Schmallenberg virus pathogenesis, tropism and interaction with the innate immune system of the host

Abstract

Schmallenberg virus (SBV) is an emerging orthobunyavirus of ruminants associated with outbreaks of congenital malformations in aborted and stillborn animals. Since its discovery in November 2011, SBV has spread very rapidly to many European countries. Here, we developed molecular and serological tools, and an experimental in vivo model as a platform to study SBV pathogenesis, tropism and virus-host cell interactions. Using a synthetic biology approach, we developed a reverse genetics system for the rapid rescue and genetic manipulation of SBV. We showed that SBV has a wide tropism in cell culture and "synthetic" SBV replicates in vitro as efficiently as wild type virus. We developed an experimental mouse model to study SBV infection and showed that this virus replicates abundantly in neurons where it causes cerebral malacia and vacuolation of the cerebral cortex. These virus-induced acute lesions are useful in understanding the progression from vacuolation to porencephaly and extensive tissue destruction, often observed in aborted lambs and calves in naturally occurring Schmallenberg cases. Indeed, we detected high levels of SBV antigens in the neurons of the gray matter of brain and spinal cord of naturally affected lambs and calves, suggesting that muscular hypoplasia observed in SBV-infected lambs is mostly secondary to central nervous system damage. Finally, we investigated the molecular determinants of SBV virulence. Interestingly, we found a biological SBV clone that after passage in cell culture displays increased virulence in mice. We also found that a SBV deletion mutant of the non-structural NSs protein (SBVΔNSs) is less virulent in mice than wild type SBV. Attenuation of SBV virulence depends on the inability of SBVΔNSs to block IFN synthesis in virus infected cells. In conclusion, this work provides a useful experimental framework to study the biology and pathogenesis of SBV.

Figure 1. In vitro growth kinetics of SBV.

A. Growth kinetics of SBV in a variety of established cell lines as indicated in the panel. Cells were infected with SBV at a MOI of 0.05 and supernatants were collected at 4, 8, 24 and 48 h post-infection. Supernatants were then titrated in CPT-Tert cells by limiting dilution analysis.

Figure 2. Reverse genetics protocols for the rescue of SBV.

A. Schematic representation of the antigenome plasmids used to rescue SBV. B and C. Schematic representation of the strategies used for the rescue of SBV in BSR-T7/5 and 293T cells as described in the text. The two methods are very similar with the exception that BSR-T7/5 stably express the T7 RNA polymerase and therefore these cells are transfected only with the SBV antigenome plasmids. On the other hand 293T cells are transfected with the antigenome plasmids and an expression plasmid for the T7 RNA polymerase.

Figure 3. In vitro phenotypic characterization of SBV and sSBV.

A. Comparison of plaques produced by wild type SBV and sSBV rescued using BSR-T7/5 and 293T cells. B. Growth kinetics of SBV and sSBV. CPT-Tert and BFAE cells were infected at a MOI of 0.05 for 90 min. Supernatants were collected at the indicated times post-infection and virus titer was measured using standard plaque assays in CPT-Tert cells. C. The presence of SBV in CPT-Tert and BFAE cells infected with wild type and rescued virus was confirmed by western blotting using antibodies against the SBV N (nucleocapsid) protein.

Figure 4. Sequences of 3′ and 5′ UTRs of SBV obtained by RACE PCR.

A schematic representation of the genomes of the 3 viral segments is shown. Sequences on the top of each segment (sSBV) indicate the sequence of the UTRs of the plasmids used for reverse genetics. The middle sequences indicate the UTRs sequences inferred by RACE PCR (RACE1, RACE 2 etc). The bottom sequences indicate SBV UTRs reported in GenBank (HE649914, HE649913 and HE649912). Positions highlighted in red correspond to nucleotides inferred for the construction of sSBV sequences based on the AKAV sequence. Scores represent positions in the genome segments for which there was no sequence available in the submitted GenBank sequences. Note that 54 nucleotides previously reported in the 5′ UTR of the M segment that appeared to be a sequence artifact (i.e. not part of the SBV genome) were not, as expected, detected by RACE and are not shown in the figure. Numbers shown in this figure take into account the corrected 3′ and 5′ UTRs.

Figure 5. Neuropathology of SBV infection in mice inoculated intracerebrally.

A. Left panel: survival plots of 2-day old NIH-Swiss mice inoculated intracerebrally using the indicated viruses or cell culture media as a control. Right panel: survival plots of 10 and 18-day old NIH-Swiss mice inoculated as described in A. B–E histopathology of brain sections from SBV-infected infected mice stained with hematoxylin. B and C. brain sections from SBV-infected mice at 72 h post-infection. Arrows indicate areas of malacia and hemorrhage. D and E brain sections from SBV infected mice at 120 h post-infection, showing areas of vacuolation in low and high magnification respectively. F–K immunohistochemistry of brain sections of mock-infected mice (F–G), or SBV infected mice at 48 h (H–I) or 72 h (J–K) post-infection using an SBV N antiserum as described in Materials and Methods. Bars = 2 mm in B, F, H, J; 200 µm in C; 500 µm in D; 100 µm in E, G, I, K.

Figure 6. SBV tropism in the central nervous system of naturally in utero infected lambs and calves.

A. Stillborn lamb showing signs of congenital SBV infection including artrogryposis, brachygnatia inferior and torticollis. B–D: Tissue sections stained with hematoxylin and eosin derived from brain of a lamb congenitally infected with SBV. Panel B illustrates cerebral cortex with porencephaly. The white matter is lacking as indicated by (*). Adjacent grey matter is reduced in thickness (bar = 2 mm). C. Higher magnification micrographs showing areas of malacia and the presence of myelin laden macrophages (arrow; bar = 20 µm). D. Glial nodule (arrow) and mild lymphohistiocytic perivascular infiltrate (bar = 50 µm). E–L: Immunohistochemistry of tissue sections derived from brain (E–H) or spinal cord (I–J) of lambs congenitally infected with SBV. In sections shown in panels E, F, I and J a SBV N polyclonal rabbit antiserum was used, while panels G–H show instead serial sections (of E and F) incubated with the pre-immune serum. The use of SBV N antiserum reveals a strong positive reaction (characterized by the intracytoplasmic dark brown staining) in the cell body and processes of neurons of the grey matter, while no staining was observed with serial sections incubated with the pre-immune serum. K–L: Immunohistochemistry of control tissue sections derived from brains of sheep reared in Scotland and probed with the antiserum towards the SBV N protein showed no immunoreactive cells.

Figure 7. In vitro phenotypic characteristics of SBVΔNSs and SBVp32.

A. Strategy used to generate SBVSΔNSs. Nucleotide (antigenome) and amino acid sequences showing the translation start site of the NSs in an overlapping reading frame within the N gene are indicated (top sequences). Bottom sequences indicate silent mutations (low case) introduced in N that abrogate expression of the NSs protein. B. Growth kinetics of SBV lacking the NSs protein and SBVp32. CPT-Tert cells were infected at a MOI of 0.05 for 90 min. Supernatants were collected at the indicated times post-infection and the virus titer was measured using standard plaque assays in CPT-Tert cells. C. Schematic representation showing the nucleotide differences (red dots) found in SBVp32 in relation to SBV. Only the coding regions of the antigenomes of the 4 known genes are shown. Non-synonymous mutations are indicated with an asterisk.

Figure 8. Virulence of SBV mutants in suckling NIH-Swiss mice.

A. 3 and 7-day old mice were inoculated intracerebrally with either 100 or 400 PFU with the indicated viruses or cell culture media as a control. Survival plots show that SBVΔNSs possesses an attenuated phenotype while SBVp32 is more virulent than sSBV. B. Immunohistochemistry of brain sections derived from NIH-Swiss mice inoculated with sSBV or SBVp32 and killed at various time points post-infection as indicated in the figure. Immunohistochemistry was performed using an SBV N antiserum as described in Materials and Methods. At the early time points SBV antigens are detected only in sections derived from SBVp32 infected mice (Bar = 500 µm).

Figure 9. SBVSΔNSs induces the production of IFN.

A. IFN protection assay. 2fGTH cells were infected with the indicated viruses and supernatants were collected 24 h post infection. Supernatants were UV treated to remove infectious virus and fed to CPT-Tert cells after serial dilution. CPT-Tert cells were later infected with EMCV and the presence of CPE monitored and compared to cells supplemented with known amounts of universal IFN. B. The induction of IFN-β mRNA was investigated by RT-PCR from RNA extracted from 2fGTH cells infected with virus as indicated or transfected with Poly I∶C as a positive control. To control for the presence of residual genomic DNA all the samples were amplified after reverse transcription performed without reverse transcriptase (top panel, indicated as+/−RT). The quality of the extracted RNA was verified by the amplification of the 45S ribosomal RNA (middle panel). The presence of virus was confirmed by the amplification of part of the SBV S segment (bottom panel). C. IFN protection assay performed in primary ovine trophoblast cell (oTr-1) and primary ovine endothelial cells as described in A. D. Survival plots of 7 day old IFNAR^(−/−) mice inoculated intracerebrally with sSBV, SBVSΔNSs or cell culture media as a control. Data indicate that SBVΔNSs is as virulent as sSBV in these mice that lack an intact IFN system.