3.1 Identification of activation sites in MEV NS1
Critical amino acid residues in the MEV transactivation domain were determined by alignment of amino acid sequences of the MEV NS1 protein and the well-studied AAV Rep protein. As illustrated in Fig. 1A, the origin-binding domain is located within aa 6 ~ 254 and the helicase domain within aa 340 ~ 556. Between them, there is also an oligomerization domain. The transactivation region of the MEV NS1 protein was identified at aa 557–668 (Fig. 1A). To further characterize transactivation activity, 6 truncation fragments of NS1 were constructed, as shown in Fig. 1B.
F81 cells were transfected with these truncated plasmids together with pMEV-ΔNS1 in which the NS1 expression had been stopped by introduction of a stop codon (Fig. 1C). Western blot analysis showed that transfection with any of the truncated fragments of the NS1 transactivation domain significantly reduced expression of the VP2 protein, and especially with pNS1-Δc63 (Fig. 1D). Detection of VP2 mRNA by qPCR showed that inhibition occurred at the transcription level (Fig. 1E).
3.2 NS1-Δc63 mutation lost its ability to activate transcription of P38 and VP2-5’UTR
The expression of VP2 is controlled by the P38 promoter and the 5’ untranslated region (VP2-5’UTR) at the transcriptional level. We have previously shown that the expression of VP2 from a full-length infectious clone of MEV was abolished when its 5’UTR was deleted. To test whether deletion of 63 amino acids at the C-terminal of NS1 protein inhibited the transcriptional level of VP2, plasmid pP38 + 5’UTR was constructed with a pGL3 backbone in which the luciferase gene had been replaced by the MEV VP2 gene (Fig. 2A). Expression of VP2 protein in F81 cells harvested 24 h post transfection with either pNS1 or pNS1-Δc63 showed that VP2 protein was detectable by Western blot only in the presence of pNS1 (Fig. 2B) and the level of VP2 mRNA in the pNS1-Δc63 transfected group, as measured by qPCR, was significantly decreased (Fig. 2C).
The dual-luciferase reporter assay system was used to measure the ability of pNS1-Δc63 to activate the P38 and VP2-5’UTR. In this system, firefly luciferase was translated by P38 promoter and 5’UTR (pGL3-P38 + 5’UTR) together, by P38 promoter (pGL3-P38) alone, or by 5’UTR (pGL3-5’UTR) alone, and the Renilla luciferase was translated by pRL-TK as an internal reference. F81 cells were co-transfected with reporter plasmids, pRL-TK and pNS1 or pNS1-Δc63 (Fig. 2D), and the ratio of firefly luciferase: Renilla luciferase activity was measured to assess the transcriptional activity of the P38 promoter and 5’UTR.
Results showed that the VP2-5’UTR significantly enhanced transcriptional activity of the P38 promoter (Fig. 2E), and the highest transcript levels were observed in transfection with pNS1 and pFL3-P38 + 5’UTR together whereas pNS1-Δc63 significantly reduced the transcriptional activity of pP38 + 5UTR (Fig. 2F). F81 cells were also transfected with plasmids containing either the P38 promoter or 5’UTR with pNS1 or pNS1-Δc63. Results showed that pNS1-Δc63 inhibited both P38 and 5’UTR transcriptional activities (Figs. 2G and H).
3.3 NS1-Δc63 failed to activate the transcriptional activity of P38 and VP2-5'UTR in the whole MEV genome
As described above, NS1 with a C-terminal 63 aa deletion failed to activate transcription of the P38 promoter and VP2-5’UTR in the dual-luciferase reporter assay system or the VP2 protein expression system controlled only by P38 and 5UTR. To further determine whether the NS1 C-terminal 63 amino acids is necessary for the regulation of P38 and VP2-5’UTR in the MEV whole genome, a full-length infectious clone of MEV with the VP2 5’UTR (pMEV-Δ5’UTR) or the P38 promoter (pMEV-ΔP38) deletion was used. Since the VP2-5’UTR is indispensable for the translation of VP2 in MEV, the expression of VP2 protein was abolished when the VP2-5’UTR was deleted due to transcriptional repression of the VP2 protein itself. Instead, a plasmid with the VP2 gene replaced with EGFP (pMEV-5’UTR -EGFP) in which the EGFP protein was expressed under the control of only the P38 promoter (Fig. 3A) was used.
F81 cells were co-transfected with pMEV-ΔP38 or pMEV-5’UTR -EGFP and pNS1 or pNS1-Δc63, samples were harvested at 24 h and the expression of VP2 or EGFP was determined by Western blot analysis. Results showed that the 63 aa C-terminal deletion in the NS1 protein rendered it inadequate to promote the expression of VP2 (Fig. 3C) or EGFP (Fig. 3D). EGFP was also assayed by fluorescence and found to be expressed only in the presence of the intact NS1 protein (Fig. 3B). Additionally, detection of EGFP and VP2 mRNA by qPCR showed that inhibition occurred at the transcription level (Fig. 3E and F). These data support the conclusion that transcriptional activity of P38 and VP2-5’UTR within the MEV genome is abolished when the 63 aa of NS1 C-terminal are deleted.
Collectively, these results indicate that both P38 and VP2-5’UTR are required for efficient VP2 mRNA expression in both the truncated vector and in the full length cloned MEV, and that NS1 protein is critical for transcriptional activation of the VP2 gene, while a C-terminal deletion of 63 aa is sufficient to abolish transcriptional activation activity.
3.4 NS1-Δc63 failed to rescue the virus
Reverse genetics methodologies were used to determine the impact of the 63 amino acids of NS1 C-terminal on viral rescue and replication. Starting with the full-length MEV clone, pMEV [16], a deletion of 63 aa from the NS1 C-terminal (pMEV-NS1-Δc63) was constructed (Fig. 4A). F81 cells were transfected with pMEV-NS1-Δc63, with pMEV itself used as a positive control. When cultures were harvested after 5 days, cells transfected with the pMEV controls showed typical cytopathic changes, including cell detachment and elongation.
After 3 freeze-thaws, the cell lysates (passage 0, P0) were added to fresh F81 cell cultures and incubation was continued. No lesions were observed in the pMEV-NS1-Δc63 transfected cells (Fig. 4C). When P3 cell lysates were incubated with fresh cells and expression of VP2 protein was measured by IFA (Fig. 4C) and Western blot (Fig. 4D) 24 h later, results showed that VP2 protein was expressed only when the recombinant virus was rescued successfully, i.e., via pMEV, and no VP2 expression was detected following transfection with pMEV-NS1-Δc63, and VP2 mRNA in the pMEV-NS1-Δc63 transfected cells was significantly reduced as measured by qPCR (Fig. 4E). It can therefore be concluded that viruses with the pMEV-NS1-Δc63 mutation did not yield infectious virus, and that the NS1 protein C-terminal was critical for virus rescue.
3.5 NS1-Δc63 affects the integrity of DNA sequences that interact with the NS1 protein
The above results showed that deletion of 63 aa from the C-terminal of NS1 suppressed P38 promoter and VP2-5’UTR transcriptional activity, leading to inhibition of VP2 protein expression and virus rescue. This posed the question as to whether transcriptional activity could be restored after providing normal NS1 protein. F81 cells were therefore co-transfected with pMEV-NS1-Δc63 and pNS1 or pNS1-Δc63 (Fig. 5A), and harvested after 24 h. Western blots showed that the VP2 protein was not detected in the pMEV-NS1-Δc63 transfected cells although pNS1 was co-transfected (Fig. 5B), thereby demonstrating the need for an intact NS1 C-terminal for production of infectious virus.