Comparative Molecular and Structural Characterisation of Chikungunya Virus Isolated Before and After 2006 Epidemics

Background: Chikungunya virus caused incapacitating infection during 2006 epidemics spanning 28 countries. The expansion of the disease and increased virulence resulted in neurological and ocular complications in the affected patients implicating changes in the structural and functional properties of the newly emerging strains of Chikungunya virus. Objective: Hence the study was designed to understand the disparity between two strains isolated before and after 2006 epidemics by both in vitro and in silico approaches with respect to E1 gene. Materials and Methods: Sequencing of E1 gene and phylogenetic analysis of the two strains were carried out followed by the determination of growth pattern. The impact of aminoacid substitutions on the structural properties of E1 protein between the strains of Chikungunya virus was identified by different Bioinformatic tools. Results: Sequencing and Phylogenetic analysis revealed the two strains as Asian (isolated before 2006) and East Central South African (isolated after 2006). ECSA strain produced 1.5 fold log10 titre increased viral production than the Asian strain at the multiplicity of infection 1. Influence of aminoacid difference on the structure of E1 protein between two strains by Bioinformatic analysis Original Research Article Sangeetha et al.; BMRJ, 17(2): 1-11, 2016; Article no.BMRJ.27880 2 had shown a change in the conformations by the loss of two intermolecular hydrogen bonds in 121 position in Asian strain and electrostatic differences in 211 positions between the two strains was also observed. Conclusion: As the interacting aminoacid 121 and 211 position lies near the fusion loop hence the difference in aminoacid position between the two strains could better influence fusogenicity and stable trimer formation. This forms a preliminary insight on the impact of aminoacid substitutions on the structural properties of E1 protein between the strains of Chikungunya virus and further experimental investigations are warranted.


INTRODUCTION
Mutation in RNA viruses occurs rapidly because of the high error frequency of viral RNA dependent RNA polymerases leading to adaptive mutation in host cells [1]. Mutation in Chikungunya virus has led to the emergence of new strains with higher degree of virulence and expansion into other countries [2]. So far Phylogenetically, Chikungunya virus has been classified into three different clades, primarily by geography into West African, Central / East African and Asian CHIKV [3,4]. But the epidemics after 2007 had led to the emergence of Indian Ocean lineage [IOL], a new lineage evolved from an existing ECSA enzootic genotype [5]. E1 protein plays a major role in the fusion of virus to the host cell receptors and it occurs through a p H -dependent endocytic pathway. Thus the amino acid and the structure of E1 protein highly influence the fusogenicity both in vector and host cells [6,7]. The change or substitution of amino acid from A226V mutation in E 1 protein had established the adaptability of Chikungunya virus in Aedes albopictus in addition to its rural vector, Aedes aegypti and had clearly demonstrated the role of aminoacid substitution in E1 protein with respect to the adaptation and invasion of different lineages in the vector [8]. E1 T98A substitution had shown increased susceptibility and adaptation to Ae. albopictus however E1 211K was not found to be associated with a significant increase in CHIKV infectivity in Ae.albopictus and had shown epistatic interaction between position E1-226 and E1-98 of CHIKV [9].
Besides the specific adaptation pattern exhibited by different lineages to specific vectors ,the strains emerged after 2006 had caused severe pathogenicity such as neurological complications [9,10,11], ocular infections [12], dermatological manifestations [13], hepatitis,severe arthropathy [14] in the infected patients than the strains evolved before 2006.
In this study, we the authors had tried to find the impact of aminoacid substitutions on the structural properties of E1 protein between the two strains of Chikungunya virus isolated before and after 2006 epidemics by identifying the disparity through electrostatic potential and interaction of amino acids in the tertiary structure.

Estimation of Replication Kinetics
Confluent monolayer of cells were infected at multiplicity of infection of 1 and incubated at 37°C for 1 hr. After 1 hr of incubation, 2% maintenance medium was added and incubated at 37 in 5% Co 2 . Virus inoculums were then removed at an interval of 24 hrs upto 48 hrs and centrifuged at 3000 rpm for 15 min to remove the supernatant and stored at -80°C for titration. Each experiment was done in triplicates.
Virus titres from cell culture supernatant were quantified. Serial 10 -fold dilutions were made of samples collected at each point and 100 µl of samples of each dilution were added in triplicates in 96 well plates containing confluent monolayer of vero cells. Viral titre was determined as TCID 50 by Reed and Meunch method.

PCR Standardization
The cDNA was then used for PCR amplification of E1 gene, using gene specific primers.
Likewise, PCR II was done using the primers E1-Seq-FP1 & E1-Gene-RP. Expected size of the amplicon was ~1137 bp and the PCR amplicon obtained were sequenced using the primers E1-Seq-FP1; E1-Seq-FP2 & E1-Gene-RP, to read the entire stretch of the gene amplified. PCR was performed for 35 cycles at initial denaturation at 94°C for 5min followed by renaturation at 55°C for 30 sec and annealing at 72°C for 1.30 sec to obtain the ~ 522 bp, and the denaturation at 94°C for 30 sec followed by renaturation at 55°C for 30 sec and annealing at 72°C for 5 min. PCR products obtained were Gel eluted [ Figs

Sequencing
The PCR products were sequenced using Big Dye Terminator version 3.1" Cycle sequencing kit in ABI sequencing machine ABI 3500 XL Genetic Analyzer in POP_7 polymer 50 cm Capillary Array following BDTv3-KB-Denovo_v 5.2 protocol.

Phylogenetic Analysis
Relationships among the aligned amino acid sequences of the Chikungunya virus were determined using Mega version 5.0. The Kimura two-parameter algorithm was applied to calculate the evolutionary distances and the neighbourjoining method was used to construct the phylogram, which was viewed using the treeview program. Bootstrap analysis was performed on 1000 replicas using the programs seqboot and consense to ascertain support for the major branches of the tree. The existing sequences from the NCBI GenBank with the accession numbers were retrieved excluding the sequences collected from mosquitoes and were analysed for the construction of phylogenetic tree to deduce the clade of the subjected query sequence.
Pairwise comparison was performed using Lalign between the protein sequence of the two tested strains.

Determination of Structural and Functional Variations
The aminoacid sequence of Asian and ECSA of CHIKV were determined and the secondary structure were predicted using Psi-Pred [http://bioinf.cs.ucl.ac.uk/psipred/ psiform.html] [15,16]. Also the coiled coil structures were determined using multicoil [http://groups.csail.mit.edu/cb/multicoil/cgibin/multicoil.cgi] by Hidden Markov model [HMM] to predict any significant mutation in the fusogenicity region, if any ,in the coiled coil region [17,18]. The 3D structure and the atomic surface arrangements were identified using the Swiss Model Server. Molecular surface model of proteins and the influence of aminoacid substitution on the structure were calculated with Deep view 3.7 [19,20].
Electrostatic potentials were calculated with partial atomic charges of the Gromos 4.3 A1 force field and with dielectric constants of 4 and 80 for the protein and solvent respectively using Poisson -Boltzmann method. The absolute and relative solvent accessibility of amino acids with substitutions were predicted using NetSurfP [http://www.cbs.dtu.dk/services/NetSurfP/] [21].

Growth Kinetics
Both the strains showed similar characteristic programmed cell death features such as membrane rupture, nuclear condensation, cell fragmentation resulting in total monolayer destruction of Vero cells within 36 hrs of viral inoculation. Although both produced a similar pattern of CPE, the isolates showed a slight difference in their replication kinetics. Strain M 60613 produced 1.5 fold log 10 titre increased viral production than the strain M 4944 at the multiplicity of infection 1 after 36 hrs.

Structure Cum Function Analysis
E1 fusion peptide loop comprises the region from E 1 73 to E 1 90 [-VVPFMWGGAYCFCDHENT-] and there were no difference in aminoacid between the Asian and ECSA strains as revealed by sequence analysis. The aminoacid present near the fusion loop showed substitutions at S72N and T98 A positions and thus their electrostatic influence on the fusion loop were analysed.

Fig. 2. Comparative analysis performed between Asian and East Central South African strains by LAlign
Calculations of the Electrostatic potential were performed with Swiss-PDB viewer. The orientation of structures were identical in all panels and the electrostatic calculations were performed in an identical manner. The electrostatic potential calculated was identical /same irrespective of the amino acid substitution observed between Asian and ECSA strains at positions 72, 98, 121, 142, 145, 215, 225, 315 and 322 positions and had not produced any significant change in electrostatic potential except at the position 211 [Figs. 3.1 to 3.6]. The differences in the net solvent accesability of the exposed and buried aminoacids between the Asian and ECSA strain are shown in Table 1.  Fig. 3

Effects of A to S Mutation and E to K Mutation in Structure Formation of Asian and ECSA Strain
Further analysis on the conformational changes of the specific mutations had shown that all the  A slight Variation in the secondary structure between the Asian and ECSA were observed. The number of random coils observed were the same in both the strains however variations in beta turn ,alpha helix and extended beta strand were detected as analysed by SOPMA. But the superimposition of the two different structures did not reveal any differences between them.
Both the strains used in our study were wild type E226A and had not the A226V mutation, a factor involved in the Cholesterol dependence of vectors but had A226G mutation signifying that the two isolates were virulent. Earlier studies on Chikungunya viruses had shown that S, T,G,P aminoacids in 226 position in the E1 protein were almost indistinguishable from parental wild type viruses in their virulence properties but the strains showing I,F,M or H and L had shown slight attenuation or complete attenuation respectively [22]. The most conserved aminoacids G91 and H230 were present in the two isolates that are found to be playing a prime role in fusogenicity [23].
We hypothesised that an increased 1.5 fold magnitude growth of ECSA than Asian strain in Vero cells could be due to mutations observed in the membrane fusion protein and thus tried to explore the influence of amino acid substitution on the structure and functional characteristics of E1 protein by Molecular Modelling. In order to further understand the difference in the monomer interaction, the substituted aminoacids were analysed by computing Hydrogen bond formation and the distances with other monomers were analysed using Swiss PDB viewer v.4.01. The exposed aminoacids that had undergone substitution did not show any loss in the intermolecular bond formation with the adjacent monomers and were stable, however the buried aminoacid substitution A to S had shown significant difference in the intermolecular hydrogen bond formation with the monomer present at its backbone.
During fusion, E1 inserts into the target membrane via the fusion loop forms a core trimer composed of domains I & II and refolds to a hairpin like confirmation in which DIII and the stem [DIII] pack against the central core trimer. This refolding reaction moves the TM domain and the fusion loop to the same side of the trimer, bringing the viral and target membranes together and driving membrane fusion [11,18]. Thus the low PI of Serine [5.8] in comparison to alanine [6.02] at 121 position of ECSA along with the slight higher electrostatic potential exhibited by lysine at 211 position on the surface structure of neighbouring amino acids could better influence the fusogenicity and the formation of stable trimer in ECSA than the Asian strain .The study also highlights that the mutation were higher and notable only in the extracellular topological domain 1 of E1 protein of CHIKV and the other domains [II and III] ,fusion loop were highly conserved. Thus the aminoacid substitutions observed especially at 121 and 211 positions could highly influence the fusogenicity and interaction with the host receptors.
E1 protein plays major role in pathogenicity by remarkable cell tropism and mediating entry into susceptible cells. The aminoacid substitution in Japanese encephalitis virus at 138[Glutamic acid to lysine] and 123 [methionine to lysine] [24,25] and 335 [Glutamic acid to Lysine ]in the HN glycoprotein of mumps virus had highly attributed towards neurotropism [26] likewise the specific mutations at E211K position from glutamic acid to lysine in ECSA strain could also feature for higher virulence as the strains emerged after 2006 epidemics were, in particular, found to exhibit neurological complications. Thus further molecular insights and in vivo experiments are also warranted to support our findings.

CONCLUSION
As the significant conformational change occur at the aminoacid positions 121 and 211 lies near the fusion loop, the difference in aminoacid position between the two strains could influence their fusogenicity and stable trimer formation. This impact had provided a new insight into the influence of the aminoacid substitutions pertaining to the three dimensional structural of E1 protein between the strains of Chikungunya virus.