Characterization of the GXXXG motif in the first transmembrane segment of Japanese encephalitis virus precursor membrane (prM) protein

Abstract

The interaction between prM and E proteins in flavivirus-infected cells is a major driving force for the assembly of flavivirus particles. We used site-directed mutagenesis to study the potential role of the transmembrane domains of the prM proteins of Japanese encephalitis virus (JEV) in prM-E heterodimerization as well as subviral particle formation. Alanine insertion scanning mutagenesis within the GXXXG motif in the first transmembrane segment of JEV prM protein affected the prM-E heterodimerization; its specificity was confirmed by replacing the two glycines of the GXXXG motif with alanine, leucine and valine. The GXXXG motif was found to be conserved in the JEV serocomplex viruses but not other flavivirus groups. These mutants with alanine inserted in the two prM transmembrane segments all impaired subviral particle formation in cell cultures. The prM transmembrane domains of JEV may play importation roles in prM-E heterodimerization and viral particle assembly.

Figure 1

Schematic dagram of flavivirus prM and E protein where the stem and anchor regions located at their C-terminal ends.

Figure 2

Positions of the alanine insertion mutations in the two transmembrane regions (TM1 and TM2) of prM protein of JEV CH2195LA strain. A series of mutants was constructed; Bac-prM, Bac-prM A139, Bac-prM A143, Bac-prM A147, Bac-prM A157, Bac-prM A161 and Bac-prM A165. Arrows indicate the positions of alanine insertion.

Figure 3

(A) The total expression of prM and E proteins in Sf9 cells co-infected with Bac-prM (wild type and mutants A139, A143, A147, A157, A161, A165) and Bac-E. GAPDH was measured as an internal control; (B) The mRNA transcript levels of prM and E in the co-infected Sf9 cells measured by real-time RT-qPCR. The wt prM/E expression level was taken as 100%.

Figure 4

Treatment of recombinant prM with endoglycosidase PNGaseF and EndoH. The prM mutants with inserted alanine were treated with PNGaseF (P) and EndoH (E) to analyze the presence and composition of their N-linked glycans.

Figure 5

Observation of subcellular localization using confocal laser scanning microscopy. The subcellular location of wild type and mutant prM was visualized by double-staining of the infected Sf9 cells. The ER compartment was stained with 10 nM DiIC13(3), and the prM proteins were stained with mAb 5B1, followed by goat anti-mouse Alexa Fluor 647 IgG. Photographs were taken with appropriate excitation laser wavelengths and merged to reveal the co-localization of prM and ER compartment.

Figure 6

Sucrose gradient sedimentation analysis of alanine-insertion mutagenesis of the anchor region of prM protein. Sf9 cells were coinfected with Bac-E and Bac-prM mutants. Cell lysates labeled with [³⁵S] were applied for centrifugation in a 3 to 60% (wt/wt) sucrose gradient. Each fraction was immunoprecipitated with monoclonal antibody E3.3. The immune complexes were analyzed by reducing SDS-PAGE and fluorography. (A) Sf9 cells coinfected with Bac-prM and Bac-E; (B) Sf9 cells coinfected with prM A139 and Bac-E; (C) Sf9 cells coinfected with prM A143 and Bac-E; (D) Sf9 cells coinfected with prM A147 and Bac-E; (E) Sf9 cells coinfected with prM A157 and Bac-E; (E) Sf9 cells coinfected with prM A157 and Bac-E; (F) Sf9 cells coinfected with prM A161 and Bac-E; (G) Sf9 cells coinfected with prM A165 and Bac-E. The data presented in this figure are three independent experiments.

Figure 7

prM-E heterodimerization of prM mutants with inserted alanine as a percentage of the binding intensity of wild type prM. Each prM to E ratio was calculated. The wild type prM/E ratio was taken as 100% in evaluating the mutant prM-E binding affinity. Each data represents the average of the experiments for a single prM mutant and the error is standard deviation. The positions of the alanine insertions in the TM region of prM are indicated.

Figure 8

Alanine-insertion mutagenesis of the TM regions of prM protein: effect on SP formation. Recombinant SPs secreted from Sf9 cells co-infected with with Bac-E and (A) Bac-prM; (B) Bac-prM A139; (C) Bac-prM A143; (D) Bac-prM A147; (E) Bac-prM A157; (F) Bac-prM A161, and (G) Bac-prM A165 were quantified by ELISA. The size and shape of the secreted recombinant SP were observed by TEM using immuno-gold labeling with (H) mAb E3.3 and (I) and mAb 5B1. (J) The total E protein in Sf9 cells was determined in the presence of Brefeldin A, a fungal a fungal macrolide antibiotic which blocks E protein secretion. GAPDH was taken as an internal loading control.

Figure 9

Quantification of the SP release was determined by calculating the underneath area of the E protein concentrations from fractions 7 to 10 in Fig. 7 A-G. The underneath areas were normalized by the corresponding total amount of E protein in Fig 7J and the percentage of SP release represented the relative value of each prM mutant (Fig. 7B-G) compared to the wild type (Fig. 7A)

Figure 10

Characterization of the GXXXG motif of the TM1 region of the prM protein by glycine-substitution mutagenesis. Glycine residues at 142 and 146 were substituted by alanine, leucine, or valine to investigate the relationship between the GXXXG motif and the prM-E binding affinity.

Figure 11

(A) The total expression of prM and E proteins in Sf9 cells co-infected with Bac-prM (wild type and mutants G142A, G142L, G142V, G146A, G146L, G146V) and Bac-E. GAPDH was measured as an internal control; (B) The mRNA transcript levels of prM and E in the co-infected Sf9 cells measured by real-time RT-qPCR. The wt prM/E expression level was taken as 100%.

Figure 12

The percentage of the binding intensity of wild type prM. Each prM to E ratio was calculated where wild type prM/E ratio was taken as 100% in evaluating the glycine-substituted mutant prM-E binding affinity. Each data represents the average of the experiments for a single prM mutant and the error is standard deviation.