Table 1.
Sequences of the primers used for RT-qPCR.
Fig 1.
Susceptibility of human and rodent cells to orthohantavirus infection.
Cells grown on glass coverslips were inoculated with a MOI of 1 with the different orthohantaviruses, then used at dpi3 and dpi7 for intracellular immunofluorescence staining of the viral N protein using the A1C5 monoclonal antibody to evaluate the infectivity of PUUV (orange bars), TULV (blue bars), and PHV (green bars). The histogram in (A) shows the percentage of infected cells (N+) of human origin derived from liver (HuH7), lung (A549), intestine (Caco2), as well as differentiated monocytes (THP1dif) and vascular endothelial cells (HUVEC). The non-human primate Vero E6 cells used to prepare virus stocks were added as a control. Histogram in (B) shows the susceptibility of rodent cell lines to orthohantaviruses, including renal cells (MyglaSWRecB and BVK168) and airway epithelia (MyglaSWTrach and MyglaAECcl2) from bank vole. The results were obtained from at least three independent experiments and the error bars correspond to standard deviation to the mean. Statistical analysis between dpi3 and dpi7 was performed using Two-way ANOVA and are shown according to p-values represented by *p-value <0.0332, **p-value<0.0021, ***p-value<0.0002, ****p-value<0.0001. Non-significant variations are not marked and are further detailed in S1 Fig. Pictures in (C) show N immunofluorescence detection at dpi 3 in Vero E6, HuH7 and MyglaSWRecB cells infected by PUUV, TULV or PHV. Viral N protein is stained in green and the nucleus in blue with DAPI (4′,6-diamidino-2-phenylindole). The scale bar corresponds to 100 μm.
Fig 2.
Ability of human and bank vole cells to produce infectious viruses.
A schematic representation of the workflow of experiments performed to evaluate the capacity of production of infectious virus by the cell lines further used in the study is shown in (A). Supernatants of Vero E6, HuH7 and MyglaSWRecB cells primarily infected (1st) at MOI 0.5 by PUUV (B), TULV (C) or PHV (D) were recovered at dpi 7 for determination of their infectious titers on Vero E6 cells (left panels) and then for their ability to secondary (2nd) infect HuH7 and MyglaSWRecB cells (middle panels). Infectious titers were calculated from the percentage of N+ cells determined by immunofluorescence staining at dpi 3 as described in the Materials and Methods section. In parallel, copies of viral RNA produced in the three cell lines infected by PUUV (B), TULV (C) and PHV (D) were quantified at dpi 3 and dpi 7 (right panels) both, from the cellular fraction (intracellular viral RNA) and from their supernatants (released viral particles). The histograms show the copy numbers of viral RNA per sample calculated based on RNA standard curves corresponding to S segments specific of each orthohantavirus. The same viral preparation was used for each virus to infect the different cell lines (input). The error bars represent the standard deviation to the mean of technical replicate of biological triplicate for each sample. In (E), an IFN-λ neutralization assay was performed. PUUV, TULV or PHV were either left untreated (NT) or were pretreated with anti-human IFN-λ1 or with anti-human IFN-λ2 antibodies prior to infection at a MOI of 0.5, then their infectious titers on Vero E6 and HuH7 human cells were compared at dpi 3. Statistical analyses were performed to compare the amount of RNA between day 3 and day 7 in the cell lysates, as well as the amount of RNA among day 3, day 7 and the input in the supernatant. For neutralization assay, infection in absence of neutralizing antibodies was compared to those in presence of either anti-IFNλ1 or anti-IFNλ2. “ns” indicates not significant differences, while the stars (*) mark significant variations as explained in Fig 1. RecB was used as an abbreviation of MyglaSWRecB in the middle B-D panels.
Fig 3.
Maturation of viral particles assessed by electron microscopy and immunofluorescence assay.
Panels A-G show thin sections of fixed cells processed by transmission EM. Vero E6 cells were either infected with TULV (A-C) or remained non-infected (D). The lower panels in A and B correspond to a magnification of the plasma membrane area framed by a square in the upper panels. Viral particles (insets), at the plasma membrane and in the extracellular space, are marked with large black arrows and tubular structures are indicated with thin black arrows. In (C), the white asterisk highlights the filamentous structure induced by TULV infection. HuH7 cells infected by TULV (E, F) only show tubular structures at the plasma membrane and the extracellular space (thin black arrows). Multivesicular vacuoles containing viral particles (black arrowheads) and their magnification are seen in MyglaSWRecB cells infected by PHV (G). Immunofluorescent staining of membrane (Mb) and intracellular (IC) viral glycoprotein Gn is shown in panel H for Vero E6 cells infected with PUUV, TULV and PHV. The scale bar corresponding to 20 μm indicates the magnification.
Fig 4.
Distribution of N proteins relative to cellular compartments.
Vero E6 cells grown on glass coverslip were infected with PUUV, TULV or PHV at MOI 1. Intracellular co-localization appeared in yellow in the merge panels following staining of the N protein in green, and markers of different cell compartments in red. In (A) the co-localization of N with compartments of the secretory pathway (ER, ERGIC and Golgi) is shown, while in (B) the co-localization with the early (EE), late (LE) and recycling endosomes from the endocytic pathway is shown. Recycling endosomes were revealed via transferrin trafficking (Tf) in presence of BFA. DAPI labels nuclei in blue.
Fig 5.
Co-localization of viral RNA with filamentous structures of N protein.
Vero E6 cells infected with PUUV, TULV or PHV were treated for in situ hybridization by incubation with fluorescent RNA probes either complementary of the viral genomic (-)RNA (green) or of the antigenomic (+)RNA (red). The viral nucleocapsids were detected by immunofluorescence staining with the A1C5 antibody but using secondary antibodies conjugated to Alexa 555 (red) or Alexa 488 (green) according to the fluorescence of the RNA probes. Colocalization of the N filaments with viral RNA appears in yellow in the merge panels. White arrows indicate co-localisation of N dots with PUUV RNA probes.
Fig 6.
Interaction of the nucleocapsid of PUUV, TULV and PHV with cellular proteins.
Panel (A) shows the different organizations of PUUV-N, TULV-N or PHV-N, identified by intracellular detection at dpi 3 in infected Vero E6 cells (green staining). Small dots of N protein are marked with yellow arrows, filaments with red arrows, bigger granules with white arrows and aggregates are circled in pale purple. Nuclei are counterstained with DAPI (blue staining). In (B), kinetics of TULV-N distribution is shown at dpi 1, 2 and 3. In (C), interactors of the N of PUUV, TULV and PHV identified by mass spectrometry analysis are shown. Cellular partners, statistically significant (p-value<0.05), are colored in orange, blue and green for PUUV, TULV and PHV, respectively. Further explored interactors are colored in violet. In (D), HuH7 cells were infected with TULV or PHV at a MOI of 0.5 and stained for ribophorin I (red) and viral N (green). Co-localization appears in yellow in the merge panel. In (E), HuH7 infected with PHV were stained for NKRF (red) and viral N (green). White arrows highlight the distribution of small dots of NKRF in the cytoplasm of infected cells while NKRF mostly localized in the nucleus of non-infected cells.
Table 2.
Human proteins found in complex with PUUV-N in HEK293T cells.
Table 3.
Human proteins found in complex with TULV-N in HEK293T cells.
Table 4.
Human proteins found in complex with PHV-N in HEK293T cells.
Fig 7.
Immunostaining validation of identified interactions of N protein with human proteins.
Immunofluorescence co-staining of infected Vero E6 cells was performed as in Fig 4. The localization of viral N protein (green) with markers of P bodies, DDX6, and lipid droplets is shown in (A) and (B) respectively. In (B), the distribution of lipid droplets in non-infected cells (NI) is shown for comparison with infected cells. The co-staining of N and tubulin is shown in (C). Cellular markers are labelled in red while nuclei are counterstained in blue (DAPI).
Fig 8.
Regulation of IFN expression in infected human and bank vole cells.
Human A549 (A) and HuH7 (B) cells and rodent MyglaSWRecB cells (C) were left non-infected (grey bars) or were infected with PUUV (orange bars), TULV (blue bars) or PHV (green bars) at a MOI of 0.5. As positive control, a condition where cells were treated with poly-IC was included (yellow bars). For each condition, the mRNA expression of the different IFN type I (IFNα and β) and type III (IFNλ1, λ2 and λ3), was compared with non-treated cells (2-ΔΔCT) by RT-qPCR. Error bars correspond to the standard deviation from the mean determined from at least three replicates of three independent samples. Statistical analysis showed non-significant (ns) and significant variations relative to non-infected cells with p-values p<0.0332 (*), p<0.0021(**), p<0.0002 (***) and p<0.0001(****).
Fig 9.
Transcriptome analysis of MyglaSWRecB bank vole cells infected with PUUV and PHV.
Graphs in (A) represents the upregulated genes in MyglaSWRecB cells infected with PUUV (left panel) and PHV (right panel), as compared to non-infected (NI) cells, which were identified by RNA-Seq using the Mus musculus reference genome, GRCm38/mm10. Graphs were generated using GraphPad Prism software by plotting the -log10 of the adjusted p-value against the log2 of the fold change in gene expression. Genes colored in blue were up regulated under both viral conditions, while PUUV- and PHV-specifically induced genes are shown in orange and green, respectively. In (B) M. musculus transcriptome data for PHV-infected bank vole cells were analyzed using the on line-software STRING. The up-regulated genes are represented as an interaction network, with the nodes involved in the immune response colored in red, according to the GO terminology. In (C), the most significantly enriched pathways, compared to non-infected cells, are listed based on Reactome classification. The number of genes identified in RNA samples of PHV-infected cells (entities found, light green), is shown relative to the total amount of genes involved in the specific corresponding functions that are indicated (total entities, dark green).
Fig 10.
Quantitative RT-qPCR confirmation of gene expression in bank vole cells.
In (A), relative differential expression of Irgm1, Ifit1, Iigp1, Mpeg1, Ednrb, Msr1 genes measured by RNA-Seq was validated by RT-qPCR using RNA extracted at dpi5 from MyglaSWRecB that were either not infected (NI, grey bars), or infected with PUUV (orange bars) or PHV (green bars). In (B), the expression of Xaf1, Mx1 and Rsad2, was examined in MyglaSWRecB cells and, the expression in infected HuH7 and A549 cells of their human counterparts RSAD2, XAF1, and MX1 is shown in (C) and (D) respectively. The expression of RNA corresponding to the S segments of PUUV and PHV is shown in the right panels of (A) for MyglaSWRecB and of (D) for A549-infected cells. Error bars correspond to the standard deviation from the mean of samples tested in triplicate and statistical analysis performed as shown in Fig 8.
Fig 11.
Schematic representation of the different interactions of orthohantaviruses with human and rodent cells.
The cell susceptibility to PUUV- (orange), TULV- (blue) and PHV- (green) infection, viral replication and production of infectious particles in human A549 and HuH7 cells and bank vole MyglaSWRecB cells, are schematically summarized. Colored arrows, as indicated for each of the three viruses, indicate the main steps of the viral cycle i.e. entry, replication and release. Levels of expression of IFNs and ISGs are visualized as black RNA strands and their relative amount by grey arrows next to their name.