Continued evolution of H5N1 influenza viruses in wild birds, domestic poultry, and humans in China from 2004 to 2009

Abstract

Despite substantial efforts to control H5N1 avian influenza viruses (AIVs), the viruses have continued to evolve and cause disease outbreaks in poultry and infections in humans. In this report, we analyzed 51 representative H5N1 AIVs isolated from domestic poultry, wild birds, and humans in China during 2004 to 2009, and 21 genotypes were detected based on whole-genome sequences. Twelve genotypes of AIVs in southern China bear similar H5 hemagglutinin (HA) genes (clade 2.3). These AIVs did not display antigenic drift and could be completely protected against by the A/goose/Guangdong/1/96 (GS/GD/1/96)-based oil-adjuvanted killed vaccine and recombinant Newcastle disease virus vaccine, which have been used in China. In addition, antigenically drifted H5N1 viruses, represented by A/chicken/Shanxi/2/06 (CK/SX/2/06), were detected in chickens from several provinces in northern China. The CK/SX/2/06-like viruses are reassortants with newly emerged HA, NA, and PB1 genes that could not be protected against by the GS/GD/1/96-based vaccines. These viruses also reacted poorly with antisera generated from clade 2.2 and 2.3 viruses. The majority of the viruses isolated from southern China were lethal in mice and ducks, while the CK/SX/2/06-like viruses caused mild disease in mice and could not replicate in ducks. Our results demonstrate that the H5N1 AIVs circulating in nature have complex biological characteristics and pose a continued challenge for disease control and pandemic preparedness.

FIG. 1.

Phylogenetic analyses of the H5N1 viruses isolated from 2004 to 2009 in China. The phylogenetic trees were generated with the PHYLIP program of the CLUSTALX software package (version 1.81). The five trees were generated based on the following sequences: HA nucleotides (nt) 29 to 1732, NA nt 21 to 1730, PB2 nt 28 to 2307, PB1 nt 25 to 2298, and PA nt 25 to 2175. The phylogenetic tree of HA was rooted to A/mallard/Denmark/64650/03 (H5N7), the NA phylogenetic tree was rooted to A/green-winged teal/Ohio/72/99 (H1N1), and the PB2, PB1, and PA phylogenetic trees were rooted to A/Memphis/1/90 (H3N2). The colors of the viruses in the NA, PB2, PB1, and PA trees match with those used in the HA tree. Abbreviations: BHG, bar-headed goose; CK, chicken; DK, duck; GS, goose; GC, great cormorant; SK, shrike; AH, Anhui; FJ, Fujian; GD, Guangdong; GX, Guangxi; HB, Hubei; HN, Hunan; HeB, Hebei; HeN, Henan; JX, Jiangxi; LN, Liaoning; NX, Ningxia; QH, Qinghai; SD, Shandong; SX, Shanxi; TB, Tibet; XJ, Xinjiang. Sequences labeled with a red “#” in the HA tree were downloaded from the available databases. Viruses labeled with a red arrow were selected for antiserum generation.

FIG. 2.

Genotypic evolution of the H5N1 viruses isolated in China from 2004 to 2009. The eight gene segments are indicated at the top of each bar. The number in each bar shows the group of genes indicated in Fig. 1 and Fig. S1 in the supplemental material.

FIG. 3.

Replication and virulence of H5N1 influenza viruses in mice. Virus replication was tested as described in Materials and Methods. The data shown are the mean titers for three mice. A value of 0.5 was assigned if the virus was not detected from the undiluted sample. The MLD₅₀ is shown as the log₁₀ EID₅₀. Genotypes were determined on the basis of the diversity of the gene nucleotide sequences, as described in the legends of Fig. 1 and 2. The red dashed line indicates the lower limit of detection.

FIG. 4.

Replication and virulence of H5N1 influenza viruses in ducks. (A) Viral titers and shedding in ducks on day 3 p.c. Data shown are means ± standard deviation. For statistical purposes, a value of 0.9 was assigned if virus was not detected from the undiluted sample. *, P < 0.01 compared with the titers in the corresponding samples in the DK/GD/23/04, DK/HB/43/05, and BHG/QH/1/06 virus-inoculated ducks. (B) Death pattern for ducks inoculated with different H5N1 influenza viruses. The dashed line indicates the lower limit of detection.