Yet another avian influenza virus, H10N8, infects humans

chicken market

To the collection of avian influenza viruses known to sporadically infect humans – H5N1, H7N9, H7N2, H7N3, H7N7, H9N2, and H10N7 – we can now add H10N8, recently found in two individuals in China.

Avian influenza virus H10N8 was first detected in tracheal aspirates from a 73 year old woman who was hospitalized in November 2013 for severe respiratory illness. The patient, who died, had previously visited a live poultry market. A second infection with this virus was detected in January 2014.

Virus isolated from tracheal aspirates on day 7 of illness was named A/Jiangxi-Donghu/346/2013(H10N8). Nucleotide sequence analysis of the viral genome reveals that it is a reassortant. The HA gene most closely resembles that of a virus isolated from a duck in Hunan in 2012, while the NA gene resembles that of a virus isolated from a mallard in Korea in 2010. All six other RNA segments resemble those from circulating H9N2 viruses in China. These viruses have also provided genes for H7N9 and H5N1 viruses.

Examination of the viral protein sequences provides some clues about virulence of the virus. The HA protein sequence reveals a single basic amino acid at the cleavage site, indicating that the virus is of low pathogenicity in poultry, like H7N9 virus. The sequence in the sialic acid binding pocket of the HA protein indicates a preference for alpha-2,3 linked sialic acids, typical  for avian influenza viruses (human influenza viruses prefer alpha-2,6 linked sialic acids). A lysine at amino acid 627 in the PB2 protein is known to enhance the ability of the virus to replicate at mammalian temperatures; the H10N8 virus has a mixture of lysine and glutamic acid, the residue associated with less efficient replication. The sequence of the M2 protein indicates that the virus is resistant to the antiviral adamantanes. In vitro testing indicated sensitivity to NA inhibitors Tamiflu and Relenza.

It is not known if this novel H10N8 virus will spread further in the human population. A novel influenza H7N9 virus was first detected in humans in early 2013 and has since caused 250 human infections with 70 deaths. Similar incursions of avian influenza viruses into humans have probably taken place for as long as humans have had contact with poultry. We are now adept at detecting viruses and therefore we are noticing these infections more frequently.

Live poultry markets are clearly a risk factor for humans to acquire infections with avian influenza viruses, as noted by Perez and Garcia-Sastre:

Live bird markets in Asia are undoubtedly the major contributor in the evolution of avian influenza viruses with zoonotic potential, a fact for which we seem to remain oblivious.

Given their role in transmitting new viruses from animals to humans, I wonder why live poultry markets are not permanently closed.

Update: George Gao agrees that the live poultry markets in China should be closed.

Avian influenza H7N9 viruses isolated from humans: What do the gene sequences mean?

Influenza A virionThere have been over 60 human infections with avian influenza virus H7N9 in China, and cases have been detected outside of Shanghai, including Beijing, Zhejiang, Henan, and Anhui Provinces. Information on the first three cases has now been published, allowing a more detailed consideration of the properties of the viral isolates.

The first genome sequences reported were from the initial three H7N9 isolates: A/Shanghai/1/2013, A/Shanghai/2/2013, and A/Anhui/1/2013. These were followed by genome sequences from A/Hongzhou/1/2013 (from a male patient), A/pigeon/Shanghai/S1069/2013), A/chicken/Shanghai/S1053/2013), and A/environment/Shanghai/S1088/2013, the latter three from a Shanghai market.

Analysis of the viral genome sequences reveals that all 8 RNA segments of influenza A/Shanghai/1/2013 virus are phylogenetically distinct from A/Anhui/1/2013 and A/Shanghai/2/2013, suggesting that the virus passed from an animal into humans at least twice. Similar viruses have been isolated from pigeons and chickens, but the source of the human infections is not known. There is as yet no evidence for human to human transmission of the H7N9 viruses, and it seems likely that all of the human infections are zoonotic – transmission of animal viruses to humans. Since the H7N9 viruses are of low pathogenicity in poultry, infected animals may not display disease symptoms, further facilitating transmission to humans.

The RNA sequences reveal that the H7N9 viruses isolated from humans are all triple reassortants, which means that they contain RNA segments derived from three parental viruses. The gene encoding the hemagglutinin protein (HA) is most closely related to the HA from A/duck/Zhejiang/12/2011 (H7N3), while the NA gene is most similar to the NA gene from A/wild bird/Korea/A14/2011 (H7N9). The remaining 6 RNA segments are most related to genes from A/brambling/Beijing/16/2012-like viruses (H9N2). The type of animal(s) in which the mixed infections took place is unknown.

Some observations on the relatedness of these sequences:

  • A/Shanghai/2/2013, A/Anhui/1/2013, and A/Hangzhou/1/2013 were isolated in distant cities yet have over 99% identity. The pigeon, chicken, and environmental isolates are also very similar except for one gene of A/pigeon/Shanghai/S1069/2013. Long-range shipping of infected poultry might explain these similarities.
  • There are 53 nucleotide differences between A/Shanghai/1/2013 and A/Shanghai/2/2013. Perhaps A/Shanghai/1/2013 and the remaining viruses originated from different sources.

When the gene sequences of these human viral isolates are compared with closely related avian strains, numerous differences are revealed. The locations of the proteins in the influenza virion are shown on the diagram; click for a larger version (figure credit: ViralZone).

  • All seven H7N9 viruses do not have multiple basic amino acids at the HA cleavage site. The presence of a basic peptide in this location allows the viral HA to be cleaved by proteases that are present in most cells, enabling the virus to replicate in many organs. Without this basic peptide, the HA is cleaved only by proteases present in the respiratory tract, limiting replication to that site. This is one reason why the H7N9 viruses  have low pathogenicity in poultry.
  • All seven viruses have a change at HA amino acid 226 (Q226L) which could improve binding of the viruses to alpha-2,6 sialic receptors, which are found throughout the human respiratory tract. Avian influenza viruses prefer to bind to alpha-2,3 sialic acid receptors. This observation suggests that the H7N9 isolates should be able to infect the human upper respiratory tract (alpha-2,3 sialic acid receptors are mainly located in the lower tract of humans). However, viruses which bind better to alpha-2,3 sialic acids still bind to alpha-2,6 receptors and can infect humans.
  • All seven viruses have a change at HA amino acid 160 from threonine to alanine (T160A). This change, which has been identified in other circulating H7N9 viruses, prevents attachment of a sugar to the HA protein and could lead to better recognition of human (alpha-2,6 sialic acid) receptors.
  • Five amino acids are deleted from the neuraminidase (NA), the second viral glycoprotein, in all seven viruses. In avian H5N1 influenza virus this change may influence tropism for the respiratory tract and enhance viral replication, and might regulate transmission in domestic poultry. This change is believed to be selected upon viral replication in terrestrial birds.
  • One of the viruses (A/Shanghai/1/2013) has an amino acid change in the NA glycoprotein associated with oseltamivir resistance (R294K).
  • An amino acid change in the PB1 gene, I368V, is known to confer aerosol transmission to H5N1 virus in ferrets.
  • An amino acid change in the PB2 gene, E627K, is associated with increased virulence in mice, higher replication of avian influenza viruses in mammals, and respiratory droplet transmission in ferrets.
  • Changes of P42S in NS1 protein, and N30D and T215A in M1 are associated with increased virulence in mice, but these changes are also observed in circulating avian viruses.
  • All seven viruses have an amino acid change in the M2 protein known to confer resistance to the antiviral drug amantadine.
  • All seven viruses lack a C-terminal PDZ domain-binding motif which may reduce the virulence of these viruses in mammals.

For the most part we do not know the significance of any of the amino acid changes for viral replication and virulence in humans.

I believe that these H7N9 viruses might take one of two pathways. If they are widespread in birds, they could spread globally and cause sporadic zoonotic infections, as does avian influenza H5N1 virus. Alternatively, the H7N9 viruses could cause a pandemic. Influenza H7N9 virus infections have not occurred before in humans, so nearly everyone on the planet is likely susceptible to infection. Global spread of the virus would require human to human transmission, which has not been observed so far. Some human to human transmission of avian H7N7 influenza viruses was observed during an outbreak in 2003 in the Netherlands, but those viruses were different from the ones isolated recently in China. Whether or not these viruses will acquire the ability to transmit among humans by aerosol is unknown and cannot be predicted. If a variant of H7N9 virus that can spread among humans arises during replication in birds or humans, it might not have a chance encounter with a human, or if it did, it might not have the fitness to spread extensively.

What also tempers my concern about these H7N9 viruses is the fact that the last influenza pandemic (H1N1 virus) took place in 2009.  No influenza pandemics in modern history are known to have taken place 4 years apart, although only 11 years separated the 1957 (H2N2) and 1968 (H3N2) pandemics. I suppose that is not much consolation, as there are always exceptions, especially when it comes to viruses.

Meanwhile a vaccine against this H7N9 strain is being prepared (it will be months before it is ready), surveillance for the virus continues in China and elsewhere, and health agencies ready for a more extensive outbreak. These are not objectionable courses of action. But should this be our response to every zoonotic influenza virus infection of less than 100 cases?


Human Infection with a Novel Avian-Origin Influenza A (H7N9) Virus.

Genetic analysis of novel avian A(H7N9) influenza viruses isolated from patients in China, February to April 2013.

TWiV 223: EEEV and the serpent

On episode #223 of the science show This Week in Virology, Vincent, Alan, and Kathy discuss new influenza virus NA inhibitors, detection of EEEV antibody and RNA in snakes, and replication of the coronavirus EMC in human airway epithelial cells.

You can find TWiV #223 at

The abundant and diverse viruses of the seas

earthWhat is the most abundant biological entity in the oceans?

Viruses, of course! The quantity and diversity of viruses in the seas are staggering. Each milliliter of ocean water contains several million virus particles – a global total of 1030 virions! If lined up end to end, they would stretch 200 million light years into space. Viruses constitute 94% of all nucleic-acid containing particles in the sea and are 15 fold more abundant than bacteria and archaea.

Because viruses kill cells, they have a major impact on ocean ecology. About 1023 virus infections occur each second in the oceans; in surface waters they eliminate 20-40% of prokaryotes daily. Viral lysis converts living organisms into particulate matter that becomes carbon dioxide after respiration and photodegradation. Cell killing by viruses also liberates enough iron to supply the needs of phytoplankton, and leads to the production of dimethyl sulphoxide, a gas that influences the climate of the Earth. Because of these activities, marine viruses have a significant impact on global microbial communities and geothermal cycles.

Most of the marine viruses are bacteriophages, but there are also significant numbers that infect eukaryotic phytoplankton, invertebrates, and vertebrates. The best studied viruses are those that infect commercially important species. Novel viruses are frequently discovered; for example, white spot syndrome virus of panaeid shrimp is a member of a new virus family. Viruses of commercially important finfish include herpesviruses, reoviruses, nodaviruses, birnaviruses, and rhabdoviruses. How these viruses are transmitted among marine species is not understood. Many viruses move between marine and fresh waters, posing threats to fishing industries. The rhabdovirus viral hemorrhagic septicemia virus, which causes disease in European farmed trout, has been isolated from 40 marine fish species, from fish farms in Alaska, and from fish in the Great Lakes.

Many ocean viruses cause disease in marine mammals. Phocid distemper virus is a morbillivirus of Arctic phocid seals that has killed thousands of harbor seals in Europe. Similar viruses kill dolphins and other cetaceans. Many other viruses infect marine mammals and even cause disease in humans, including adenoviruses, herpesviruses, parvoviruses, and caliciviruses. The natural reservoirs of most of these viruses are unknown.

Massive sequencing projects have been used to provide information on the diversity of marine viruses. In these studies, seawater is filtered to remove large particles, virions are purified by centrifugation, and nucleic acids are extracted, amplified, and subjected to pyrosequencing. Bioinformatic approaches are used to sift through megabase data sets to identify viral sequences. In one study the viral genomes (‘viromes’) from the Arctic Ocean, the Sargasso Sea, and the coastal waters of British Columbia and the Gulf of Mexico were compared. Over 90% of the sequences were not found in the GenBank collection. There was also little sequence overlap among the samples from the four sites. Similar studies have revealed a rich array of RNA viruses in two different coastal environments; again, most of the sequences were not present in current databases. From the results of these studies it has been estimated that the oceans probably harbor several hundred thousand viral species.

Much more work is required to understand the diversity of marine viruses and their role in the global ecosystem. From the studies done to date, one conclusion is quite clear: the numbers of viruses in the oceans, and their impact on marine life, is far greater than we we ever imagined. And the zoonotic pool may be much larger than we suspected.

Suttle, C. (2007). Marine viruses — major players in the global ecosystem Nature Reviews Microbiology, 5 (10), 801-812 DOI: 10.1038/nrmicro1750

Angly, F., Felts, B., Breitbart, M., Salamon, P., Edwards, R., Carlson, C., Chan, A., Haynes, M., Kelley, S., Liu, H., Mahaffy, J., Mueller, J., Nulton, J., Olson, R., Parsons, R., Rayhawk, S., Suttle, C., & Rohwer, F. (2006). The Marine Viromes of Four Oceanic Regions PLoS Biology, 4 (11) DOI: 10.1371/journal.pbio.0040368

Culley, A., Lang, A.S., & Suttle, C.A. (2006). Metagenomic Analysis of Coastal RNA Virus Communities Science, 312 (5781), 1795-1798 DOI: 10.1126/science.1127404

The zoonotic pool

2209806245_9b9b88e0e5_mI previously discussed the idea that new human virus infections will continue to emerge from animal hosts. Stephen Morse, my colleague here at Columbia, has called this collection of viruses the ‘zoonotic pool’. How many viruses are in this pool?

Here are Dr. Morse’s calculations: assume that there are 50,000 vertebrates on earth, each of which harbors 20 different viruses. That gives a total of 1 million vertebrate viruses. We have only identified about 2,000 viruses; therefore over 99.8% of vertebrate viruses have not yet been discovered!

In other words, the zoonotic pool is very large – providing many opportunities for new human infections, and for the scientists that study them. This realization has lead to the rapidly growing field of pathogen discovery, of which Ian Lipkin and Joe DeRisi are masters.

Morse, SS. 1993. Emerging viruses. Oxford University Press.

Lipkin WI 2008. Pathogen Discovery. PLoS Pathog 4(4): e1000002.