Plaque formation on stable cell lines

The selected stable cell lines NS2A-EGFP, NS2B-EGFP(pro), NS4B-EGFP(pro), EGFP, and pcDNA3 were infected with DV2 virus PL046 strain. Seven days post-infection, the cells were fixed and stained with crystal violet. The number of plaques of BHK-21 was defined as 100% and the relative percentage of plaque numbers of the stable cell lines were recorded in Table 5 and Fig. 3.18. The data were analyzed by ANOVA and Scheffe test. For NS2A-EGFP, there was a 44% reduction (down to 81.46% from that of control EGFP, 126.32%) with significance (P<0.05). Whereas, the differences of the percentage of plaque numbers between NS2B-EGFP(pro), NS4B-EGFP(pro) and controls, pcDNA3, and EGFP, were not significant.

(II) Discussion

3.8 Construction of pNS2A-EGFP, pNS2B-EGFP, pNS4A-EGFP, and pcDNA3-D24B-EGFP expression plasmids

Previously, researchers in the laboratory have cloned the four nonstructural genes of DV2 PL046 strain into the expression vector pcDNA3 along with C-terminal HA and His tags (徐 婕 琳, 2003, 交 大 碩 士 論 文 ; 楊 馥 嘉 , 2006, 交 大 碩 士 論 文 ). They were named pNS2A-HAHis, pNS2B-HAHis, pNS4A-HAHis, and pcDNA3-D24B-HAHis (Appendix 2).

Therefore, it was convenient to investigate the expression and to select the stable cell lines of these four nonstructural genes by the C-terminal EGFP tag. I replaced the HA-His tags with the EGFP tag and obtained the expression clones of the four nonstructural genes on the pcDNA3 backbone, named pNS2A-EGFP, pNS2B-EGFP, pNS4A-EGFP, and pcDNA3-D24B-EGFP. These four constructs were assessed by restriction digestions (Fig. 3.3) and confirmed by sequencing.

3.9 Expression of pNS2A-EGFP, pNS2B-EGFP, pNS4A-EGFP, and pcDNA3-D24B-EGFP in mammalian cells, BHK-21

3.9.1 Confocal fluorescence microscopy analysis

The EGFP gene has been optimized for brighter fluorescence and higher expression in mammalian cells (excitation maximum= 488 nm; emission maximum=507 nm). To confirm the expression of pNS2A-EGFP, pNS2B-EGFP, pNS4A-EGFP, and pcDNA3-D24B-EGFP, these four expression plasmids and pEGFP-N2 were transfected into BHK-21. All have the green fluorescence expression through the C-terminal EGFP tag in BHK-21 (Fig. 3.4 A, D, G, J, and M). Additionally, the four nonstructural proteins were colocalized against the ER, using calnexin as the marker. 2A-EGFP, 2B-EGFP, 4A-EGFP, and 4B-EGFP proteins co-localized with the ER markers in the cytoplasm of BHK-21, compared to EGFP proteins displayed the

whole cells (Fig. 3.4 A). It suggested that the constructs of the four nonstructural genes could express in mammalian cells.

3.9.2 Transient expression of pNS2A-EGFP, pNS2B-EGFP, pNS4A-EGFP, and pcDNA3-D24B-EGFP in BHK-21

The predicted molecular weights of NS2A-EGFP, NS2B-EGFP, NS4A-EGFP, and NS4B-EGFP are 50.9, 41.2, 43.7, and 54.0 kDa, respectively. The nonstructural proteins NS2B and NS4B could be detected both in the supernatants of the cells with the expected size by Western blot (Fig. 3.5 B c, f). For NS4A, even though there was a band at about 43 kDa, the dominant band was between 34 and 26 kDa (Fig. 3.5 B c). For NS2A, the dominant band was between 34 and 26 kDa, and there was a band at about 43 kDa which was smaller than expected. This was perhaps due to the protein charge affecting mobility or a partial cleavage at the N-terminal of NS2A. All in all, the transient expression of nonstructural proteins NS2B and NS4B in mammalian cells were confirmed by Western blot clearly, but NS2A and NS4A were not. Nevertheless, I proceeded to select the stable cell lines expressing these four proteins.

3.10 Selection of stable transfected cells of pNS2A-EGFP, pNS2B-EGFP, pNS4A-EGFP, pcDNA3-D24B-EGFP, and pEGFP-N2

3.10.1 Western blot analysis of the selected stable cell lines

pNS2A-EGFP, pNS2B-EGFP, pNS4A-EGFP, pcDNA3-D24B-EGFP, and pEGFP-N2 were transfected into BHK-21 for stable clone selection by G418. The selected cells were analyzed by Western blot analysis with anti-GFP-HRP antibody against the C-terminal EGFP tag and were observed by fluorescence microscope (Fig. 3.6. b; 3.7. b; 3.8. b; 3.9. b; 3.10. b).

The predicted molecular weights of NS2A-EGFP, NS2B-EGFP, NS4A-EGFP, NS4B-EGFP, and EGFP are 50.9, 41.2, 43.7, 54.0, and 27 kDa, respectively. The EGFP were detected in the

supernatants and pellets of the cells with the expected size by Western blot (Fig. 3.6 c, d). The 2A-EGFP was detected at about 43 kDa in the supernatants and pellets which was smaller than the expected 50.9 kDa (Fig 3.7. c). It was similar with the transient expression of NS2A-EGFP (Fig. 3.5 B. a). The smaller size may be due to the protein charge affecting mobility or a partial cleavage at the N-terminal of NS2A. For NS2B-EGFP, NS4A-EGFP, and NS4B-EGFP, no expected bands were detected but a band at about 27 kDa, which was the same as the molecular weight of EGFP (Fig. 3.8. c, 3.9. c, 3.10. c).

3.10.2 Materials and RNA expression confirmation of the selected stable cell lines of NS2A-EGFP, NS2B-EGFP, NS4A-EGFP, and NS4B-EGFP

Because of the incorrect bands in Western blot of NS2A-EGFP (smaller than expected), NS2B-EGFP, NS4A-EGFP, and NS4B-EGFP, first I tried to examine the materials by PCR amplification of the genomic DNA extracted from the stable cell lines. The PCR amplification of the genomic DNA extracted from the stable cell lines of NS2A-EGFP, NS2B-EGFP, NS4A-EGFP, and NS4B-EGFP were performed with forward primers 2AE-F, 2BE-F, 4AE-F, and 4BE-F and reverse primers 2AE-R located at the C-terminal of EGFP tag. For NS2A-EGFP, NS2B-EGFP, and NS4A-EGFP, expected bands were obtained by PCR amplification, but no band was detected with NS4B-EGFP. It showed that the DNA in the selected stable cell lines of NS2A-EGFP, NS2B-EGFP, and BS4A-EGFP were correct while NS4B-EGFP was not.

Second, I isolated the total RNA of the selected stable cell lines NS2A-EGFP, NS2B-EGFP, NS4A-EGFP, and NS4B-EGFP by using RNeasy Mini Kit. The extracted RNAs were subjected to superscript one-step RT-PCR (Invitrogen) with individual primers to amplify the full length RNA of NS2A-EGFP, NS2B-EGFP, NS4A-EGFP, and NS4B-EGFP, separately. According to the results in Fig. 3.12, RNAs of NS2A-EGFP, NS2B-EGFP, and NS4A-EGFP were detected in the stable lines but not NS4B-EGFP. Therefore, genomic DNA

and RNA expression of NS2A-EGFP, NS2B-EGFP, and NS4A-EGFP were verified by PCR and RT-PCR except for NS4B-EGFP. Thus, I presumed that the transient and long-term expressions of NS2A-EGFP were achieved but the smaller size in Western blot perhaps due to the protein charge. Nevertheless, the selection of stable cell lines of NS2B-EGFP and NS4A-EGFP could not be achieved.

There were two possible reasons: (1) The C-terminal EGFP tag of NS2B-EGFP and NS4A-EGFP clones had its own start codon, ATG. Overexpression of NS2B and NS4A proteins is toxic to cells, therefore, the translation starts at the ATG of the EGFP sequence. (2) The sequence of NS2B and NS4A may contain internal ribosome entry sites (IRESs). Many pathogenic viruses, for example HCV and piconavirus, use an alternative, cap-independent mechanism that substitutes RNA structure for the cap and proteins involved in translation initiation. The RNA structure driving this process is called internal ribosome entry sites (IRESs) (Jeffrey et al., 2008; Baird et al., 2006).

Therefore, I decided to solve the problem by deleting the ATG start codon at the N-terminal of the EGFP tag of pNS2B-EGFP, pNS4A-EGFP, and pcDNA3-D24B-EGFP.

Thus, I re-cloned 2B-EGFP, 4A-EGFP, and 4B-EGFP constructs as described in the next section.

3.11 Construction of pNS2B-EGFP(pro), pNS4A-EGFP(pro), and pcDNA3-D24B-EGFP(pro) expression plasmids

The possible reason why the stable cell line selection of NS2B-EGFP, NS4A-EGFP, and NS4B-EGFP could not be carried out maybe because cells selectively expressed the C-terminal EGFP gene, which had its own start codon. Thus I obtained the EGFP fragment without the start codon by PCR amplification. This fragment is named EGFP’. The EGFP fragments of pNS2B-EGFP, pNS4A-EGFP, and pcDNA3-D24B-EGFP were then replaced by the EGFP’ fragments. The constructs were named pNS2B-EGFP(pro), pNS4A-EGFP(pro),

and pcDNA3-D24B-EGFP(pro) (Fig. 3.14 A, B, C). All constructs were assessed by BsrGI and NdeI to expected results (Fig. 3.14 D). The constructs were further confirmed by sequencing to ensure that the ATG at the EGFP tag of pNS2B-EGFP(pro), pNS4A-EGFP(pro), and pcDNA3-D24B-EGFP(pro) were actually deleted (Fig. 3.15).

3.12 Expression of pNS2B-EGFP(pro), pNS4A-EGFP(pro), and pcDNA3-D24B-EGFP(pro) in mammalian cells, BHK-21

The predicted molecular weights of NS2B-EGFP(pro), NS4A-EGFP(pro), and NS4B-EGFP(pro) are 41.1, 43.6, and 53.9 kDa, respectively. The nonstructural proteins NS2B and NS4B could be detected both in the supernatants of the cells with the expected size by Western blot (Fig. 3.16 B a, d). For NS4A, the dominant band was between 34 and 26 kDa (Fig. 3.16 B c), even though the expected band was detected at about 43 kDa (Fig. 3.16 B b).

All in all, the transient expression of nonstructural proteins NS2B-EGFP(pro) and NS4B-EGFP(pro) in mammalian cells were confirmed by Western blot but NS4A-EGFP(pro) was not. The sequence of NS4A may contain internal ribosome entry sites (IRESs) or a partial cleavage at the end of NS4A (Preugschat et al., 1991). Nevertheless, these three re-cloned constructs were proceeded to the stable cell line selection.

3.13 Selection of stable transfected cells of pNS2B-EGFP(pro),