Influenza virus is a typical enveloped (Schulze, 1973 and Laver, 1973] negative-strand RNA virus which is composed of eight single-stranded genomic segments coding for 10 or more polypeptides (Anwar and Khan, 2006; Lamb, 1983; Lamb et al., 1983). The M1 matrix protein of influenza virus is a multifunctional factor involved in several steps of the life cycle of the virus (Helenius, 1992; Martin and Helenius, 1991). It is the most abundant protein found within the virion and has been shown to play a central role in virus assembly (Allen et al., 1980; Lamb 1983;Lamb et al., 1983;Lamb et al., 1981; Lamb et al., 1985). This protein is located at the inner surface of the lipid bilayer of the virion envelope in close proximity to the ribonucleocapsid protein (RNP) complex (Apostolov et al., 1969;Compans et al., 1975; Compans et al., 1979; Schulze 1972). As with all RNA viruses, the influenza virus lacks a proofreader for replication, allowing the virus to mutate quickly. The host immune system selects for mutants by making antibodies to the original strain of virus. This leads to antigenic drift, whereby the virus gradually changes its types or sub-type. Antigenic variation in influenza comes in a multitude of forms, enabling it to effectively evade the immune system. In humans, changes in certain genes can lead to increasing virulence. Interestingly, antigenic drift in avian influenza is at a standstill; mutant viruses contain only silent changes in amino acid sequences [http://www.brown.edu/Courses/Bio_160/ Projects1999/av/]. The most important aspect in the production of antibodies or drug is the design of the peptideantigen. The peptide-antigen is a small segment (15-18 amino acids) of the protein sequence of interest. These peptideantigens can be used for immunization in order to produce antibodies against the protein for which the peptide is formed or the peptides can be used as a basis for small-molecule/ drug targeting. There is a renewed interest to understand the antigenic diversity of influenza virus because of recent outbreaks of influenza epidemics (CDC, 2006). The knowledge thus gained will play a decisive role in influenza vaccine development (Amexis et al., 2001). We have undertaken this study to compare the antigenic proprieties of the Indian isolates of Influenza H5N1 with the homologous region of different subtypes of the Avian flu virus. The predicted antigenic sites of the matrix protein (M1) of H5N1, found in India, have been compared with the antigenic sites of the homologous domains in other subtypes, H5N1.

Antigenic Site Analysis

The sequences of Matrix protein (M1) of Influenza A Virus, spread in India during 2005-06 were obtained, which includes sub-types A/chicken/Navapur/Maharashtra/India/ 7972/2006 (H5N1) [GenBank: ABG88883] and A/chicken/ Jalgaon/India/9386/2006 (H5N1) [GenBank: ABJ96491] isolated from chicken were analysed in the present work. Sequences similar to M1 of A/chicken/Navapur/ Maharashtra/India/7972/2006 (H5N1) were extracted from NCBI FTP server ftp://ftp.ncbi.nih.gov/genomes/INFLUENZA/ from the file influenza.faa. Protein sequences of the sub-type H5N1 were sorted out for analysis. The sequences selected for the study are given in table 1. Antigenic sites were predicted using the tool Antigenic at Emboss available at http://bio.dfci.harvard.edu/Tools/EMBOSS/.

Table 1: Predicted antigenic domains in Influenza A virus of Indian origin during the outbreak of 2006 in Navapur and Jalgaon ( ABG88883 and ABJ96491)

Antigenic Site Analysis

Distinct antigenic domains identified in H5N1 strain of Indian origin (ABG88883 & ABJ96491), are reported in Table 1. On comparing the antigenic domains in Indian strain, ABG88883 with the antigenic domains of Turkey, CAJ01905, it was noticed that in Turkey, CAJ01905 there was a single domain at position 48 to 69 while in our India strain, ABG88883, the antigenic domain at this position was splitted into two domains within the same position viz., 48->55 and 60->69. We, then compared Indian strain, ABG88883 with another strain of Indian origin i.e., strains from Jalgaon (ABJ96491). It was found that Jalgaon strain also showed two antigenic domains within the same position, 48->69. To find out the reason for this discrimination of being splitting into two different antigenic domains in some of the strains, 68 different H5N1 strains were analysed for the antigenic domain of M1 protein at 48 -> 69 (Table 2). Through in silico analysis of this antigenic domain revealed that the strains that were having mutation at position 59 (M59I) have two antigenic domains (48->55 and 60->69), within the same region, while strains with single domain (48 -> 69) found to have Isoleucine at position 59. Moreover, it was also noticed that the strains from Germany, Sudan, Russia and Nigeria found to have two antigenic domains within position 48 to 69.

Table 2: Comparison of antigenic domains at position 48 to 69 within 68 different strains of Influenza virus A H5N1 for M1 protein from different geographic regions.

To understand the function of M1 protein it is necessary to study the functional properties of its distinct domains (such as antigenic domains). Antigenic sites represent potential candidates for peptide vaccine against the virus; this type of study can help in better understanding of Influenza A H5N1 isolates spread in India. In this work, we identified distinct antigenic regions in the M1 protein of Influenza A H5N1 virus and compared these domains with other H5N1 strains for M1 protein. Sequence based analyses of M1 protein revealed that although there exists very high sequence similarity at protein level, various strains have acquired certain mutations which confer strain-specific properties.

It is known that a phenotypic property such as antigenicity is the result of 'spatio-temporal' hierarchical processes. Exact mapping of the same at molecular level is difficult due to the fact that there exists complexity in terms of intermolecular interactions which may be discrete or continuous (Flower et al., 2003). Thus, a bioinformatics approach was utilized for this analysis. It was found out that the reason for splitting of the antigenic domain is the point mutation at position 59 (M59I). Our data shows that the strains which have Isoleucine at position 59 show single antigenic domain while the strains with methionine at this position are having two antigenic domains within the region 48 to 69 (Table 2). It was also noticed that the two strains isolated from Indian subcontinent are showing similarity for the presence of antigenic domain (48 -> 69).

Antigenic sites represent potential candidates for the peptide vaccine against H5N1. In silico approach has the potential to become an important tool to understand disease pathogenesis and designing antigen- specific therapies. The peptides, derived from antigenic sites may serve as ideal antigens to develop site-specific immunoassays for serological diagnosis.

Department of Biotechnology, Ministry of Science and Technology, Government of India is acknowledged for the financial support. This work was supported by DBT grant BT/PR7507/BID/07/201/2006 to AUK.