#18,843
In 2009 we took a deep (3 blog) dive into a landmark PNAS study - Early and sustained innate immune response defines pathology and death in nonhuman primates infected by highly pathogenic influenza virus by Carole Baskin et. al. - which compared the pathogenesis of an older (2004) H5N1 virus, seasonal flu, and with two altered viruses carrying genetic material from the 1918 Spanish Flu in macaques.
The key finding from this study was that severe morbidity and mortality in macaques infected with H5N1 was linked to a severe `Cytokine storm'; a strong, early, and sustained innate immune response, marked by elevated pro-inflammatory cytokines and severe lung damage (notably, apoptosis of dendritic cells).
As we've discussed previously, the HPAI H5N1 viruses circulating today are a far cry from the clades and variants that circulated in the early 2000's. They've evolved into new clades and subclades, and have reassorted into numerous genotypes (and subtypes).
The Cambodian clade 2.3.2.1e H5N1 virus appears to produce a nearly 50% CFR (case fatality rate), but the U.S. clade 2.3.4.4b `bovine' B3.13 genotype has a reputation for producing relatively mild symptoms in humans (primarily conjunctivitis, mild fever, cough, and mild flu-like symptoms).
This B3.13 genotype encompasses a wide range of variants, including at least one isolate (huTX37-H5N1) which carried the PB2-E276K mutation (glutamic acid → lysine at position 627), which is associated with increased mammalian adaption and pathogenicity.
While there are undoubtedly other amino acid changes (e.g. HA-T143A, PB2 D701N, PB2 Q591K, HA Q226L, etc.) that could increase the pathogenicity of HPAI H5N1, the E627K mutation is always at or near the top of the list.
All of which brings us to a new study, published yesterday in Cell Immunity, which revisits the pathogenicity of the new bovine B3.13 genotype (with E627K) in two types of macaques (Cynomolgus and Rhesus).
This lengthy, and at times technical, report finds that the `mild' bovine B3.13 genotype - when aided and abetted by the PB2-E627K mutation - can be every bit as deadly in macaques as the older H5N1 viruses from Vietnam the mid 2000s.
It also finds - much like we saw in 2009 - that ` . . . acute inflammation, cytokine storm, and immune dysregulation are hallmarks of H5N1 disease pathogenesis in both cynomolgus and rhesus macaques.'
For now, E627K remains rarely reported in the B3.13 genotype of H5N1, but it has turned up in other genotypes, including the D1.1 genotype variant which caused severe infections in both Canada and the United States.
Due to its length I've only posted the abstract, and a few excerpts. Follow the link to read the full report in its entirety. I'll have a brief postscript after the break.
Immunopathogenesis of lethal H5N1 avian influenza virus clade 2.3.4.4b infection in macaques
Hanne Andersen1 ∙ Malika Aid2 ∙ Jonathan J. Stone3 ∙ … ∙ Mark G. Lewis1 ∙ Amanda J. Martinot3 ∙ Dan H. Barouch2,6,7 dbarouch@bidmc.harvard.edu
Published online August 18, 2025
DOI: 10.1016/j.immuni.2025.07.020
HighlightsSummary
- Cynomolgus macaques developed lethal H5N1 influenza pneumonia
- Rhesus macaques demonstrated dose-dependent H5N1 influenza mortality
- Surviving animals were protected against high-dose re-challenge
- H5N1 influenza was characterized by severe inflammation and immune exhaustion
The H5N1 avian influenza virus clade 2.3.4.4b outbreak represents a major pandemic threat for humans, with some reported cases of severe and fatal respiratory illness. A key unanswered question is the pathogenesis of severe H5N1 disease following respiratory infection.
In this study, we explored mechanisms of pathogenesis of severe H5N1 disease in cynomolgus and rhesus macaques following infection with the H5N1 isolate A/Texas/37/2024 (huTX37-H5N1). Cynomolgus macaques developed severe pneumonia that was lethal in 100% of macaques by 7 days post-infection. By contrast, rhesus macaques demonstrated dose-dependent mortality, and surviving animals showed protective immunity against high-dose re-challenge.
A multi-omics analysis demonstrated that H5N1 infection was characterized by robust induction of proinflammatory cytokines, innate immune cells, complement, coagulation, apoptosis, and immune exhaustion pathways.
Taken together, our data indicate inflammation and immune dysregulation as key mechanisms of H5N1 pathogenesis in nonhuman primates.
(SNIP)
Discussion
In this study, we observed that huTX37-H5N1 infection of cynomolgus and rhesus macaques by the respiratory route resulted in severe and rapidly progressive consolidative necrotizing pneumonia across a broad range of challenge doses. Infection was lethal in 100% (24 of 24) of cynomolgus macaques and in 33% (3 of 9) of rhesus macaques in the titration studies.
Lung consolidation was also more extensive and severe in cynomolgus macaques than in rhesus macaques. Transcriptomic profiling showed enrichment of proinflammatory and immune exhaustion pathways following challenge as well as reduction of NK, T, and B cell activation pathways and increased expression of T cell exhaustion markers, particularly in cynomolgus macaques.
These findings suggest that acute inflammation, cytokine storm, and immune dysregulation may be key contributors to the pathogenesis of H5N1 severe disease.
(SNIP)
In summary, our data demonstrate severe clinical disease with H5N1 clade 2.3.4.4b in cynomolgus and rhesus macaques when inoculated by the respiratory route, particularly in cynomolgus macaques. Although most human cases have been mild, several cases of severe and critical respiratory disease in humans have been reported, suggesting the importance of understanding the pathogenesis of severe H5N1 disease.
Interestingly, while cynomolgus and rhesus macaques are closely related primate species - and are both used in biomedical research - there are some key differences which may help explain the higher mortality in cynomolgus macaques over rhesus.Our findings define key features of pathogenesis and immunity to H5N1 clade 2.3.4.4b infection in cynomolgus and rhesus macaques. Future studies could define the role of T cell responses and T cell exhaustion for H5N1 disease pathogenesis. Moreover, these models of H5N1 infection in nonhuman primates should prove useful in evaluating candidate vaccines and therapeutics.
Limitations of the study
This study is limited to the evaluation of one strain of H5N1 influenza in rhesus and cynomolgus macaques, and it is possible that other H5N1 strains may have different characteristics. Moreover, the pathogenicity of H5N1 influenza in macaques may differ from that in humans, potentially because of higher inoculation doses and different challenge routes. In addition, this study involves an H5N1 isolate with the human adaptation mutation PB2-E627K as a model for severe disease. This mutation is currently rare in bovine H5N1 isolates, and thus it is possible that future human infections will have other adaptation mutations.
- Cynomolgus macaques tend to be slightly smaller than rhesus macaques, and are generally easier to handle (less aggressive).
- Cynomolgus macaques tend to have smaller lungs - and more abundant avian-like α2,3 receptors in their lower respiratory tract - which may lead to more severe pneumonia
- Influenza viruses tend to produce higher fever and more acute inflammatory cytokines in cynomolgus versus rhesus monkeys
While neither species likely serve as ideal analogs for human exposure to H5N1, they do provide us with a sobering range of non-human primate responses to infection with this specific H5N1 B3.13 isolate.
But most of all, this study reaffirms that our currently mild `bovine' genotype may only lack one or two crucial fixed mutations to make it a genuine public health concern.
Assuming that some other clade/genotype doesn't get there first.
