of a single lipid bilayer of mitochondria and other cellular organelles. According to our study, there is no reason to reject the hypothesis about mitochondrial origin of the ECTV envelope.
The evaluation of mitochondrial physiology indicates that ECTV infection contributes to an ongoing reduction of MMP in L929 and RAW 264.7 cells. Our results of mitochondrial fission in ECTV-infected cells confirm that the loss of mitochondrial membrane potential is related with mitochondrial dysfunction and fragmentation [68,69]. Further, we show that alterations in MMP appeared earlier and were higher in fibroblasts than in macrophages. This shows that RAW 264.7 cells are more resistant to the loss of MMP triggered by ECTV infection. Previous studies of mechanisms by which poxviruses regulate apoptosis have indicated that ECTV and VACV infection do not affect mitochondrial membrane potential [19][20][21]. This stands in contrast to our data and can be related to different fluorescent probes applied to measure MMP in our study. Additionally, the dissimilarity between our results and others in the level of MMP may arise from differences in the time post-infection when the measurements were taken. No changes in MMP were noted at 6-8 hpi with ECTV or VACV; however, further observations were not conducted [19][20][21]. The only exceptions are results reported from experiments performed by Melo-Silva et al. [70] in which MMP declined at 24 hpi with ECTV compared to control, and no differences were observed at 8 hpi. Therefore, it can be assumed that similar alterations could have been observed during the other cited studies.
The collapse of MMP is linked to mitochondrial damage and is often a manifestation of apoptosis activation. In this context, occurrence of apoptosis in ECTV-infected L929 and RAW 264.7 cells was determined. Our results indicated no apoptotic changes in L929 fibroblasts; however, RAW 264.7 macrophages exhibited an apoptotic rate of~40%. The discrepancy between results may arise from the different properties of cell lines that influenced the duration of the virus replication cycle. Higher tropism of ECTV to macrophages or greater efficiency of virus entry into this cell type could be a reason for both faster virus replication and apoptosis activation. On the other hand, lack of cell death in L929 cell during ECTV infection suggest no virus-associated cytotoxicity effect on this cell line at MOI = 5, even in the late stages of infection.
Increased levels of ROS were evident in both L929 and RAW 264.7 cells during the later stages of infection with ECTV. Mitochondria are the main source of ROS both in normal and pathological conditions but are simultaneously one of the most endangered ROS targets [71]. Zorov et al. have shown that higher levels of ROS result in the loss of MMP, which consequently leads to mitochondrial permeability transition and damage [72]. In turn, defective mitochondria generate even more ROS which contributes to even greater damage to mitochondria. Similar results were obtained during HCV infection [73]. Intended inhibition of ROS production in HCV-infected cells resulted in both a reduction of ROS level and prevention of mitochondrial degradation. Based on these reports, it can be suggested that the increased level of ROS generated during ECTV infection triggers mitochondrial damage, manifested by the loss of MMP and mitochondrial fragmentation.
The molecular mechanism of ECTV-induced attenuation of MMP and progression of mitochondrial fragmentation is not known and should be elucidated in future studies to extend our knowledge about the implications of mitochondrial dynamics in orthopoxviral pathogenesis. We speculate that ECTV may encode protein(s) that accumulate in mitochondria and therefore affect mitochondrial functions. It has been shown that the PB1-F2 protein of the influenza A virus constitutively localizes to the mitochondrial inner membrane space via Tom40 channels, and its accumulation leads to remarkable mitochondrial fragmentation due to reduced MMP [74]. MMP reduction was correlated with impaired mitochondrial-mediated innate immune responses, including the RIG-I signaling pathway and NLRP3 inflammasomes [74].
Our results show that later stages of ECTV infection are critical to the reduction of mitochondrial mass in L929 and RAW 264.7 cells. It appears that there is successive mitochondrial damage, presumably as a result of disruption of mitochondrial function during infection. Defective mitochondria can be targeted for degradation via selective (mitophagy) or nonselective autophagy, due to the activation of mitochondrial permeability transition and/or loss of mitochondrial membrane potential [75][76][77]. Our current and previous studies have shown the presence of autophagy in ECTV-infected L929 cells [77], which suggests that some type of autophagy may be responsible for mitochondrial removal in infected murine cell lines, manifested by the decrease of mitochondrial mass.
Mitophagy is tightly related to mitochondrial dynamics, and together they are identified as the key pathways in mitochondrial quality control [78]. A number of studies have shown that mitochondrial fission is responsible for segregation and elimination of damaged organelles out of the cell, pointing to the protective role of this process [79,80]. Fragmented mitochondria with irreparable defects constitute a risk to cell survival, due to the harmful effects of ROS production [80]. Oxidative stress generates damage to DNA, proteins, and lipids, consequently leading to cell death. During poxviral infection, including ECTV, prolonged maintenance of the vital functions by the host cell is critical for efficient viral replication. Poxviruses encode anti-apoptotic proteins [19,21]; therefore, induction of cell death is not supportive of viral survival. Kim et al. [47,50] have demonstrated that HBV and HCV infection lead to autophagy (mitophagy) activation that is presumably a counteraction against virus-induced apoptosis. HBV or HCV infection stimulates synthesis of Drp1 and triggers its translocation to mitochondria, leading to mitochondrial fission and then mitophagy. Intended inhibition of mitochondrial fragmentation by silencing of Drp1 resulted in reduction of mitophagy and enhancement of apoptosis [47,50]. In the light of findings such as fragmentation of mitochondrial network, mitochondrial mass reduction, and presence of autophagy, it is likely that ECTV infection contributes to the active process of mitochondria removal from L929 and RAW 264.7 cells leading to inhibition of apoptosis. We cannot rule out the possibility that mitochondrial elimination impairs MAVS-dependent anti-viral immunity and promotes virus infection as well [81]. Higher ROS levels increase MAVS downstream signaling resulting in elevated type I IFN synthesis and limited development of virus infection [81]. Based on our results, removal of mitochondria from ECTV-infected cells may not only protect from host cell death, but may also allow the virus to avoid mitochondrial-dependent immunity. It should be noted that the increased level of ROS during the late stages of ECTV infection would theoretically indicate an opposite effect; however, the inability to eliminate damaged mitochondria could possibly lead to even greater amount of ROS than was observed.

Conclusions
Our study shows that ECTV infection alters the mitochondrial network morphology and distribution, especially through fragmentation of mitochondria and their accumulation in close proximity to the viral factories. Mitochondrial fission at the later stages of infection led to the formation of small, punctate mitochondria that co-localized with progeny virions and were found on cytoskeleton components, like actin tails, filopodia, and microtubules. Mitochondrial function in infected cells was altered, and involved the decrease of mitochondrial membrane potential, increase of ROS level, reduction of mitochondrial mass, and presence of autophagy, suggesting degradation of mitochondria and their elimination from the cell at the later stages of infection. ECTV replicated at a higher rate in RAW 264.7 cells and consequently induced more rapid changes in mitochondrial network morphology and distribution than in L929 cells. Disruption of mitochondrial function occurred at a later time point in macrophages compared to in fibroblasts. The observed alterations in the mitochondrial network morphology and organization as well as mitochondrial function during ECTV infection provide suitable conditions for viral replication and morphogenesis, but the rate of such changes is cell-dependent.