Increased public health threat of avian-origin H3N2 influenza virus caused by its evolution in dogs

Curation statements for this article:
  • Curated by eLife

    eLife logo

    eLife assessment

    The authors characterize an H3N2 influenza A virus that jumped from birds into dogs in 2006. Through its evolutionary adaptation to dogs, the virus is now gaining properties that are increasingly consistent with the potential to infect humans. Using experiments with canine H3N2 influenza isolates, the authors found that more recent viruses have acquired receptor specificity for both avian- and human-like receptors, enhanced low-pH stability and in vitro growth, as well as improved replication and transmission in the dog and ferret models.

This article has been Reviewed by the following groups

Read the full article See related articles

Abstract

Influenza A viruses in animal reservoirs repeatedly cross species barriers to infect humans. Dogs are the closest companion animals to humans, but the role of dogs in the ecology of influenza viruses is unclear. H3N2 avian influenza viruses were transmitted to dogs around 2006 and have formed stable lineages. The long-term epidemic of avian-origin H3N2 virus in canines offers the best models to investigate the effect of dogs on the evolution of influenza viruses. Here, we carried out a systematic and comparative identification of the biological characteristics of H3N2 canine influenza viruses (CIVs) isolated worldwide over 10 years. We found that, during adaptation in dogs, H3N2 CIVs became able to recognize the human-like SAα2,6-Gal receptor, showed gradually increased hemagglutination (HA) acid stability and replication ability in human airway epithelial cells, and acquired a 100% transmission rate via respiratory droplets in a ferret model. We also found that human populations lack immunity to H3N2 CIVs, and even preexisting immunity derived from the present human seasonal influenza viruses cannot provide protection against H3N2 CIVs. Our results showed that canines may serve as intermediates for the adaptation of avian influenza viruses to humans. Continuous surveillance coordinated with risk assessment for CIVs is necessary.

Article activity feed

  1. Author Response

    Reviewer #1 (Public Review):

    Although the authors have identified some properties/molecular markers of canine H3N2 influenza viruses that highlight the potential for infecting humans, it needs to be cautious to emphasize the threat of these viruses to public health. One fact is that despite the increasing prevalence of these viruses in dogs and the close proximity between dogs and humans, there is so far no report of human infection with canine H3N2 influenza viruses. The authors are wished to discuss this in their manuscript so that the readers can have a more comprehensive understanding of their findings and the public health importance of canine influenza viruses.

    We agree with the reviewer. We added the related discussion and revised some words to not emphasize the threat of these viruses to public health (lines 342-346).

    Reviewer #3 ( Public Review):

    1. The investigators should run neuraminidase inhibition assays to established the level of cross reactivity of human sera to the canine origin NA (one of reasons proposed as to the lower impact of the H3N2 pandemic was the presence of anti0N2 antibodies in the human population).

    We performed neuraminidase inhibition assays as suggested for both ferret sera against human H3N2 virus and human sera. The results showed that the NI titers of ferret sera against human H3N2 virus to canine H3N2 viruses were <10 (lines 147- 148, Supplementary file 2). Additionally, 2.0%–3.0% of the children's serum samples, 1.0%–2.0% of the adult's serum samples, and 1.0%–2.0% of the elderly adult's serum samples had NI antibody titers of ≥10 to canine origin NA (lines 158-161, Table 1, and lines 435-445).

    1. Please tone down the significance of ferret-to-ferret transmission as a predictor of human-to-human transmission. Although flu viruses that transmit among humans do show the same capacity in ferrets, the opposite is NOT always true.

    We agree with the reviewer. To tone down the significance of ferret-to-ferret transmission as a predictor of human-to-human transmission, we added the related discussion and deleted or revised some words (lines 342-346, line 37, line 302, line 308, line 322, and line 341).

  2. eLife assessment

    The authors characterize an H3N2 influenza A virus that jumped from birds into dogs in 2006. Through its evolutionary adaptation to dogs, the virus is now gaining properties that are increasingly consistent with the potential to infect humans. Using experiments with canine H3N2 influenza isolates, the authors found that more recent viruses have acquired receptor specificity for both avian- and human-like receptors, enhanced low-pH stability and in vitro growth, as well as improved replication and transmission in the dog and ferret models.

  3. Reviewer #1 (Public Review):

    The authors conducted an extensive characterization of canine H3N2 influenza viruses. By analyzing gene sequences of canine H3N2 influenza viruses isolated in their laboratory and those that are available in public databases, they identified various genetic clades (also somehow correlate with antigenic groups identified in serological assays) and human-like amino acid substitutions in these viruses, which indicated the evolution of these viruses towards potentially more adaptive to humans. By experiments with several selected canine H3N2 influenza isolates, they found that more recent canine H3N2 influenza viruses have acquired receptor specificity for both avian- and human-like receptors, enhanced low-pH stability and in vitro growth as well as improved replication and transmission in the dog and ferret models. They further identified amino acid substitutions underlying the improved transmissibility of these canine H3N2 influenza viruses. The study was well-designed and the conclusions in the manuscript are in general well supported by the experimental data. Findings from the study will certainly help understand the evolution of canine influenza viruses and assessing the risk posed by these viruses to public health.

    Although the authors have identified some properties/molecular markers of canine H3N2 influenza viruses that highlight the potential for infecting humans, it needs to be cautious to emphasize the threat of these viruses to public health. One fact is that despite the increasing prevalence of these viruses in dogs and the close proximity between dogs and humans, there is so far no report of human infection with canine H3N2 influenza viruses. The authors are wished to discuss this in their manuscript so that the readers can have a more comprehensive understanding of their findings and the public health importance of canine influenza viruses.

  4. Reviewer #2 (Public Review):

    The authors show how an avian influenza A virus that jumped into dogs is now evolving in real time. Though its evolutionary adaptation to dogs, the virus is gaining properties that are increasingly consistent with the potential to infect humans.

    The data are alarming, although it should be emphasized that this dog H3N2 influenza virus has not yet infected humans, and perhaps never will. It is also unknown how pathogenic (medically serious) the virus would be in humans if it were to jump. The authors show preliminary data that prior exposure to human seasonal H3N2 will not render us resistant to this dog virus should it jump to humans.

    What is most remarkable about this study is the breadth of experimental approaches taken, and the holistic analysis of what is bound to become a classic tale in virus evolution and emergence through an intermediate host.

  5. Reviewer #3 (Public Review):

    The manuscript by Chen et al shows solid evidence that canine origin influenza viruses are evolving towards a more mammalian adapted phenotype. The data also show that humans may lack proper protection against these viruses if they were to evolve more prone to cross to humans. There are some aspects of the ms that need to be addressed: 1) The investigators should run neuraminidase inhibition assays to established the level of cross reactivity of human sera to the canine origin NA (one of reasons proposed as to the lower impact of the H3N2 pandemic was the presence of anti0N2 antibodies in the human population), 2) please tone down the significance of ferret-to-ferret transmission as a predictor of human-to-human transmission. Although flu viruses that transmit among humans do show the same capacity in ferrets, the opposite is NOT always true.