The WH2012 strain of TBEV has undergone several passages in our lab since it was first isolated. To determine the genetic homogeneity in the viral stock, RT-PCR products were amplified from the viral genome and cloned into the T vector. Multiple clones targeting at the same viral segment were sequenced. As shown in Table 1, there were thirteen non-synonymous substitutions that resulted in ten amino acid mutation in the parental WH2012 strain. In addition, there were also two nucleotide mutations and a deletion of two nucleotides in 5' UTR as well as twenty-three synonymous substitutions in the ORF region. As shown in Fig. 1, TBEV Pn replicated well and caused strong cytopathic effect (CPE) which could be used in endpoint dilution assay on BHK-21 cells (TCID50). However it did not produce plaques in BHK-21 cells (Fig. 1C). Even after five rounds of blind passages, TBEV Pn+5 still did not produce visible plaques in BHK-21 cells (Fig. 1A and 1C). In addition, qRT-PCR was also performed to analyze the viral copies of the serially passaged supernatants. Both TBEV Pn and TBEV Pn+5 reached viral loads of > 108 copies/mL with no significant differences (P > 0.05, Fig. 1D). Thus, there were a variety of variants in the present viral stock, the consensus sequence of the original WH2012 virus (each nucleotide only confirmed in more than 75% of the cloned T vectors was adopted as the consensus nucleotide) was used in the following construction of infectious TBEV cDNA clone.
Base position WH2012 Sequencing resultsa Amino acid change Location 50-52 - CA - 5'UTR 71 C T - 5'UTR 75 T C - 5'UTR 3498-3499 TC CG S-R NS1347 3597 G A G-R NS2A28 5811 A G N-D NS3405 6132 C G R-G NS3512 6281 T C - NS3561 6317 C T - NS3573 6404 T C - NS3602 7207 C T A-V NS4B100 7408 C T P-L NS4B167 7670 T C - NS52 8561 C T - NS5299 9468 C G R-G NS5602 9471 A T T-S NS5603 9475 A G E-G NS5604 9503 C T - NS5613 9506 T A - NS5614 9612-9614 AAC GAT N-D NS5650 9659 T C - NS5665 9743 G A - NS5693 9791 C T - NS5709 9794 T C - NS5710 9824 A G - NS5720 9836 C T - NS5724 9914 C T - NS5750 9920 T C - NS5752 9968 T C - NS5768 10, 121 T C - NS5819 10, 145 C T - NS5827 10, 199 C T - NS5845 10, 226 G A - NS5854 10, 247 T C - NS5861 10, 304 C T - NS5880 10, 334 T C - NS5890 aEach nucleotide substitution was observed and confirmed in more than 75% of the cloned T-vectors.
Table 1. Sequence differences in the parental WH2012 virus.
The full-length TBEV cDNA clone, TBEV-FL, was constructed based on the above consensus sequence of the WH2012 strain. As is shown in Fig. 2, we used a low-copy vector pACYC177 which works efficiently in the ZIKV infectious clone (Li et al. 2018). A cytomegalovirus (CMV) promoter was engineered right before the authentic 50 end of the viral sequence for in vivo transcription, and the connecting of a HDVr sequence and a SV40 poly(A) was placed downstream the 3' UTR to ensure that the 3' end of the viral RNA was generated correctly. Synonymous substitutions of C5859T, G8416A, and C9354T were introduced to delete the three redundant KpnI, XhoI, and KpnI sites. Synonymous substitutions of C1088A and G4464C were introduced to generate the single ClaI and BinI sites. During cloning of TBEV-FL, we found that fragment B, spanning the viral E-NS1-NS2A-NS2B genes, was unstable in E. coli, even with the use of pACYC177, leading to aberrant deletions/mutations in plasmids. To circumvent this issue, we inserted an intron in NS1 to disrupt the sequence of ORF, and postulated that the viral RNA would be spliced and restored in mammalian cells (Fig. 2). Overall, TBEV-FL assembled as depicted in Fig. 2 propagated stably E. coli, and sequence analysis of the plasmid revealed no nucleotide substitution.
BHK-21 cells were transfected with TBEV-FL to characterize its transcription in vivo and protein expression. Meanwhile, BHK-21 cells transfected with pACYC177 were used as the negative control, and BHK-21 cells infected with the wild-type TBEV (TBEVwt) at an MOI of 0.01 were used as the positive control. Three days later, total RNAs of the transfected cells were extracted and subjected to qRT-PCR analysis. As is shown in Fig. 3A, significant increasing copies of viral RNA were detected in the TBEV-FL-transfected cells. To further determine if splicing occurs in the transfected cells, total RNA was extracted from the supernatants at 4 d.p.t or infection with TBEVwt. A DNA fragment that contains the intron was amplified with a pair of primers surrounding the intron (Supplementary Table S1). As is shown in Fig. 3B, the similar single DNA fragments in size were amplified from the supernatants of both transfection with TBEV-FL and infection with TBEVwt. Sequence analysis of these RTPCRs showed that all the fragments from the supernatants of transfection with TBEV-FL were found in the absence of the inserted intron (Fig. 3C), which meant the introninterrupted fragments were recreated in BHK-21 cells.
Figure 3. Characterization of viral RNA replication and splicing in TBEV plasmid-transfected cells. A Viral copies in cells transfected with TBEV-FL were determined at 3 d.p.t. by qRT-PCR. B Agarose gel analysis of RT-PCR products from the supernatants of the TBEV-FL-transfected cells at 4 d.p.t. M, Marker; band 1, Water; band 2 and 3, Unspliced TBEV-FL; band 4, MOCK; band 5, Supernatants of transfection with TBEV-FL; band 6, Parental WH2012. C Sequence results of RT-PCR products from the supernatants of the cells transfected by TBEV-FL. The two reference sequences were parental WH2012 virus and unspliced TBEV-FL plasmid. Data shown (A) are the mean ± SD analysed by Student's t-test (two tailed) (***, P < 0.001).
To analyze the infectious TBEVic production, BHK-21 cells transfected with TBEV-FL were used to rescue infectious virion. Obvious CPEs were observed in the transfected cells since 2 d.p.t. (Fig. 4A). After harvesting the supernatant, the robust expression of NS5 was detected by Western blotting in cell samples (Fig. 4B), in another way, suggesting that the inserted intron was precisely excluded in mammalian cells, otherwise, the frameshift mutations would generate in the polyprotein following closely on the heels of the inserted intron. Though the parental WH2012 virus did not produced plaques (Fig. 1C), plaque assay was still performed in BHK-21 cells with the supernatant (TBEVic) harvested from transfected cells. Unexpectedly and interestingly, classic plaques with a size of 0.8–0.9 mm in diameter were observed at 4 d.p.i. (Fig. 4C, 4E) with a peak titer of 106.39±0.41 PFU/mL. Protein E and NS5 expression in BHK-21 cells infected with TBEVic could be also detected by IFA (Fig. 4D, 4F). To visualize the ultrastructural features of TBEVic directly, TEM assay was used to determine the infected cells. As is shown in Fig. 5, BHK-21 cells infected with TBEVic and TBEVwt presented with similar spherical particles ranging from 35 to 50 nm in diameter. Collectively, these data demonstrated that infectious TBEVic could be rescued by TBEV-FL cDNA clone.
Figure 4. Rescue of TBEVic from the infectious TBEV clone in cell culture. A The CPE on BHK-21 cells transfected with the TBEV-FL at 2 d.p.i. All the images were captured at a × 20 magnification. B Western bolt analysis of TBEV NS5 protein expression in BHK-21 cells transfected with the TBEV-FL. C and E The plaque morphology of the TBEVic virus in BHK-21 cells, visualized using 0.5% crystal violet solution following incubation for 4 days. D IFA of TBEV E protein expression in BHK-21 cells infected with either TBEVic or mock-infected at 2 d.p.i. (MOI = 0.02), Red represents TBEV E protein, and blue represents nuclei (stained with Hoechst 33, 258). F IFA of TBEV NS5 protein expression in BHK-21 cells infected with either TBEVic or mock-infected at 2 d.p.i. (MOI = 0.1). Red and blue represent TBEV NS5 protein and nuclei (stained with Hoechst 33258), respectively. D and F All the IFA images were captured at a × 10 magnification.
Figure 5. Transmission electron micrographs of BHK-21 cells infected with TBEVic and TBEVwt. Red arrows represented sphere particles in BHK-21 cells infected with the corresponding viruses. The right panels showed the enlarged views of the corresponding regions in the left panels. Scale bars are 2 lm and 500 nm for low magnification images and the high magnification images, respectively.
To compare the replicative fitness of the rescued and parental WH2012 virus, BHK-21 cells were infected by the two viruses at the same MOIs and the infectious titers were determined by a standard TCID50 assay and a qRT-PCR assay. As is shown in Fig. 6A, both TBEVic and the parental virus caused similar typical CPEs, such as cell rounding, necrosis, and detachment. Compared with the parental TBEVwt, TBEVic exhibited a slightly low replication kinetic in BHK-21 cells, however, there was no significant differences (P > 0.05) before 4 dpi (Fig. 6B). Both TBEVic and TBEVwt reached undifferentiated maximum viral titers (106.88±0.17 TCID50/mL and 107.13±0.12 TCID50/mL, P > 0.05) at 3 dpi (Fig. 6B). Only the endpoint titer of TBEVic was significant lower than that of the parental TBEVwt (P < 0.05, Fig. 6B). We next compared the viral RNA levels of TBEVwt and TBEVic at different time points after infection of BHK-21 cells. Results demonstrated that there were no significant differences between TBEVwt and TBEVic in RNA replication kinetics. Overall, infectious TBEVic reproduced from the infectious clone of TBEV-FL exhibited the similar replicative characteristic to its parental virus WH2012, enabled extensive analysis of the biological properties, potential virulent determinants, and immunogenicities of TBEV in vitro and in vivo.
Figure 6. Replication of TBEVic and the parental TBEVwt virus in cell culture. A Apparent and similar CPE of TBEVic and TBEVwt on BHK-21 cells. B and C The supernatants from BHK-21 cells transfected with the full-length TBEV-NS5DGDD (the conserved polymerase motif GDD in NS5 was mutated to GAA, G663G-D664A-D665A) cDNA clone were used as the replication-deficient negative control. B BHK-21 cells were infected with viruses at an MOI of 0.1. Virus titers were measured by a TCID50 assay on BHK- 21 cells. C BHK-21 cells were infected with viruses at an MOI of 0.0001. Viral RNA copies were measured by a qRT-PCR assay on BHK-21 cells. Data shown (B and C) are the mean ± SD analyzed by Student's t-test (two tailed) (*, P < 0.05).