Pathfinder
Editor, Senior Moderator
Surveillance on California dairy farms reveals multiple sources of H5N1 transmission
Posted August 01, 2025.
Download PDF
View ORCID ProfileA.J. Campbell, View ORCID ProfileMeredith Shephard, View ORCID ProfileAbigail P Paulos, View ORCID ProfileMatthew Pauly, View ORCID ProfileMichelle Vu, View ORCID ProfileChloe Stenkamp-Strahm, View ORCID ProfileKaitlyn Bushfield, Betsy Hunter-Binns, Orlando Sablon, View ORCID ProfileEmily E Bendall, William J Fitzimmons, View ORCID ProfileKayla Brizuela, View ORCID ProfileGracie Quirk, Nirmal Kumar, Brian McCluskey, View ORCID ProfileNishit Shetty, View ORCID ProfileLinsey C Marr, View ORCID ProfileJenna Guthmiller, Kevin Abernathy, View ORCID ProfileAdam S Lauring, Blaine T Melody, Marlene Wolfe, Jason Lombard, View ORCID ProfileSeema S Lakdawala
doi: https://doi.org/10.1101/2025.07.31.666798
Abstract
Transmission routes of highly pathogenic H5N1 between cows or to humans remain unclear due to limited data from affected dairy farms. We performed extensive air, farm wastewater, and milk sampling on 14 H5N1 positive dairy farms across two different California regions. Virus was detected in the air in milking parlors and from exhaled breath of cows. Infectious H5N1 virus was detected in the air and water streams; sequence analysis revealed viral variants on a farm in these locations. Longitudinal analysis of milk from the individual quarters of cows revealed a high prevalence of subclinical H5N1 positive cows and a heterogeneous distribution of infected quarters that maintained a consistent pattern over time. Our data highlight potential modes of H5N1 transmission on dairy farms.
Competing Interest Statement
SSL and LCM receive funds from Flu Lab and NIH. ASL receives funds from Flu Lab, NIH, and CDC. ASL receives consulting fees and research support from Roche, outside of the submitted work.
...
https://www.biorxiv.org/content/10.1101/2025.07.31.666798v1
----------------------------------------------------------
From the PDF link above:
...
Discussion
Elucidating the routes of transmission of H5N1 between cows is critical to defining
successful mitigation strategies. In this study we successfully detected H5N1 in the air and in
reclaimed farm wastewater on separate dairy farms on multiple days. This included infectious air
samples from three different milking parlors and viral RNA from the exhaled breath of rows of
cows on two distinct farms. Additionally, we detected viral RNA in farm wastewater at multiple
sites on various farms and infectious virus at two different sites on the same farm. Together, these
results highlight the expansive environmental contamination of H5N1 on affected dairy farms and
identify additional sources of viral spread between cows and to humans. Surprisingly, we also
observed an HA mutation in a residue (H5 HA 189, H3 numbering 193) known to alter sialic acid
specificity26 in the air during milking of infected cows, suggesting that HA mutations are emerging
during this outbreak but are not currently maintained. Notably, mutations at 189 are associated
with improved binding to human a2,6-linked sialic acids26 and a structure of a 2.3.4.4b H5 with
an a2,6-linked sialic acid analog shows position 189 helps stabilize receptor binding24. Whether
the N189D mutation observed in this study alters receptor binding and improved tropism to
humans remains to be determined. These data support that emerging viral variants can pose a risk
to humans working in the dairy parlors through exposure to viruses in the air and on surfaces.
Given the high viral loads of H5N1 found in milk, milking equipment has been implicated
as the likely source for viral transmission. However, data from our longitudinal study revealed a
heterogeneous pattern of positivity across the subset of cows analyzed, both in the number and
location of positive quarters. If milking equipment alone was the driver of cow-to-cow H5N1
transmission it is likely that a similar pattern of positivity would have been observed between the
animals. Experimental transmission studies have also failed to induce H5N1 spread between cows
using contaminated milking equipment30. Thus, alternative mechanisms of transmission of H5N1
between cows must exist. The variations in H5N1 quarter positivity may reflect that, and animals
with all 4 quarters positive may be those with H5N1 viremia as reported in a subset of cows31.
While animals with fewer quarters H5 positive may have been from contaminated milking
equipment or infection in damaged teats that have encountered H5N1 virus through aerosols or
contaminated wastewater.
Surprisingly, we observed H5N1 positivity was not always associated with mastitis,
suggesting many subclinical infected cows on a farm whose milk will enter the commercial stream.
Assessment of mastitis-affected quarters in a distinct set of cows from the same farm done prior
to initiation of the longitudinal study identified a clear preference for the FR quarter. In contrast,
analysis of the five cows with only a single H5N1 positive quarter revealed two animals with a
positive FL quarter, one BL, one BR, and one FR. The lack of a consistent quarter infected across
all cows, as observed for mastitis, suggests that these two features are not linked and likely have
distinct modes of transmission.
While field surveillance is critical to understanding ongoing infectious disease outbreaks,
there are many limitations that are important to note when considering the data presented in this
manuscript. First, longitudinal sampling was conducted on a single farm with 14 cows over a short
period of time, which may have missed patterns that could be revealed with a larger number of
animals. Additionally, sampling across multiple farms, for longer periods of time, or earlier in an
outbreak may reveal different patterns. Constraints on time and personnel, shared milking
equipment could not be traced in our study, so patterns of the milking equipment usage and order
of milking for all 14 longitudinal animals are not known and likely varied each day of sampling
routes could include contaminated milking equipment from an infected cow, aerosols generated
within the milking parlor, and/or contact of teats with contaminated water used to clean housing
pens. Multiple mitigation strategies should therefore be implemented to reduce the risk of H5N1
spread within a herd and to humans. Respiratory and ocular personal protective equipment (PPE)
for farm workers to prevent deposition of virus-laden aerosols on these sites, especially in the
milking parlor. Disinfection of milking equipment between each milking of each cow, such as with
consistent use of backflush system, could also reduce spread of H5N1 between cows. Treatment
of milk from sick cows to inactivate H5N1 prior to disposal as well as treatment of waste streams
prior to their use on fields or on farms should also be considered. Finally, identification of infected
cows, regardless of clinical signs, for isolation will help reduce the transmission of H5N1 on farms.
routes could include contaminated milking equipment from an infected cow, aerosols generated
within the milking parlor, and/or contact of teats with contaminated water used to clean housing
pens. Multiple mitigation strategies should therefore be implemented to reduce the risk of H5N1
spread within a herd and to humans. Respiratory and ocular personal protective equipment (PPE)
for farm workers to prevent deposition of virus-laden aerosols on these sites, especially in the
milking parlor. Disinfection of milking equipment between each milking of each cow, such as with
consistent use of backflush system, could also reduce spread of H5N1 between cows. Treatment
of milk from sick cows to inactivate H5N1 prior to disposal as well as treatment of waste streams
prior to their use on fields or on farms should also be considered. Finally, identification of infected
cows, regardless of clinical signs, for isolation will help reduce the transmission of H5N1 on farms.
...
Posted August 01, 2025.
Download PDF
View ORCID ProfileA.J. Campbell, View ORCID ProfileMeredith Shephard, View ORCID ProfileAbigail P Paulos, View ORCID ProfileMatthew Pauly, View ORCID ProfileMichelle Vu, View ORCID ProfileChloe Stenkamp-Strahm, View ORCID ProfileKaitlyn Bushfield, Betsy Hunter-Binns, Orlando Sablon, View ORCID ProfileEmily E Bendall, William J Fitzimmons, View ORCID ProfileKayla Brizuela, View ORCID ProfileGracie Quirk, Nirmal Kumar, Brian McCluskey, View ORCID ProfileNishit Shetty, View ORCID ProfileLinsey C Marr, View ORCID ProfileJenna Guthmiller, Kevin Abernathy, View ORCID ProfileAdam S Lauring, Blaine T Melody, Marlene Wolfe, Jason Lombard, View ORCID ProfileSeema S Lakdawala
doi: https://doi.org/10.1101/2025.07.31.666798
Abstract
Transmission routes of highly pathogenic H5N1 between cows or to humans remain unclear due to limited data from affected dairy farms. We performed extensive air, farm wastewater, and milk sampling on 14 H5N1 positive dairy farms across two different California regions. Virus was detected in the air in milking parlors and from exhaled breath of cows. Infectious H5N1 virus was detected in the air and water streams; sequence analysis revealed viral variants on a farm in these locations. Longitudinal analysis of milk from the individual quarters of cows revealed a high prevalence of subclinical H5N1 positive cows and a heterogeneous distribution of infected quarters that maintained a consistent pattern over time. Our data highlight potential modes of H5N1 transmission on dairy farms.
Competing Interest Statement
SSL and LCM receive funds from Flu Lab and NIH. ASL receives funds from Flu Lab, NIH, and CDC. ASL receives consulting fees and research support from Roche, outside of the submitted work.
...
https://www.biorxiv.org/content/10.1101/2025.07.31.666798v1
----------------------------------------------------------
From the PDF link above:
...
Discussion
Elucidating the routes of transmission of H5N1 between cows is critical to defining
successful mitigation strategies. In this study we successfully detected H5N1 in the air and in
reclaimed farm wastewater on separate dairy farms on multiple days. This included infectious air
samples from three different milking parlors and viral RNA from the exhaled breath of rows of
cows on two distinct farms. Additionally, we detected viral RNA in farm wastewater at multiple
sites on various farms and infectious virus at two different sites on the same farm. Together, these
results highlight the expansive environmental contamination of H5N1 on affected dairy farms and
identify additional sources of viral spread between cows and to humans. Surprisingly, we also
observed an HA mutation in a residue (H5 HA 189, H3 numbering 193) known to alter sialic acid
specificity26 in the air during milking of infected cows, suggesting that HA mutations are emerging
during this outbreak but are not currently maintained. Notably, mutations at 189 are associated
with improved binding to human a2,6-linked sialic acids26 and a structure of a 2.3.4.4b H5 with
an a2,6-linked sialic acid analog shows position 189 helps stabilize receptor binding24. Whether
the N189D mutation observed in this study alters receptor binding and improved tropism to
humans remains to be determined. These data support that emerging viral variants can pose a risk
to humans working in the dairy parlors through exposure to viruses in the air and on surfaces.
Given the high viral loads of H5N1 found in milk, milking equipment has been implicated
as the likely source for viral transmission. However, data from our longitudinal study revealed a
heterogeneous pattern of positivity across the subset of cows analyzed, both in the number and
location of positive quarters. If milking equipment alone was the driver of cow-to-cow H5N1
transmission it is likely that a similar pattern of positivity would have been observed between the
animals. Experimental transmission studies have also failed to induce H5N1 spread between cows
using contaminated milking equipment30. Thus, alternative mechanisms of transmission of H5N1
between cows must exist. The variations in H5N1 quarter positivity may reflect that, and animals
with all 4 quarters positive may be those with H5N1 viremia as reported in a subset of cows31.
While animals with fewer quarters H5 positive may have been from contaminated milking
equipment or infection in damaged teats that have encountered H5N1 virus through aerosols or
contaminated wastewater.
Surprisingly, we observed H5N1 positivity was not always associated with mastitis,
suggesting many subclinical infected cows on a farm whose milk will enter the commercial stream.
Assessment of mastitis-affected quarters in a distinct set of cows from the same farm done prior
to initiation of the longitudinal study identified a clear preference for the FR quarter. In contrast,
analysis of the five cows with only a single H5N1 positive quarter revealed two animals with a
positive FL quarter, one BL, one BR, and one FR. The lack of a consistent quarter infected across
all cows, as observed for mastitis, suggests that these two features are not linked and likely have
distinct modes of transmission.
While field surveillance is critical to understanding ongoing infectious disease outbreaks,
there are many limitations that are important to note when considering the data presented in this
manuscript. First, longitudinal sampling was conducted on a single farm with 14 cows over a short
period of time, which may have missed patterns that could be revealed with a larger number of
animals. Additionally, sampling across multiple farms, for longer periods of time, or earlier in an
outbreak may reveal different patterns. Constraints on time and personnel, shared milking
equipment could not be traced in our study, so patterns of the milking equipment usage and order
of milking for all 14 longitudinal animals are not known and likely varied each day of sampling
routes could include contaminated milking equipment from an infected cow, aerosols generated
within the milking parlor, and/or contact of teats with contaminated water used to clean housing
pens. Multiple mitigation strategies should therefore be implemented to reduce the risk of H5N1
spread within a herd and to humans. Respiratory and ocular personal protective equipment (PPE)
for farm workers to prevent deposition of virus-laden aerosols on these sites, especially in the
milking parlor. Disinfection of milking equipment between each milking of each cow, such as with
consistent use of backflush system, could also reduce spread of H5N1 between cows. Treatment
of milk from sick cows to inactivate H5N1 prior to disposal as well as treatment of waste streams
prior to their use on fields or on farms should also be considered. Finally, identification of infected
cows, regardless of clinical signs, for isolation will help reduce the transmission of H5N1 on farms.
routes could include contaminated milking equipment from an infected cow, aerosols generated
within the milking parlor, and/or contact of teats with contaminated water used to clean housing
pens. Multiple mitigation strategies should therefore be implemented to reduce the risk of H5N1
spread within a herd and to humans. Respiratory and ocular personal protective equipment (PPE)
for farm workers to prevent deposition of virus-laden aerosols on these sites, especially in the
milking parlor. Disinfection of milking equipment between each milking of each cow, such as with
consistent use of backflush system, could also reduce spread of H5N1 between cows. Treatment
of milk from sick cows to inactivate H5N1 prior to disposal as well as treatment of waste streams
prior to their use on fields or on farms should also be considered. Finally, identification of infected
cows, regardless of clinical signs, for isolation will help reduce the transmission of H5N1 on farms.
...
