Role of mitochondria in parvovirus pathology

Abstract

Proper functioning of the mitochondria is crucial for the survival of the cell. Viruses are able to interfere with mitochondrial functions as they infect the host cell. Parvoviruses are known to induce apoptosis in infected cells, but the role of the mitochondria in parvovirus induced cytopathy is only partially known. Here we demonstrate with confocal and electron microscopy that canine parvovirus (CPV) associated with the mitochondrial outer membrane from the onset of infection. During viral entry a transient depolarization of the mitochondrial transmembrane potential and increase in ROS level was detected. Subsequently, mitochondrial homeostasis was normalized shortly, as detected by repolarization of the mitochondrial membrane and decrease of ROS. Indeed, activation of cell survival signalling through ERK1/2 cascade was observed early in CPV infected cells. At 12 hours post infection, concurrent with the expression of viral non-structural protein 1, damage to the mitochondrial structure and depolarization of its membrane were apparent. Results of this study provide additional insight of parvovirus pathology and also more general information of virus-mitochondria association.

Figure 1. CPV colocalizes with mitochondria.

(A) Confocal immunofluorescence images of CPV infected cells. Mitochondria were labelled with anti-COX IV antibody (red) and CPV with anti-capsid antibody (green). Nucleus has been cut out from the images due to really intense fluorescence from CPV at longer time points. Colocalization of CPV with mitochondria is shown in white in lower images. Bars 5 µm. (B) Percentage of colocalization of CPV with mitochondria. Colocalization analysis of confocal microscopy images was done with BioimageXD software. Mitochondria were labelled either with anti-COX IV antibody or with MitoTrackerRed (MTR). CPV was labelled with antibody recognizing intact capsids. Mock 24 h p.i. Results are shown as means from 30 cells from 3 independent experiments ± S.D.

Figure 2. CPV infection influences the morphology of mitochondria.

(A) Transmission electron microscopy pictures from mock and CPV infected cells at different time points post infection showing immunogold-labelled CPV to associate with mitochondria. (B) Following a CPV infection damage to the mitochondria can be observed. Images are from 22 h p.i. Infection leads to (i) membrane blebbing (arrowhead), (ii) damage to the membrane (*) and disappearance of cristae (C). (iii) Damaged mitochondria were seen inside autophagosome like structures. C) Percentage of damaged mitochondria was counted with transmission electron microscope from 30 cells from 3 independent experiments. Mock 2 and 22 h p.i. Results are shown as mean ± S.D and there is a statistically significant (p≤0.05) change between mock and CPV infected samples at all time points post infection. Bars 200 nm.

Figure 3. Biphasic loss of mitochondrial membrane potential during CPV infection.

Flow cytometry of JC-1 labelled cells was used to detect depolarization of mitochondrial membrane potential during CPV infection. Results are shown as mean of relative amount of cells showing mitochondrial depolarization from three independent repeats (m, mock infected 18 h p.i.). * The difference is statistically significant (p≤0.05) between marked time point and mock infected cells.

Figure 4. Production of ROS during CPV infection.

The level of ROS was determined by loading the cells with DCFDA and analyzed by flow cytometry. Results are shown as mean of relative amount of cells showing production of ROS from three independent repeats (m, mock infected 24 h p.i.). * The difference is statistically significant (p≤0.05) between marked time point and mock infected cells.

Figure 5. Calcium is not released to the cytoplasm during the entry phase of infection.

Release of calcium from intracellular stores, like the mitochondria, was studied with Fluo-4 dye. Cells were loaded with Fluo-4 and infected with CPV. At indicated time points p.i. fluorescence intensity was measured. Thapsigargin was used as a positive control. Results are means from 3 repeat ± S.D.

Figure 6. ERK1/2 signalling is activated early in CPV infection.

(A) Activation of ERK1/2 was analyzed based on phosphorylation of ERK1/2 (p-ERK1/2). Cell lysates were collected at indicated time points post infection and immunoblotting with antibody towards ERK or p-ERK1/2 was performed. (B) Intensity of band signals were determined with densitometry (ImageJ software) and ratio of p-ERK to ERK was calculated. * The difference is statistically significant (p≤0.05) between marked time point and mock infected cells (m, mock infected 6 h p.i.; U, U0126). (C) CPV infection was carried out in the presence of 20 µM U0126 that inhibits the activation of ERK1/2. At 24 h p.i. cells were fixed and CPV infected cells were visualized with anti-capsid antibody. Infection percentage was determined with immunofluorescence microscopy. The observed decrease in infection percentage was statistically significant (p≤0.05). (D) Cells were infected with CPV or mock infected with or without 20 µM U0126 and JC-1 was used to detect the depolarization of mitochondrial membrane potential. Results are shown as mean of relative amount of cells showing mitochondrial depolarization from three repeats.