Surprising diversity found in H5N1 in Asian chickens

Robert Roos News Editor

Apr 14, 2009 (CIDRAP News) – H5N1 influenza viruses found in chickens seized at Vietnam's border with China in 2008 show a surprising level of genetic diversity, suggesting that the viruses are evolving rapidly and raising questions about disease surveillance and the effectiveness of prepandemic H5N1 vaccines, according to a team of Vietnamese and US scientists.

The findings suggest that the subgroup of viruses identified in the study has been circulating in the region either undetected or unreported for a "considerable amount of time," the report says. The viruses are in clade 7, a group that has not been found in an outbreak since one in China's Shanxi province in 2006.

Also, given the degree of difference between these clade 7 viruses and those used to make human H5N1 vaccines, "it is unclear whether or not humoral antibodies elicited by current vaccine candidates will have cross-neutralization activity against the new viruses," says the report, published recently in Virology.

The study was conducted by scientists from Vietnam's National Center for Veterinary Diagnostics in Hanoi and the US Centers for Disease Control and Prevention (CDC), with the CDC's Ruben O. Donis as the senior author.

Vietnam bans the importation of poultry from H5N1-infected countries. The viruses analyzed in the study came from imported illegal chickens seized at border control stations in Vietnam's Lang Son province, on the Chinese border. Authorities confiscated 495 chickens at the border stations in the first 5 months of 2008 and tested them for avian flu. Fifteen birds were found infected with H5N1 virus, and two others had an H9N2 virus.

The authors extracted RNA from the 15 H5N1 isolates, amplified it, and analyzed the sequence of the hemagglutinin (HA) genes. They found that all 15 HA genes fit within a subgroup of previously identified clade 7 viruses. They determined that 13 of the 15 HA genes fell into two further subgroups, which they labeled A and B.

The isolates within groups A and B were very closely related, but the groups differed considerably from each other, with an average divergence of 4.05% in HA nucleotide sequences and 5.69% in amino acid sequences, the report says. The authors also determined that the isolates differed by an average of 3.67% and 5.81% from the most closely related previous clade 7 isolate, which came from a chicken in China's Shanxi province in 2006.

In comparing the viruses with representatives of other H5N1 clades, the investigators found they were most closely related a clade 4 virus (a 2006 isolate from Guiyang province of China) and most distantly related to a clade 1 virus, a 2004 isolate from Vietnam. The respective differences in amino acid sequences were 8.20% and 9.07%.

The genetic distance between the HAs of these clade 7 viruses and the isolates used to make prepandemic vaccines raises doubt as to whether antibodies generated by the vaccines would react with the clade 7 viruses, the authors write. They say the viruses used in the vaccines represent clade 1 (from Vietnam, 2004), clade 2.1.3 (Indonesia, 2005), clade 2.2 (Qinghai province, China, 2005), and clade 2.3.4 (Anhui province, China, 2005).

The authors used ferret serum samples to test whether antibodies generated against five of the clade 7 viruses would cross-react with viruses from other clades and vice versa. Using hemagglutination inhibition assays, they found that samples (antisera) with antibodies against clades 1 and 2.3.4 did not cross-react with a clade 7 virus, and antisera against that same clade 7 virus did not react with clade 1 and clade 2.3.4 viruses. Also, antisera to the clade 7 virus showed limited cross-reactivity with each of four other clade 7 isolates, suggesting "some, albeit minimal" antigenic similarity among the viruses.

The authors write that the identification of two distinct subgroups (groups A and B) of clade 7 viruses "may suggest that these viruses originated from two distinct regions or host populations isolated from each other."

They say the isolates they analyzed were remarkably diverse, given that other studies of H5N1 viruses in a single clade have shown a nucleotide divergence of no more than 1.5%. They also observe, "The high degree of genetic divergence from previously identified clade 7 viruses also indicates that this subgroup of viruses has likely been circulating undetected or unreported for a considerable amount of time as poultry outbreaks caused by clade 7 H5N1 viruses were last reported in Shanxi province, China in 2006."

The findings point up the need for border controls and strong virologic surveillance to detect and prevent the spread of "exotic" H5N1 viruses from country to country, the authors conclude.

Dr. Les Sims of Australia's Asia-Pacific Veterinary Information Services called the study "an important piece of work" that offers another sign of gaps in avian flu surveillance systems in Asia.

"Unexplained gaps in the evolution of H5N1 virus HA genes suggest that surveillance systems have not been able to detect all cases of infection," Sims, who is a consultant to the United National Food and Agriculture Organization, told CIDRAP News by e-mail.

Other signs of holes in surveillance efforts include the detection of H5N1 virus in humans before finding it in nearby poultry, finding the virus in intermittent testing in poultry markets, and infected dead poultry washing up on Hong Kong's beaches, Sims said.

Clade 7 viruses have been found in various places in China, including Shanxi, Yunnan, and Hebei provinces, he reported, adding, "One of the recent human cases in Jiangsu was apparently due to a clade 7 virus, based on newspaper reports."

"The antigenic variation is not surprising given that clade 7 viruses have been shown to be antigenic variants previously," Sims said. But the findings from the ferret sera should be tested further with challenge trials in poultry vaccinated with existing vaccines, he added.

The difference between a 2006 clade 7 virus found in China—cited in the study as the closest previously identified relative of the viruses the authors analyzed—and existing vaccine antigens "led to the incorporation of a new antigen into some Chinese poultry vaccines," he said. "These vaccines containing a clade 7 antigen have been used extensively in China." He added that there have been no reports of vaccine failures in Vietnam resulting from antigenic variants.

Sims said the antigenic changes in the clade 7 virus might potentially have been spurred by poultry vaccination efforts. "We need to balance measures to reduce the amount of circulating virus, which vaccination of poultry can do, against the potential 'harm' that might occur if antigenic variants emerge," he stated. "We should expect antigenic variants to emerge and ensure that appropriate measures are in place to detect these at an early stage."

Nguyen T, Davis CT, Stembridge W, et al. Characterization of a highly pathogenic avian influenza H5N1 virus sublineage in poultry seized at ports of entry into Vietnam. Virology 2009 (early online publication) [Abstract]

hat-tip Sally