Pathological changes in brain tissue in mice infected with JEV
In the present study, 6-week-old BALB/c mice were used because the mortality rate of this strain after infection with JEV was less than 50%, which was conducive to the study of the persistent inflammatory response. The results showed that there was no death of mice after infection. Histological changes in the brain at different time points after JEV infection in mice were observed (Fig. 1). The results showed no significant pathological histological changes in brain tissue in the control group (Fig. 1A) or on day 3 post infection (dpi) (Fig. 1B). There were obvious pathological changes in the brain tissue at 6 dpi (Fig. 1C~1E), as well as significant glial cell hyperplasia in all regions of the brain, including typical glial nodules and vascular cuffs formed by large numbers of lymphocytes exuding around the blood vessels, neuronal degeneration and necrosis, increased cytoplasmic eosinophilia and significant consolidation and fragmentation of neuronal cell nuclei. Brain tissue showed significant pathological histological changes at 9 dpi (Fig. 1F), mainly in the form of glial cell proliferation and glial nodules, occasional necrosis of neurons, which was mostly degenerative, increased cytoplasmic eosinophilia, and vascular cuffing. Brain histopathology was compared on different days after infection and revealed significant pathological changes at 6 dpi, including histopathological changes typical of viral encephalitis. However, the pathological changes were decreased at 9 dpi, probably due to the activation of glial cells to resist the inflammatory response triggered by the virus.
Microglia and virus-infected cells on different days after infection with JEV were labeled by double immunofluorescence assay (IFA) (Fig. 2A). The results showed that in the absence of infection, a small number of microglia underwent activation (IBA-1+), but the cells were rounder and had fewer protrusions. The number of microglia was increased compared to that of uninfected cells at 3 dpi, the cells were still not significantly deformed, and there was no obvious viral antigen present. A large number of virus-positive (JEV+) cells appeared at 6 dpi, microglia were significantly more activated and had longer and elongated protrusions, and microglia were clearly clustered around virus-infected cells. The number of virus-infected cells was increased at 9 dpi, the cells were surrounded by a large number of activated microglia, and some microglial protrusions were in contact with the infected cells.
A comparative analysis of the number of positive cells and the mean fluorescence intensity (MFI) of JEV-infected cells and microglia was performed (Fig. 2B~2C). The number of JEV+ cells and the MFI at 6 and 9 dpi were significantly higher than those in the control group (P<0.05), showing a higher viral load in mouse brain tissue at 6~9 dpi. The number of positive cells and the MFI of microglia at 3~9 dpi were significantly higher than those in the control group (P<0.05), indicating that glial cell activation dominated neuroinflammation in brain tissue, and the difference was most pronounced at 6 dpi (P<0.0001), suggesting a major role for microglia during postinfection pathological changes. Pearson's correlation coefficient (PCC) was used to show the colocalization of JEV-infected cells with other cells, and the results are shown in Table 1. The values for JEV+ cells and IBA-1+ cells before and after infection were greater than 0 and less than 0.5 and were randomly correlated; that is, JEV+ cells and microglia were randomly colocalized.
JEV infects dopaminergic neurons
To confirm whether JEV infects neurons, dopaminergic neurons that mainly secrete dopamine and are concentrated in the midbrain were selected as research objects, and tyrosine hydroxylase-positive (TH+) neurons and JEV+ cells were stained (Figure. 3A). The results showed that TH+ neurons were distributed in the midbrain and were not infected, and dopaminergic neurons were also distributed in the brain and diencephalon. No significant JEV+ cells were found at 3 dpi, and the number of dopaminergic neurons in the midbrain was not significantly reduced. A large number of JEV+ cells appeared at 6 dpi, there was obvious colocalization with TH+ neurons, and there was an obvious colocalization signal in the midbrain region. JEV+ cells were abundant and colocalized with TH+ neurons in the diencephalon and midbrain at 9 dpi, which indicated JEV-infected dopaminergic neurons. The number and MFI of JEV-infected cells and dopaminergic neurons were analyzed (Fig. 3B~3C). The JEV+ cell results were consistent with the dual immunofluorescence analysis of microglia and JEV+ cell colocalization. The JEV+ cell results were significantly different from those of the control group at 6 and 9 dpi (P<0.05). Statistical analysis of TH+ neurons showed that there was no significant difference in the number of positive cells before and after infection, but the number of positive cells at 6~9 dpi was slightly less than that in the control group and at 3 dpi, which may be because the number of positive cells decreased due to the death of neurons caused by JEV infection. On Day 6, the MFI was significantly lower than that in the control group, indicating that JEV affected the ability of dopaminergic neurons to secrete dopamine, resulting in a decrease in tyrosine hydroxylase release, which could affect the subsequent functional role of dopamine. JEV+ cells and TH+ cells were colocalized, and the results are shown in Table 1. Before infection, the PCC value was close to 0, and at 6 dpi, it was close to 1, while at 9 dpi, it was close to 0.5. Compared with that in the control group and 3 dpi, the PCC value was close to a positive correlation; that is, JEV-infected cells and dopaminergic neurons had more obvious colocalization.
Dopaminergic neurons exhibit NLRP3/caspase-1/GSDMD-dependent pyroptosis
When pyroptosis occurs, it involves inflammatory changes in tissue, and one of the keys to the occurrence of pyroptosis is inflammasome activation. It has been demonstrated that the NLRP3 inflammasome is present during the development of neuroinflammation. Immunohistochemistry was used to examine NLRP3 and ASC (Fig. 4), which are components of the NLRP3 inflammasome. The staining showed positive signals for NLRP3 and ASC in the whole brain before and after infection with JEV, but the positive rate was less than 50% before infection, suggesting that the expression of NLRP3 and ASC was too low to cause an inflammatory response. However, after infection with JEV, the positive rate of NLRP3 gradually increased with time, and there was a significant difference between that at 9 dpi and the control group (P<0.01), which reached 80%. The positivity rate of ASC was increased after infection, and there was a significant difference (P<0.05) between 6 dpi and the control, while there was a slight decrease at 9 dpi.
To examine the pyroptosis pathway in dopaminergic neurons, TH+ cells were stained with caspase-1, which is a key factor in the classical pathway (Fig. 5A). The quantitative results (Fig. 5B~5C) showed that a low amount of caspase-1 expression was observed before infection and at 3 dpi, while the number of caspase-1+ cells was significantly higher than that in the control group at 6 dpi (P<0.05). The MFI was significantly increased at 6 dpi (P<0.001). The number of TH+ cells was consistent with the relevant indices of GSDMD, and the MFI at 6 dpi was significantly higher than that in the control group (P<0.0001). The PCC index was higher than 0.5 and close to 1 at 3~9 dpi, indicating a positive correlation, but the correlation was not significant. However, pyroptosis may involve the classical pathway mediated by caspase-1.
To examine pyroptosis in dopaminergic neurons induced by JEV, dopaminergic neurons and GSDMD were stained (Fig. 5A). TH+ cells were found in the whole brain before and after infection and were concentrated in the mesencephalon and diencephalon. The MFI showed that the number of TH+ cells at 6 dpi was significantly higher than that in the control group (P<0.05) (Fig. 5D~5E), which may be because the selected area is rich in dopaminergic neurons. TH+ cells have a compensatory effect in response to death and secrete a large amount of dopamine to meet the needs of the body. GSDMD staining showed that there were GSDMD+ cells before and after infection, but the positive signal was significantly stronger than that in the control group at 6 and 9 dpi. The quantitative results showed that the number of GSDMD+ cells at 6 and 9 dpi was significantly higher than that in the control group (P<0.05), which was consistent with the MFI and the number of positive cells and was significantly higher than that in the control group (P<0.05) (Fig 5E). The PCC was greater than 0 before and after infection, was close to 0.5 at 3 dpi and was close to 1 at 6~9 dpi, indicating a positive correlation and the presence of pyroptosis in dopaminergic neurons after infection; GSDMD was the key factor.
Table 2 Results for different indicator colocation coefficients Pearson coefficients (PCC)
Group
|
MOCK
|
3d
|
6d
|
9d
|
JEV+/IBA-1+
|
0.56±0.30
|
0.47±0.15
|
0.24±0.21
|
0.46±0.12
|
JEV+/TH+
|
0.24±0.16
|
0.38±0.27
|
0.54±0.3
|
0.46±0.27
|
TH+/caspase-1+
|
0.32±0.28
|
0.52±0.31
|
0.55±0.30
|
0.52±0.16
|
TH+/GSDMD+
|
0.46±0.10
|
0.51±0.05
|
0.88±0.08
|
0.82±0.06
|
Morphologically, JEV-infected mice underwent pyroptosis in an NLRP3/caspase-1/GSDMD-dependent manner on Day 6 via the classical pathway. To verify the morphological results, relevant molecules were examined at the RNA level (Fig. 6A). GSDMD expression was significantly higher than that in the control group at 6 and 9 dpi (P<0.001), caspase-1 expression was significantly higher than that in the control at 6 dpi (P<0.05), NLRP3 expression was significantly higher at 6~9 dpi (P<0.01) and ASC expression was significantly higher at 6 and 9 dpi (P<0.05 and P<0.01). The inflammatory factors IL-1β and IL-18 were examined after the onset of pyroptosis, and IL-1β was significantly elevated at 6~9 dpi (P<0.05), while IL-18 was elevated post infection and was significantly higher than that in the controls at 9 dpi (P<0.001). TNF-α and HMGB1, which are two inflammatory factors that play important roles in microglial activation, were also examined and were elevated post infection. HMGB1 was significantly higher than that in the controls at 6 dpi (P<0.001). The increase in these genes and inflammatory factors at 6~9 dpi shows that JEV induces pyroptosis and that the inflammatory response continues to occur in the later stages of JEV infection, further activating microglia and promoting pyroptosis.
Staining confirmed that the expression levels of GSDMD, caspase-1, NLRP3 and ASC were significantly increased after infection with JEV. Moreover, the mRNA levels of these factors were increased during JEV infection. To verify pyroptosis at the protein level, these factors were examined by western blotting (Fig. 6B~6C). The expression of GSDMD was low before infection but gradually increased with time after infection, and the difference was significant at 6 and 9 dpi compared with that in the control group (P<0.05 and P<0.01). When caspase-1 is not activated, it exists in the form of pro-caspase-1. After being activated, it is cleaved into caspase-1, which has a size of approximately 20 kDa. Therefore, pro-caspase-1 and caspase-1 were detected to confirm the activation of caspase-1. The results showed that pro-caspase-1 expression was high before infection, increased with time after infection and was significantly higher than that in the control group at 9 dpi (P<0.05), suggesting increased caspase-1 activation after infection to meet the needs of the body. Caspase-1 expression was significantly elevated at 6 and 9 dpi (P<0.05), indicating that caspase-1 plays an important role in the middle and late stages of infection. NLRP3 and ASC levels were elevated after infection, and ASC expression was significantly higher at 6~9 dpi than in the control group (P<0.01), indicating that the NLRP3 inflammasome plays a key role in the pathogenesis of pyroptosis.
Obvious changes in dopaminergic neurons infected with JEV
MN9D cells were infected with JEV and observed under a light microscope (Fig. 7). Changes in the cells were observed 24 h post infection (hpi). The control group had long spindles with regular morphology and no secretions. The JEV infection group was treated at different time points after infection. The cells began to change at 24 hpi; the cells were shriveled and rounded, the shape was irregular, the secretions increased, the cell membrane was broken at the later stage, a large number of cell fragments appeared, and the number of floating cells increased.
To determine whether viral infection caused the rupture of the cell membrane, the level of cell damage was examined by measuring LDH levels in cell culture supernatants (Fig. 7B). LDH release gradually increased with time after infection with JEV and was significantly higher at 48~60 hpi than in the control group (P<0.01), suggesting that cell membrane permeability increased significantly in the later stages of infection and that cell membrane rupture most likely occurred, causing changes in cell morphology and promoting the release of inflammatory factors, which would further induce pyroptosis.
JEV infection of dopaminergic neurons was examined by immunofluorescence staining of JEV+ cells (Fig. 8A). Positive signals appeared 24 hpi and were mainly in the cytoplasm, and with time, an increasing number of cells became infected, indicating JEV infection of dopaminergic neurons. However, this phenomenon became less evident at the later stage when the number of surviving cells decreased due to viral infection, and more green signals appeared at the edges of the cells, which influenced the interpretation of the results.
The susceptibility of dopaminergic neurons to JEV was examined, but whether viral proliferation changed at different time points was not known. Therefore, the changes in viral load at different time points after infection were examined by qPCR (Fig. 8C). The virus increased significantly after JEV infection (P<0.0001). JEV was amplified at 12 hpi; the highest levels were observed at 24 h and were slightly decreased at the later stage. This may be because the number of cells decreased in the later stage, and a large number of virions were released into the cell culture supernatant, which reduced virus levels. To confirm JEV entry into cells, E protein was examined (Fig. 10B). The results showed that the virus protein proliferated in cells at 12 hpi, the levels of virions increased with time, and there was a significant difference at 36 and 60 h compared with that in the control group (P<0.05).
JEV induces pyroptosis in dopaminergic neurons through the NLRP3 inflammasome
JEV induced pyroptosis through the NLRP3 inflammasome in vivo, and thus, changes in NLRP3, ASC, caspase-1 and GSDMD levels in MN9D cells were examined (Fig. 9). The results showed that the expression of NLRP3, ASC, caspase-1 and GSDMD was low before infection, NLRP3 was significantly increased from 24~48 hpi after infection, and its expression decreased at 60 hpi due to a decrease in cell number. ASC showed an obvious positive signal at 48~60 hpi. Caspase-1 and GSDMD showed positive signals at 24 hpi, and their expression gradually increased with time, proving that dopaminergic neurons underwent significant pyroptosis in the middle and late stages of infection, which promoted inflammation. Based on IFA of factors associated with the NLRP3/caspase-1/GSDMD signaling pathway at different times, the results showed that this inflammatory response was mediated by the NLRP3 inflammasome in the late stages of infection.
Morphological examination showed that JEV-infected dopaminergic neurons underwent pyroptosis. To determine whether JEV infection of dopaminer[5]gic cells promoted this process at the transcriptional level, qPCR was used to measure the mRNA levels of cytokines related to pyroptosis and other inflammatory factors (Fig. 10A). The mRNA levels of GSDMD were significantly increased at 36~60 hpi (P<0.01), caspase-1 increased gradually at 36~60 hpi and was significantly increased at 60 hpi (P<0.05), NLRP3 and ASC were significantly increased at 48 and 60 hpi (P<0.05), and IL-1β was decreased briefly after infection and was significantly increased at 24~60 hpi compared with that of the control group (P<0.05), suggesting that dopaminergic neurons underwent typical pyroptosis in the late stage of virus infection, and the transcription levels of related factors were significantly increased. IL-18 was significantly decreased at 24~36 hpi (P<0.01) but was significantly increased at 60 hpi (P<0.0001). TNF-α was significantly decreased at 12~48 hpi (P<0.05) but was significantly increased at 60 hpi (P<0.01), suggesting that these two cytokines may have different transcription levels at this stage of infection and then exhibit different expression levels.
GSDMD was expressed before and after infection, and its expression gradually increased at 36~60 hpi and was significantly different from that in the control group at 36 hpi and 60 hpi (P<0.0001) (Fig. 10C). Pro-caspase-1 was expressed before and after infection and gradually increased after infection but was not significantly different from that in the control group, suggesting that caspase-1 underwent cleavage-mediated activation and was involved in the assembly of inflammasomes. Caspase-1 was not expressed before infection, and its expression increased significantly at 36~60 hpi (P<0.01). NLRP3 was expressed before infection, decreased significantly at 12~24 hpi (P<0.05), increased at 36 hpi, and decreased significantly at 48~60 hpi (P<0.05). ASC was expressed at low levels before infection and gradually increased with time, showing a significant increase at 36 hpi and 60 hpi (P<0.001, P<0.01) and a large decrease at 24 hpi. The results of protein analysis indicated that JEV-infected cells exhibited a significant increase in the expression of pyroptosis-related factors at 36~60 hpi, promoting the development of pyroptosis.