In this study, we performed a detailed genetic analysis of HN glycoprotein gene in HPIV1 isolates from patients with ARI during 2002-2009 in Yamagata prefecture, Japan. The phylogenetic tree constructed by the NJ method showed that the present HPIV1 isolates were divisible into two major genetic clusters (Figure 1). The other tree constructed by the ML method showed that the year of the first major division was estimated at 1950, and the ancestral strains further subdivided at around 1987, resulting in three clusters (one minor and two major, Figure 2). The strains belonging to the two major clusters subdivided into many clusters after 2000. The present HPIV1 isolates showed an overall high level of nucleotide sequence identity (92.6-100%) of the HN coding region. Pairwise distance values based on the nucleotide sequences among the present strains were relatively low (less than 0.06). In addition, there were no positively selected sites found. These results suggest that several lineages of highly conserved HN gene in HPIV1 were prevalent in Yamagata prefecture. The present strains could not be provisionally type assigned from the pairwise distance values. Thus, the accumulation of large amounts of data may be needed to genotype HPIV1 based on pairwise distance.
Homology and phylogenetic analysis by the NJ method is frequently used in molecular epidemiological studies of various viruses. Homology analysis based on nucleotide sequences mainly shows the similarities of the analyzed genes among the strains. Phylogenetic analysis by the NJ method can give an estimation of the viral evolution rate and cluster classification. Furthermore, the ML method can enable analysis of the time scale of the evolution of viral genes. In the present study, we were able to estimate the viral evolution rate of HPIV1, cluster classification, and the evolutionary time scale of the present isolates by applying the NJ and ML methods to the detailed phylogenetic analysis of the HN coding region in HPIV1. There is currently little information available regarding the molecular evolution of the HN coding region in HPIV (HPIV1 to 4). Furthermore, the rate of molecular evolution is very low (7.68 × 10-4 substitutions/site/year) in the present strains. Previous reports suggest that the rate of another gene of respiratory viruses belonging to Paramyxoviridae, such as respiratory syncytial (RS) virus, is higher (1.8 × 10-3 substitutions/site/year) than that of the present data . The reason for the difference is unknown at present. However, it is possible that genome properties other than size, such as polarity or structure, may be associated with substitutions of the viral genome . Further studies on the detailed mechanisms of viral genome substitution may be needed. Additional sequence data and further structural analysis are required to demonstrate the mechanisms of the molecular evolution of HPIV1.
HPIV is classified into four serotypes (HPIV1 to 4), all of which can cause various ARI in humans, such as URI, croup, bronchitis, and pneumonia . Previous reports suggest that HPIV1 and 3 are the dominant viruses in children with ARI . In addition, HPIV is a major causative agent of virus-induced asthma . Thus, HPIV1 is a major agent of ARI, along with other viruses, such as adenovirus, RS virus, human metapneumovirus, and rhinovirus . However, the molecular epidemiology of HPIVs is poorly understood, and only a few reports on the molecular epidemiology of HPIV1 are available. For example, Henrickson and Savatski analyzed the longitudinal evolution of the HN coding region in 13 strains of HPIV1 isolated in the United States . The results showed that the antigenic and genetic subgroups are very stable. Another report suggested that two distinct genotypes of HPIV were detected during the 1991 Milwaukee epidemic . In the present study, we used HPIV1 isolates from patients with ARI and studied the evolution of HN protein, based on phylogenetic analyses using both the ML and NJ methods and the rate of the substitutions of nucleotides. The results showed that HN protein is highly conserved. In addition, no positively selected sites were detected. To our best knowledge, this is the first report of these findings in HPIV1.
The distribution of amino acids affects the structure of the HN coding region in HPIV1, and previous reports show that substitutions of amino acids in HN glycoprotein reveal second receptor binding sites [33, 34]. For example, substitutions at Asn173 and Asn523 are critical for the formation of a second binding site. In particular, these substitutions affect, for example, the inhibitor in hemagglutination inhibition (HI) assays and infection of culture cells. However, the second receptor binding site did not significantly affect the growth or fusion activity of HPIV1. Substitutions at N523S were found in seven of the present strains, but there were no substitutions at Asn173. Thus, we thought that N523S may not be significantly associated with infectivity or pathogenicity.
Furthermore, we then examined selective pressure by counting and the ML method. Analysis of selection pressure in the present strains showed that dS substitutions predominated over dN substitutions, and no positively selected sites (substitution) were found in HN protein in the present HPIV1 strains. The evolution of the present strains may be largely driven by purifying selection. Compared with HPIV3, little is known about the detailed biological properties of HN glycoprotein in HPIV1. As an essential molecule of these viruses, further analysis of the biological properties of HN glycoprotein in HPIV1 is required .
Although detailed data of the antigenic and catalytic sites of HN molecules in HPIV3 is relatively clear , such information regarding HPIV1 is not yet known. Moreover, the epidemiology and molecular epidemiology of HPIV is not exactly known. Thus, further and larger epidemiological/molecular epidemiological studies are required to give better understanding of the etiology of HPIVs, including HPIV1.