When HPAI H5N1 was unexpectedly discovered in American dairy cows last March, many countries immediately assumed this spillover was due to a unique - and likely geographically limited - genotype (B3.13) of the virus.
The UK's rationale for not testing their cattle (see UK HAIRS Risk Statement On Avian Influenza (H5N1) In Livestock) for HPAI was that the H5N1 genotype B3.13 had never been seen in Europe or the UK, and was unlikely to cross oceans.Cattle had never been thought susceptible before (see A Brief History Of Influenza A In Cattle/Ruminants), and many countries seemed to content not to look for similar cases.
A little over a month ago, researchers from Germany's Friedrich-Loeffler-Institut (FLI) announced that in early experiments, they found that several European strains of HPAI H5N1 were capable of infecting bovine cells (see FLI Statement On Experimental Infection Of Dairy Cows With European H5N1 Virus).
While unwelcome news, it appears to have had an effect. Earlier this week the UK raised their pre-pandemic risk level (from 3 to 4), and warned:
With the current evidence we should not assume that the risk of a mammalian outbreak is limited to clade B.3.13.
In last Thursday's Eurosurveillance, researchers from Germany's FLI published their full results, including their study on the effectiveness of heat inactivation on different strains of the HPAI H5N1 virus.
They not only confirmed that cattle are susceptible to a range of HPAI H5N1 viruses, they demonstrate that heat treatment at 56 °C for 30 minutes did not completely eliminate the infectivity of the tested AIV in milk.
This is a lengthy and detailed report, with a lot to take in. I've only posted some excerpts, so follow the link to read it in its entirety. I'll have a brief postscript after the break.
Diana I Palme1 , Juliane Lang1 , Dajana Helke1 , Maryna Kuryshko1 , Elsayed M Abdelwhab1
Cattle were once thought to be resilient to avian influenza virus (AIV) infections [1], until the recent widespread influenza A(H5N1) virus infections in dairy cattle in several states in the Unites Stares (US) since March 2024 [2]. Surprisingly, high viral loads were found in the milk of infected cows, but not in the respiratory tract [3]. Consumption of unpasteurised influenza A(H5N1)-contaminated milk represents a new niche of possible public health concern for an avian virus. Cats on the dairy farms [3] and laboratory mice got infected and died after drinking influenza A(H5N1)-contaminated unpasteurised colostrum and milk [4]. To date, four dairy farm workers have been reported with influenza A(H5N1)-associated conjunctivitis and upper respiratory influenza-like illness [5].
Little is known about the efficiency of H5N1 clade 2.3.4.4b, particularly non-US isolates, to replicate in bovine cells and remain infectious in milk with different fat contents. Here we investigated the effect of different fat contents on the thermostability and duration of heat inactivation of H5N1 influenza viruses and assessed the replication of recent German H5N1 viruses of clade 2.3.4.4b in bovine kidney and lung cells.
(SNIP)
Four influenza A(H5N1) viruses isolated in Germany were used in this study. These included three recent clade 2.3.4.4b viruses A/chicken/Germany/AI04286/2022 (designated H5N1-chicken), A/wood pigeon/Germany-NW/AI00951/2022 (H5N1-Pigeon; accession number EPI_ISL_10261376) and A/red knot/AI000616/2022 (H5N1-Knot; accession number EPI_ISL_18006920), as well as a clade 2.2.2 A/swan/Germany/R65/2006 (designated H5N1-Swan; accession number EPI_ISL_10142). In addition, the historic A/turkey/England/384/79 (H10N4) virus was used as a control.
(SNIP)
Heat inactivation at 75 °C for 30 min or 56 °C for 120 min effectively reduced the infectivity of all viruses below the detection limit of the plaque assays (Figure 1B–F) [6]. However, heat inactivation for 30 min at 56 °C did not completely eliminate the infectivity of the tested AIV: Despite reduced titres, H10N4, H5N1-Swan and H5N1-Chicken retained infectivity in whole milk, and H5N1-Swan and H5N1-Pigeon were detectable in semi-skimmed milk (Figure 1). In addition, infectious H10N4 and, to a lesser extent, H5N1-Knot and H5N1-Pigeon viruses were detected in virus-spiked MEM (Figure 1). Overall, thermostability varied by virus strain and fat content of the milk. Complete virus inactivation required heat treatment at 56°C for 120 minutes.
Replication of influenza A(H5N1) viruses in bovine cell culture
Viral replication was assessed in Madin–Darby bovine kidney (MDBK) cells infected with a multiplicity of infection (MOI) of 0.001 for 24 h. All H5N1 viruses replicated in MDBK cells (Figure 2A), while H10N4 did not induce detectable titres without trypsin (data not shown). The levels of H5N1-Knot and H5N1-Pigeon were ca 10 times lower than those of H5N1-Swan and H5N1-Chicken (Figure 2A). We further assessed the cell-to-cell spread in MDBK cells relative to Madin–Darby canine kidney type II (MDCKII) cells by measuring plaque diameters. The intercellular spread of H5N1 clade 2.3.4.4b viruses in MDBK cells was notably more homogeneous compared with clade 2.2.2 H5N1-Swan (Figure 2B). In contrast, MDCKII cells showed a mixed population of plaques, indicating a more heterogeneous spread (Figure 2C–E).
(SNIP)
Discussion
During the current H5N1 outbreak in the US, raw cow milk has been identified as a source of influenza virus infection among mammals near infected dairy cows. In addition, bovine A(H5N1) and human A(H1N1)pdm09 influenza viruses retained infectivity on milking machines for up to 3 h [7]. It is known that pasteurisation of milk at 63 °C for 30 min or, alternatively, heating to 72–154 °C for a few seconds, inactivates a wide range of pathogens and ensures safety for human consumption [8-10]. Studies have shown that American and Asian influenza viruses are effectively inactivated at pasteurisation temperatures.
It remains unclear if milk with varying fat contents presents a unique medium for influenza virus infectivity compared with allantoic fluid or serum samples. Our study demonstrated that all viruses were undetectable after treatment at 75 °C for 30 min, consistent with recent findings on American and Chinese H5N1 influenza viruses [4,11,12].
Interestingly, while viral titres significantly decreased at 56 °C, residual infectious virus was detected after 30 min in whole milk (three strains), semi-skimmed milk (one strain) and MEM (two strains). These findings suggest thermostability may vary by virus strain, and whole milk could impact the stability of certain influenza viruses. Similar strain-dependent variations were seen in Chinese virus-spiked milk [12]. To ensure complete inactivation of A(H5N1) viruses and thus biosafety under laboratory conditions, higher temperatures should be considered.
Furthermore, all H5N1 viruses and H10N4 (in the presence of trypsin), replicated efficiently in bovine cells, whereas pigeon and red knot viruses exhibited lower titres. This indicates that certain H5N1 viruses can replicate effectively in bovine lung and kidney cells without prior adaptation. The reasons for the cell and virus dependency to infect bovine cells are not clear, and additional studies are needed to understand the distribution of sialic acid receptors on both cell types and receptor binding specificity/affinity of these viruses.
(SNIP)
Conclusion
This study demonstrates that non-US influenza A(H5N1) viruses can also infect and replicate in bovine kidney and lung cells, albeit with varying titres, spread and infectivity. Some viruses remained infectious after exposure to 56 °C for 30 min, particularly in whole milk. Therefore, considering strain dependency is crucial when determining virus inactivation protocols at this temperature for biosafety reasons.
Once again we find that the deeper we look, the more we find.
Yet in far too many places around the globe, Don't test, don't tell' continues to be the preferred option.
As a result, we don't know how pervasive HPAI is in American cattle, and we know even less about what is going on with H5Nx in places like China, Asia, Russia, the Middle East, and Africa.
There are some signs that governments and agencies are starting to take the threat more seriously (see Colorado Orders Mandatory Statewide Weekly Bulk-Tank Testing for HPAI and UKHSA Raises Risk Assessment On HPAI H5N1), but in far too many places half-measures and bland reassurances remain the order of the day.
We may get lucky, and HPAI H5Nx may fizzle. But that is not its current trajectory, and we risk enormous damage by continuing to underestimate its future impacts.
