Sunday, February 18, 2024

Transboundary & Emerg. Dis.: Novel Human-Avian Reassortment H9N2 Virus in Guangdong Province, China


For a ubiquitous, low pathogenic, and relatively innocuous avian flu virus - one that rarely kills poultry and generally only mildly sickens humans (see FluTrackers case list) - we spend an awful lot of time looking at, and trying to understand, the pandemic potential of avian H9N2. 

What H9N2 currently lacks in lethality it makes up for with its remarkably wide host range, and its ability to reassort with an array of other avian, swine, human, and canine flu viruses. 
Over the years we've seen dozens of studies showing H9N2's genetic contribution to other novel influenza subtypes, including H5N1, H5N6, H7N9, H3N8 and H10N4.  A few (of many) recent blogs include:

Frontiers: Has Avian Influenza Virus H9 Originated From a Bat Source?

Frontiers in Public Health: Human Infections of H9N2 Avian Influenza Virus in China (in 2021)

Mainland China Reports 2 Human LPAI H9N2 Infections (Sichuan Province)

Eurosurveillance: Analysis of Avian Influenza A (H3N8) Viruses in Poultry and their Zoonotic Potential

Emerging Microbes & Inf.: H9N2 Cluster Among Humans, Chickens & Pet Cat - China, 2018

Making matters arguably worse, ineffective vaccines used in China have helped to drive H9N2's evolution (see J. Virus Erad.: Ineffective Control Of LPAI H9N2 By Inactivated Poultry Vaccines - China), leading to the creation and spread of scores of genotypes. 

In the fall of 2015, in That Touch Of Mink Flu (H9N2 Edition) we looked at a serological survey of antibodies to H9N2 (along with H5 & H7 viruses) in Chinese farmed minks, along with the results of experimental infection of minks with the H9N2 virus.

Mink inoculated with the H9N2 subtype replicated the virus in their lungs (and to a lesser extent) heart, brain, and kidney. While H9N2 infection was non-fatal for mink, they developed lung lesions, edema, and shed the virus through their respiratory tract

Two years later, in That Touch Of Mink Flu (H9N2) - Revisited, we looked at another study, published in Nature's Scientific Reports, on the seroprevalence and transmissibility of H9N2 from minks to other peridomestic animals.  The author's wrote:

Transmission experiments showed that close contact between H9N2 infected mink and naïve contact mink, foxes and raccoon dogs resulted in spread of the virus to the sentinel animals as determined by virus isolation and/or seroconversion.

Not only were mink highly susceptible to H9N2, they could pass it on to other mammals. As we've seen with other influenza viruses (and SARS-CoV-2), replication and spread in mink can often produce genetic changes that can lead to better mammalian adaptation.

Last summer, in PNAS: Mink Farming Poses Risks for Future Viral Pandemics, we looked at an excellent opinion piece penned by two well known virologists from the UK (Professor Wendy Barclay & Tom Peacock) on why fur farms - and mink farms in particular - are high risk venues for flu.

All of which bring us to a new report on the identification of a chicken–human–mink recombinant H9N2 virus (ZJ81 strain) in Guangdong province, along with 5 other strains that `. . . had a bias for the human receptor-binding site and a mutation that could cause an increase in virulence in mice.'

Among the mutations detected, the authors report:
The six viruses obtained in this research had the HA-Q226L mutation but not the G228S mutation, indicating that the viruses have a preference to bind to human receptors (α-2,6-SA) [29]. The ML3 strain showed a T190V mutation in the HA protein. This means that the virus shows increased replication in mice, and the rest of the virus strains do not have this mutation [30]. Additionally, all six virus strains have the D225G mutation in the HA protein, which suggests the ability to increase transmission and replication in pigs [31].
This is a lengthy, detailed, and often technical report, so I've only posted the abstract and a few exceprts below. Follow the link to read it in its entirety.  I'll have a bit more after the break. 

One-Health Challenge in H9N2 Avian Influenza: Novel Human-Avian Reassortment Virus in Guangdong Province, China
Qiucheng Yao,1Jing Liu,1Huizhen Liu,1Yan Zhou,1Miaotong Huo,1Yuanguo Li,2Yuwei Gao,2and Ye Ge1

Published 13 Feb 2024


China is one of the highest producers of poultry meat output in the world, with a large scale of chicken rearing. Statistically analyzed H9N2-subtype avian influenza viruses (AIVs) have become the dominant subtype in China’s live poultry market, with the highest detection rate. Although H9N2 AIV is of low pathogenicity and tends not to cause serious disease and high mortality in poultry, it poses a great challenge to the domestic poultry farming industry by causing a decrease in appetite, a decline in egg production, and deaths caused by mixed infections with another pathogenic microorganism.
Moreover, novel influenza viruses (H7N9 and H3N8) infecting humans have emerged in China, and the H9N2 AIV provides all or part of the internal genes to the new recombinant viruses, posing a potential threat to public health and safety and human health.
In this research, six H9N2 AIVs were isolated from feces or oropharyngeal swabs collected from live poultry markets and duck farms in Zhanjiang. After epidemiological investigations, phylogenetic analyses, and molecular characterization, we found that the ZJ81 strain was a chicken–human–mink recombinant virus, the ML3 strain was a chicken-human recombinant virus, and all six virus strains of the virus had a bias for the human receptor-binding site and a mutation that could cause an increase in virulence in mice. Therefore, surveillance and control of H9N2 AIV should be strengthened to provide data support for cross-species transmission of H9N2 AIV.


In this research, we mainly sampled the live poultry trading market and farms in Zhanjiang City, Guangdong Province, from 2019 to 2021. The whole genomes of these H9N2 AIVs were sequenced and analyzed by epidemiological investigations, phylogenetic analyses, and molecular characterization. Our study provides strong data to support the prevention of cross-species spread of H9 = N2 AIV in domestic poultry.


Phylogenetic analysis showed that two virus strains, ZJ81 and ML3, were differentially recombined with human influenza viruses. The M gene of the two viruses had the highest homology to that of the human influenza virus. Therefore, ZJ81 is a chicken–human–mink recombinant virus, and ML3 is a chicken–human recombinant virus.

In October 2018, an H9N2 AIV infection was detected among chicken, cat, and human populations at a backyard in Guangxi. A human-infecting AIV strain, A/Guangxi/NN10.19T-NGS/2018 (H9N2), and a cat-infecting AIV strain, A/cat/Guangxi/NN10.19H-NGS/2018 (H9N2), from this study were on the same subbranch as the DF4 strain isolated in this study, with homologies of 96.7% and 96.8%, and the homology with ZJ92 and ZJ81 was approximately 95%. This suggests that H9N2 AIV may be transmitted to humans or cats through chickens, increasing the risk of H9N2 AIV infection in humans, and shows that proactive prevention and control measures are warranted [39].

Molecular characterization showed that the HA protein of six virus strains had Q226L mutation, implying that the H9N2 strain is shifting from a predilection for avian to human receptors [29]. These six virus strains have three amino acid deletions in their NA protein stems. It has been indicated that NA protein stalk deletion of H9N2 AIV extends the host range, enhances infection of mice, and leads to the generation of mutations in amino acid 627 of the PB2 protein [40].
No amino acids associated with tolerance to anti-influenza drugs such as oseltamivir were found in the NA protein, showing that anti-influenza drugs are still effective against these six virus strains of H9N2 AIV [32]. Statistical analysis of eight gene segments with key amino acid site mutations in H9N2-subtype-AIVs worldwide from January 2017 to July 2023.
Mutations at the Q226L site are common in the H9N2 AIV HA protein, suggesting that avian-origin H9N2 viruses generally can bind the human receptor, increasing the public health risk of human infections. Moreover, the M1-N30D and M1-T215A mutations, the M2-S31N mutation, and the NS1-P43S mutation are prevalent in H9N2 AIV, with a mutation rate of more than 88% (Figure 8). All these mutations increase viral pathogenicity in mice, suggesting a potential risk of H9N2 virus infection in humans. 

5. Conclusion

The continued prevalence of AIV of the H9N2 subtype in poultry not only seriously jeopardizes the live poultry markets but also threat to public health security. Furthermore, complex viral recombination may pose a threat to poultry and humans. This study further confirms the prevalence of H9N2-subtype AIV strains worldwide and thus further emphasizes the requirement for the importance of real-time monitoring and control of H9N2 AIV transmission and its importance in poultry farming and human health.

         (Continue . . . )

Over the past 15 years we've watched as H9N2 has expanded its geographic range - spreading out of Asia into Europe, the Middle East and Africa (see 2019's Viruses: A Global Perspective on H9N2 Avian Influenza Virus). .  

As an LPAI (low path avian influenza) virus , H9N2 is not considered a `reportable' disease by WOAH (formerly the OIE), even though it is clearly zoonotic. We've seen seroprevalence studies which suggest people with exposure to infected poultry often develop H9 antibodies, suggesting mild or asymptomatic infection are likely common. 

The CDC has 2 different lineages (A(H9N2) G1 and A(H9N2) Y280) on their short list of influenza viruses with zoonotic potential (see CDC IRAT SCORE), and several candidate vaccines have been developed.

But much of what happens with this virus (and many others) invariably occurs outside of our view; in a complex and frequently intertwined eco-system that includes wild birds, poultry farms, live bird markets (LBMs), pig herds, peridomestic animals, and fur farms around the world.  

Today's report provides is a small snapshot of H9N2's evolution from a specific geographic location roughly 5 years ago.  While it provides us with valuable information, it is far from being an early warning system.

If it were easy for nature to generate pandemic pathogens, we'd be hip-deep in pandemic viruses all the time. But this study reminds us that there are countless unmonitored gain of function studies ongoing in the wild every day

While the odds of success may be long, nature gets to keep trying until it gets it right. 

When that invariably happens, we will likely have little or no warning.  Which is why we need to be preparing today, as if it were already on the way.