Influenza A(H5N1) Virus Resilience in Milk after Thermal Inactivation

Abstract

Highly pathogenic avian influenza A(H5N1) detected in dairy cows raises concerns about milk safety. The effects of pasteurization-like temperatures on influenza viruses in retail and unpasteurized milk revealed virus resilience under certain conditions. Although pasteurization contributes to viral inactivation, influenza A virus, regardless of strain, displayed remarkable stability in pasteurized milk.

Keywords: Pasteurization; United States; food safety; influenza; milk; respiratory infections; viruses; zoonoses.

Figure 1

Heat treatment of influenza virus in milk. A–C) We diluted influenza A viruses in Opti-Mem control media (Fisher Scientific, https://www.fishersci.com) or commercial off-the-shelf pasteurized whole milk and heat-treated samples of different volumes at the times and temperatures shown; we calculated time from the moment the sample was placed in the heat block. A sandwich design in a heat block ensured uniform temperature exposure. After treatment, we chilled samples on ice for 5 minutes, adjusted them to a final volume of 200 μL, and titrated by TCID₅₀ in MDCK cells (10). Results are shown for reverse genetics wild-type strain A/Puerto Rico/8/1934 (H1N1) (A); Vietnam/1203/04, a reverse genetics virus carrying the H5 hemagglutinin and N1 neuraminidase segments from A/Vietnam/1203/2004 (H5N1) in the background of PR/8/34, with the H5 segment modified with a monobasic cleavage site (Δ) (B); and a field isolate of the wild-type highly pathogenic strain A/turkey/Indiana/3707-003/2022 (H5N1) (C). D, E) A/Puerto Rico/8/1934 (H1N1) strain was spiked on pasteurized (D) and unpasteurized (E) milk samples at the times and temperatures shown. Circles indicate individual measurements; error bars indicate 95% CIs. Light gray shaded area indicates log₁₀ TCID₅₀ value of 1. NS, not significant; TCID₅₀, 50% tissue culture infectious dose.

Figure 2

Stability of influenza A in retail and unpasteurized milk. We diluted influenza A viruses in either Opti-Mem control media (Fisher Scientific, https://www.fishersci.com) (A), retail off-the-shelf pasteurized whole milk (B), or 2 different sources of unpasteurized milk: colostrum milk (C) or mature milk (D). We then incubated 200-μL samples for several days at various temperatures, as shown. We subsequently titrated samples by TCID₅₀ in MDCK cells. We tested 3 strains: PR8 (H1N1), ty/IN/22 (H5N1), and the reverse genetics version of TX/24 (H5N1). Unpasteurized colostrum milk produced during the first few days after birth contains high levels of immunoglobulins and antimicrobial peptides that might have had an effect in decreasing virus survival. PR8 (H1N1), wild-type strain A/Puerto Rico/8/1934 (H1N1); RT, room temperature; TCID₅₀, 50% tissue culture infectious dose; TX/24, wild-type strain A/Texas/37/2024 (H5N1); ty/IN/22, wild-type highly pathogenic strain A/turkey/Indiana/3707-003/2022 (H5N1).