BIOAEROSOL DISTRIBUTION AND FATE TO ENSURE MICROBIOLOGICAL SAFETY AND QUALITY IN FOOD AND AGRICULTURAL PROCESSING FACILITIES

• Sponsoring Institution: National Institute of Food and Agriculture
• Project Status: COMPLETE
• Funding Source: HATCH
• Reporting Frequency: Annual
• Accession No.: 1018600
• Project Start Date: Jan 17, 2019
• Project End Date: Jan 17, 2024
• Program Code: [(N/A)]- (N/A)

Recipient Organization
TEXAS A&M UNIVERSITY
750 AGRONOMY RD STE 2701
COLLEGE STATION,TX 77843-0001

Performing Department
Biological & Agricultural Engineering

Non Technical Summary
The food industry operates with strict rules to maintain safe and hygienic environments. Recently, aerosolized bacteria have been recognized as a threat to human health and shelf life of food (Salem and Gardner 1994). Air samples taken before and during three separate pork and beef slaughter processes using an Andersen N6 single stage impactor (Sutton, 2004) showed that the total airborne bacterial counts were less than three logs before slaughtering and greater than three logs during slaughtering. Lues et al. (2007) found 104 CFU/m3 microbial counts for potentially pathogenic bacteria and fungi in a chicken slaughtering facility, using a SAS Super 90 air sampler. Higher counts of airborne microorganisms found in the receiving-killing and defeathering areas indicate the importance of controlling microbial levels before processing to prevent the spread of organisms. Due to its clump-like structure, S. aureus was found to be able to readily adhere to surfaces in poultry slaughtering facilities (Lutgring et al., 1997). The air transmission of foodborne bacterial and viral pathogens has been recognized as one of the various factors promoting foodborne disease outbreaks including the Shiga Toxin producing Escherichia coli (STEC) (Cevallos-Cevallos et al., 2012). Although the air has been mentioned as a potential source of bacterial pathogens in produce (Beuchat, 1996), there are few, if any, reports about the prevalence of environmental pathogens such as Listeria monocytogenes (Lm) in bioaerosols formed in the packing environment during postharvest operations. Furthermore, some sanitizing treatments used in produce washing and disinfection may promote the reattachment of Salmonella onto the sanitized product, as reported for cantaloupes (Ukuku, 2006). In a produce packing environment, where air is often freely flowing through the plant, there is the potential for the presence of environmental pathogens such as Lm. Evaluating the suitability of physical and chemical methods as intervention methods to increase safety of leafy greens while maintaining their safety (Omac et al., 2017; Yang et al., 2018), and analyzing the air microbiome in packing plants could help identify the bacterial diversity. Metagenomics of the air was previously conducted for bioaerosols collected in dairy farms in California (Ravva et al., 2011), however, there is no data for produce processing facilities.Heating, ventilation, and air conditioning (HVAC) systems are used in almost every single building in the United States. The food manufacturing industry has the highest amount of non-office floor space and it is also the highest consumer of energy among industries (Kreider, 2000). The problem with HVAC systems, in this case focused particularly on meat and dairy facilities, is the filtration system and direction of airflow. Ventilation has been shown to be effective in eliminating ozone even better when particles, organic and inorganic, build up on the filter (Yu et al., 2009). However, the growing biofilms on untreated filters release bacteria after 27 days (Cecchini et al., 2004), leading to economic losses and the development of diseases (Clausen et al., 2003; Moens, 2002). High humidity in processing facilities fosters bacterial growth. Microbial contaminants, including Staphylococcus aureus, were found in all production areas of a burger plant even though the rooms were physically separated by walls (Byrne et al., 2008). A study by Prendergast et al. (2004) showed that the layout of slaughter facilities is important to airborne bacterial concentration, finding the facility with a straight production line and one level less contaminated.Another study showed that Salmonella in the air contaminates meat products even when stored in low temperatures (Buncic and Sofos, 2012), below 55.4oF.Computational fluid dynamics (CFD) modeling has been used to model the air flow pattern in different built environments. Kermani (2015) found COMSOL Multiphysics software based modeling useful to model air flow pattern in hospital rooms. Faulkner et al (2015) correlated increasing ventilation rates to reduced concentrations of 1.9 µm particles, however, found little effect on larger (5.4 and 7.9 µm) particles. Several authors modeled the effect of air curtains to restrict cold room infiltration in post-harvest facilities (Foster et al., 2007; Ambaw et al., 2012). Although the importance of ventilation in keeping the pathogenic bioaerosols away from processing and packing areas has been recognized for safety (Lutgring et al., 1997), as of now no study has been performed to analyze the HVAC systems of meat or dairy facilities using computational fluid dynamics (CFD) in order to find problematic areas for microbial transport within the built environment and relate that to experimentally found concentrations of bacteria in the facilities.Given the fundamental and practical implications of the ubiquitous HVAC systems and airborne bacteria often encountered in everyday situations (Estrada-Perez et al., 2018), the overarching vision of this proposal is to gain comprehensive understanding of the role of ventilation systems in the fate and transport of aerosols. A great possibility is that if a certain bacteria concentration at location B is known to cause contamination in the facility that would result in a recall, then a minimum allowable concentration at location A can be determined by backtracking through the flow patterns and sampling.The computational air flow models that we create based on the facilities' layout and HVAC design coupled with the bioaerosol concentrations will enable us to visualize the pathogen movement in the facility, and by redesigning the ventilation, optimize the air flow for improved sanitation

Animal Health Component

50%

Research Effort Categories

Basic

20%

Applied

50%

Developmental

30%

Classification

Knowledge Area (KA)	Subject of Investigation (SOI)	Field of Science (FOS)	Percent
404	3440	1100	50%
723	1430	2020	50%

Knowledge Area
404 - Instrumentation and Control Systems; 723 - Hazards to Human Health and Safety;

Subject Of Investigation
1430 - Greens and leafy vegetables; 3440 - Meat, dairy cattle;

Field Of Science
2020 - Engineering; 1100 - Bacteriology;

Keywords

food safety

bioaerosols

wetted wall cyclone

airflow modeling

antibiotic resistance

Goals / Objectives
The research will presently focus on the release and spread of potentially pathogenic and antibiotic resistant bacteria at meat, dairy and produce processing facilities, through on-site bioaerosol collection experiments and ventilation air flow pattern modeling. The work proposed here consists of four specific objectives supporting that overall vision:• Research Objective 1 (RO1): determine the levels and distribution of Shiga Toxin Producing E. coli (STEC), Salmonella spp. (S. enterica serovar Enteritidis, S. enterica serovar Typhi) and Listeria monocytogenes in large air volume samples collected at the different facilities using the wetted wall cyclone (WWC) bioaerosol sampling systems (McFarland et al., 2010), microbial plating and molecular analysis,• Research Objective 2 (RO2): apply computational fluid dynamics (CFD) modeling to simulate the air flow patterns at the selected facilities based on their floorplan and HVAC design to determine their effect on the spread of bioaerosols,• Research Objective 3 (RO3): investigate how building design, aerosolization and environmental stressors affect the viability of the airborne pathogenic bacteria and activate antibiotic resistance genes (ARGs) in their genomes,• Research Objective 4 (RO4): create and validate a universal HVAC/building computational model that can be tailored for any facility based on their parameters to reduce pathogen entrainment and increase the sanitation of the final product.

Project Methods
In the study, meat, dairy and produce processing facilities will be examined. High air volume wetted wall cyclone (WWC) bioaerosol samplers will be used to collect air samples during the entire processing and analyze the bioaerosols for culturable and total bacteria counts, antimicrobial resistance and microbiome composition. Blueprints and HVAC designs obtained from the selected facilities will be modeled using the ANSYS® Fluent software. Similarities between the collected bioaerosol concentrations and particle trace analysis will be determined. Finally, new HVAC models that are adaptable to different layouts will be generated to significantly increase sanitation in the facilities.4.1. Bioaerosol collection and analysis experiments (RO1) 4.1.1. Task 1: Bioaerosol collection: Sampling campaigns for data collection under this objective will be conducted for three days, three times per year (in the spring, summer and winter, to account for seasonal differences (Lejeune et al., 2004; Chapman et al., 2001; Conedera et al., 2001), in Year 1, 2, 3 and 4, for three meat, three dairy and three produce processing facilities. During sampling, environmental conditions will be monitored, including temperature, barometric pressure and relative humidity in the facilities and at the barns. The effect of operational and maintenance practices, including high pressure water spraying, washing and hosing on aerosol concentrations will also be studied and included in the layout model. Two large meat facilities in West Texas and a third, new facility in Kuna, Idaho have been selected for the study, in addition to the Southwest Regional Dairy Center of the Tarleton State University, the Daisy Dairy and the Wolf Brothers Dairy Farms, the Texas International Produce Association (TIPA) and the J&D Produce Inc. in Texas. At each meat establishment, sample collection will be achieved by setting up two WWC collectors to continuously sample air for three consecutive days at the dehiding area and in the fabrication room with an air flow rate of 100 liters per minute and a collection liquid inflow rate of 100 µL/min. Two dynamic samplers will be moved along the processing line, collecting air during the two day period at the five stations where the highest concentrations of aerosolized particles are released, at the hosing/stunning, bleeding, delimbing and dehiding steps, and also in the chiller area. Outdoor samples at the animal entrance location will also be collected in the morning and evening, when the process starts and ends, for 60 min periods each.At the dairy facilities, four collectors will be set up inside by the longer sides of the free-stall barns that are ventilated by axial fans to sample air for two consecutive days. The building orientation, wind speed and the type and location of the axial fans in the barns will be recorded. Additional sampling will be conducted in the milking parlor for the entire period of each milking session to monitor the potential transmission/shedding of bioaerosols to the milk. This may vary from facility to facility; for example, at the SW Dairy Center the cows are milked three times a day and each milking session lasts for 9 min. The liquid bioaerosol sample collections will be removed after 60 min periods for the spatial and temporal measurement and analysis of the bacterial content at each inner sampling location. Duplicate air samples will be taken at each sampling site for further comparison of testing methods. Downwind of the barns, one 60 min (6 m3) bioaerosol sample will be collected at the manure lagoons where the dairy wastewater is collected, to provide background information about antibiotic resistant bacteria. In addition to the bioaerosol samples, manure samples inside the barns and lagoon water samples will also be collected to test for bacteria harboring AR genes.In the field, air sampling will be conducted in two locations, for a total period of six weeks using the autonomous wetted wall cyclone (WWC), at 1250 L/min air flow, with a liquid inflow rate of 350 µL/min, one collector per site. This time span will encompass final crop maturation and harvest. Particle concentrations will be measured using the Aerodynamic Particle Sizer (APS, TSI Inc., Burnsville, MN) and environmental conditions such as wind velocity, wind direction, rainfall, air temperature and relative air humidity will be monitored with portable weather stations. As reference samplers, two Burkard spore traps will be operated at 10 L/min, simultaneously with the WWC collector, for the same period of time, one instrument per field. The fields, one irrigated (Texas AgriLife Experiment Station in Burleson county) and one dryland (Stiles Farm in Williamson county), will be planted with leafy greens (spinach, kale, collards) adapted to the production area, using recommended production practices. There will be two-row plots 30 feet long with 20-foot buffers for each crop, with four replications along row lengths will be delineated in an area of the field that is 600 feet long and 300 feet wide. Within this area, there will be eight plots that are spaced maximally from each other. Sprinkler irrigation with two droplet sizes (coarse and fine) will be randomly assigned to these plots, with blocking along rows and four replicates. There will also be four replicates of the control. Bioaerosol collection in facilities. At the packing facility, sampling areas will be selected to include those with more probabilities for generation of bioaerosols, including but not limiting to, washing and post-washing stations, and to cover the first stages of the process such as produce reception, cold storage and truck loading. The WWC systems at each station will be left on for 3 days and the samples will be split for testing for bacteria by comparative analysis using selective plating and molecular analysis.For more detailed information on 4.1, please see attached proposal4.2. Computational air flow modeling (RO2)As-built blueprint and HVAC design dwg files will be collected from each selected facility after the bioaerosol collection is completed. The facilities will be modeled in the ANSYS® Fluent program using the turbulent kinetic energy k, and turbulent dissipation, κ-ε (k-epsilon) turbulence model and the Reynolds Averaged Navier Stokes (RANS) equation.4.2.1. Task 1: Facility layout design: The HVAC systems will be modeled using the information from the blueprint of the building and the mechanical design. The positions of the workers, collection sites, and locations of importance, including walls, openings, ventilation exhaust and intake locations will be entered in the models. Heat sources, worker and cattle surface areas will be calculated based on the weight and length of the body (Elting, 1926) and added to the model. Relative Humidity (RH) and temperature (T) measurements will be continuously recorded in the facilities during the testing.4.2.2. Task 2: Computational air flow pattern modeling and validation: Collection sites will be correlated with the airflow by simulating the intakes of the WWC collectors at each testing site for simulated particle collection. Other metrics will also be analyzed such as velocity profiles and local mean age of air (LMA).4.3. The effect of aerosolization on the viability and antibiotic resistance genes of bacteria (RO3)The WWC bioaerosol samples collected at the facilities outdoor and indoor will be screened for antibiotic resistance bacteria (ARBs) and antibiotic resistance genes (ARGs).For more detailed methods on 4.3 and 4.4, please see attached proposal.

Progress 10/01/19 to 09/30/20

Outputs
Target Audience:The following facilities were contacted during this reporting period (October 1, 2019 - September 30, 2020): 1. Meat processing facility (5): - Rosenthal Meat Center at Texas A&M University (referred to as Facility A) - Rural, family owned meat processing facility in TX (referred to as Facility B) - Large, privately owned meat processing facility in West Texas (referred to as Facility C) -Large, corporate owned meat processing facility in West Texas (referred to as Facility D) - Large, privately owned meat processing facility in North America(referred to as Facility E) 2. Dairy facility (1): -Southwest Regional Dairy Center, Stephenville, TX 3. Produce processing facility (1): -J&D Produce, Edinburgh, TX Additional Agences contacted: 1. North American Meat Institute (NAMI) 2. USDA Food Safety and Inspection Service (FSIS) 3. Beef Cattle Research Council (BCRC) 4. Canadian Meat Council (CMC) Changes/Problems:Due tothe current Covid-19 pandemic, the proposed projectwill also include the monitoring, measurement and ventilation air flow modeling-based mitigation of virus aerosol particlesat the meat facilities. What opportunities for training and professional development has the project provided?In the project, the team received training in the following professional software programs and instruments: 1. ANSYS Fluent computational fluid dynamics modeling program to simulate air flow patterns in ventilated facilities. 2. ParticleTracking Velocimetryto measure velocities and trajectories of aerosolized droplets. 3. Qiime (Quantitative Insights Into Microbial Ecology) - open source bioinformatics software for microbiome analysis. 4. EDEM program for discrete element method to simulate the behavior of bulk materials. 5. Three-Dimensional (3D) Anemometry to measure air velocities in large open spaces including dairy facilities. How have the results been disseminated to communities of interest?The results of the research werepresented to the scientific community, todifferent agencies and industryin the following ways: 1. Publications in peer reviewed journals (listed in this report). 2. Platform and poster presentations at scientific conferences (listed in this report). 3. Two international webinars organized by the North American Meat Institute, the Canadian Meat Council and the Beef Cattle Research Council, for governmental agencies and meatindustry. 4. Meetings and presentations to meat industry facility owners and stakeholders. What do you plan to do during the next reporting period to accomplish the goals?The following activities are planned for the next reporting period: 1. Computational flow models will be created for three ventilated meat facilities based on their mechanical floorplans. 2. As the current pandemic regulations allow, field testing will be performed at the meat facilities to collect bioaerosols at specific locations in the facilities and validate the flow models. 3. Bioaerosol testing at the facilities will be expanded to include the SARS-CoV-2 virus collection, analysis and tracking. 4. Mitigation measures including air curtains and roofed plastic dividers will be tested at different facilities for reduced bioaerosol entrainment and increased sanitation. 5. Controlled laboratory experiments will be conducted with bacteria using particle tracking velocimetry to monitor droplet characteristics of bioaerosols and the effect of the different environmental factors (temperature, humidity, pressure, velocity) on the development of antibiotic resistance. 6. Genomic analysis will be performed to delineate the genes that are involved in the development of antibiotic resistance in aerosolized bacteria at the molecular level.

Impacts
What was accomplished under these goals? Impact of the research Our goal is to answer fundamental questions about the spread of infectious bacteria and viable viruses in indoor environments using our high air flow viable wetted wall cyclone (WWC) bioaerosol sampling systems and computational air flow modeling technologies. The outcome of this research has the potential to protect the health of meat facility workers through continuous monitoring of infectious aerosol concentrations and their movement, with sufficient throughput to detect dynamic changes in concentration levels in room-size spaces. The potential for efficient detection of viable bioaerosols and mitigation of their spread is assessed by the following objectives. Results for Research Objective 1 (RO1): Every year, millions of food-related illnesses are being reported in United States, many of which are caused by Shiga toxin-producing Escherichia coli (STEC) and Salmonella. They are introduced into food products at various stages during processing, especially in meat processing facilities. Spraying and dehiding of beef carcasses generate bioaerosols, which are then transported throughout the facilities along the air flow from the heating, ventilation, and air conditioning (HVAC) system. Molecular analysis of the aerosol samples collected by the WWC collectors at 100 L/min and quantitatively identified by quantitative polymerase chain reaction (qPCR) revealed 126 GCN/m3 (genomic copy numbers per cubic meter of air collected) of STEC and 104 GCN/m3 of Salmonella using stx gene and invA gene, respectively, in the dehiding rooms of the large meat Facility C during spring. Higher concentrations of the pathogenic bacteria were observed in summer samples where 2,198 GCN/m3 of stx and 3,194 GCN/m3 of invA genes were detected in the dehiding room. In summer, 207 GCN/m3 of stx and 460 GCN/m3 of invA were found in samples from the chiller and fabrication room, respectively. Microbiome analysis with Illumina sequencing. In comparison with qPCR data, total sequence reads correlated with total bacteria counts (TBC) as highest number of reads in spring were in dehiding rooms and the numbers increased in summer dehiding room samples. Illumina sequencing detected more sequence reads from Enterobacteriaceae family in fabrication room and chiller in spring and summer samples, in contrast to qPCR. It could be that other genera from the Enterobacteriaceae family not identified by stx, invA, or eae genes were present in fabrication room and chiller. This indicates that airborne bacteria from unclean rooms are being introduced into clean rooms. Results for Research Objective 2 (RO2): Air property measurements. In summer, the mean temperature and relative humidity values recorded on dataloggers in dehiding room were around 26 °C and 78 %, respectively, while mean measurements in fabrication room were recorded as 9.5 °C and 66 %. In spring, temperature in dehiding room gradually increased while relative humidity decreased throughout the day, however, they were both consistent in summer. High pressure hosing generated higher bioaerosol concentrations of TBC, STEC, and Salmonella in the dehiding rooms compared to other locations in Facility C. Higher temperature and contamination levels in the summer increased the frequency of water spraying in dehiding room, increasing relative humidity and aerosolization rate, and hence increasing the survivability of aerosolized bacteria. Computational fluid dynamics (CFD) modeling. Velocity profiles in the facility room openings were observed to be nearly uniform from the floor to the ceiling. Without the presence of any filters or barriers, bioaerosols generated from the dehiding room were disseminating into the chiller to the walls where carcasses would be hanging and stored. Fabrication room, in turn, was exposed to pathogenic bacteria flowing in from the chiller. Results from qPCR analysis of aerosolized bacteria revealed high counts of TBC at the dehiding and tripe rooms in spring, which were also detected in the chiller and fabrication room at a smaller amount. Similarly, STEC and Salmonella detected in dehiding rooms were also present in the chiller and fabrication room. Air flow simulation conducted in ANSYS Fluent program revealed that bioaerosols can be transported without disturbance from the dehiding rooms to the chiller and fabrication room as the connecting passageways were all open. As a possible mitigation solution, the effect of air curtains and plastic dividers on bioaerosol dissemination was examined using CFD simulation. Experimental air velocity measurements of a commercial air curtain in a model chamber illustrated that barriers of air streams were created by the air curtain that separated air between two rooms. Computational airflow models showed that 3 ft wide roofed plastic dividers installed between workers significantly reduce the number of aerosolized pathogens in the fabrication room of a North American meat plant. Further research is underway to prevent pathogenic and fecal microorganisms and infectious viruses from being introduced to processed food products. Results for Research Objective 3 (RO3): Antibiotic resistance poses a threat in many institutions such as hospitals, meat processing facilities and other agricultural and occupational entities. The goal of this research is to uncover how bacteria react to certain environmental conditions and delineate the factors that may trigger the expression of antibiotic resistance genes (ARGs) in Escherichia coli (E. coli) MG 1655. These conditions include temperature, relative humidity, air flow and duration of aerosolization. Fresh bacterial suspensions were aerosolized for increasing periods into a large sterile box at a constant airflow of 80 L/min and collected using the WWC bioaerosol collector that maintains the culturability of the collected bacteria. Each sample was analyzed and compared based on its response to each environmental factor by the Kirby Bauer susceptibility testing and qPCR using specific primers to target different ARGs (marR, and rfaC). The susceptibility test results indicated the strongest resistance response to antibiotics that are cell wall or DNA replication inhibitors, respectively. There was approximately 100% more resistance detected in E. coli aerosolized for 5 min versus 45 min, however, similar resistance was detected in the 5 minute and 30 min aerosolization which indicated that exposure for 45 min causes increased stress for the bacteria. Interestingly, no resistance was detected in bacteria aerosolized for 10 min, maybe due to a transient response of the cells to compensate for the osmotic and other stresses during aerosolization. The geographical orientation of free stall barns at dairy facilities may also play an important role in the development of antibiotic resistant bacteria. Manure samples collected evenly along the two sides of the barn showed increased numbers of AR bacteria collected on the shady side of the barn indicating stress in the bacteria growing on the darker and colder side of the barn that receives direct sunlight only in the afternoon. This information may help mitigate antibiotic resistance in aerosolized bacteria by optimizing environmental conditions to prevent the development of ARGs. Results for Research Objective 4 (RO4): Based on the main mechanical layout parameters and air properties common in most processing facility layouts a universal HVAC/building computational model was created that can be tailored for any facility based on their unique and measured parameters to reduce pathogen entrainment and increase the sanitation of the final product. The model has been customized based on the layout and air properties of a large meat plant and it is currently being experimentally tested in Facility E. Our results were presented in two international webinars, and met with large interest from governmental agencies, health authorities and industry.

Publications

• Type: Journal Articles Status: Published Year Published: 2020 Citation: 1. King, M.D., Lacey, R.E., Pak, H., Fearing, A., Ramos, G., Baig, T., Smith, B., and Koustova, A. (2020). Assays and Enumeration of Bioaerosols-Historical Approaches to Modern Practices; In Bioaerosol Research: Methods, Challenges, and Perspectives. Aerosol Science and Technology, 54 (5): 611-633. https://doi.org/10.1080/02786826.2020.1723789
• Type: Journal Articles Status: Published Year Published: 2020 Citation: 2. Mohammad, Z., Beck, S., King, M., Griffin, D. and Castillo, A. (2020). Comparison between the real-time PCR and Crystal Diagnostic Xpress immunoassay-based method for detecting Salmonella and Shiga toxin-producing Escherichia coli (STEC) in the air of beef slaughter establishments, Journal of Food Protection, In Press, https://doi.org/10.4315/JFP-19-616
• Type: Journal Articles Status: Published Year Published: 2020 Citation: 3. Fearing, A. Kalbasi-Ashtari, A. Zuniga, A., Pak, H., Haglund, J., Kim, H.Y. and King, M.D. (2020). Performance of Two Shrouded Probes for the Collection of Liquid Aerosols in a Wind Tunnel Optimized for High Air Speeds. Aerosol Science and Technology, 54(8): 972-982. https://doi.org/10.1080/02786826.2020.1753880
• Type: Journal Articles Status: Published Year Published: 2020 Citation: 4. Yu, X., Kish, L., Seguin, J. and King, M. (2020). Ternary Fingerprints with Reference Odor for Fluctuation-Enhanced Sensing. Biosensors, 10(8): 93. https://doi.org/10.3390/bios10080093
• Type: Journal Articles Status: Published Year Published: 2020 Citation: 5. Liu, C., Shi, Y., Wang, T. and King, M.D. (2020). Design of thermal wind sensor with constant power control and wind vector measurement method. PLoS ONE, 15(4): e0231405. doi.org/10.1371/journal.pone.0231405
• Type: Journal Articles Status: Published Year Published: 2020 Citation: 6. Mulgaonkar, N., Wang, H., King, M. and Fernando, S. (2020). Druggability Assessment of precursor membrane protein as a target for inhibiting the Zika virus. Journal of Biomolecular Structure & Dynamics, In Press, https://doi.org/10.1080/07391102.2020.1851304
• Type: Journal Articles Status: Submitted Year Published: 2020 Citation: 7. Pak, H-M., Zuniga, A.A., Alejandro Castillo, A. and King, M.D. Air flow modeling and mitigation of bioaerosol transport in meat processing facilities. Submitted to Scientific Reports
• Type: Conference Papers and Presentations Status: Published Year Published: 2019 Citation: 1. Ramos, G., Venkateshaiah, B., Gerlich, R., Jantrania, A. and King, M.D. Development of Antimicrobial Resistance in Bioaerosols. AAAR 37th Conference, October 14-18, 2019, Portland, OR. Platform presentation.
• Type: Conference Papers and Presentations Status: Published Year Published: 2019 Citation: 2. Pak, H-M. and King, M.D. Tracking the Movement of Antibiotic Resistant Genes in Dairy Farms using Computational Fluid Dynamics. AAAR 37th Conference, October 14-18, 2019, Portland, OR. Poster presentation.
• Type: Conference Papers and Presentations Status: Published Year Published: 2019 Citation: 3. Zuniga, A. and King, M.D. Dynamic Monitoring and Displacement Ventilation to Prevent Pathogen Spread During Meat Processing. AAAR 37th Conference, October 14-18, 2019, Portland, OR. Poster presentation.

Progress 01/17/19 to 09/30/19

Outputs
Target Audience:The following facilities were contacted during this reporting period (January - September 2019): 1. Meat processing facility(3): -Rosenthal Meat Center at Texas A&M University (referred to as Facility A) -Rural, family ownedmeat processing facility in TX (referred to as Facility B) -Large, privately ownedmeat processing facility in West Texas (referred to as Facility C) 2. Dairy facility(1): -Southwest Regional Dairy Center, Stephenville, TX 3. Produce processing facility (1): -J&D Produce, Edinburgh, TX Changes/Problems: Nothing Reported What opportunities for training and professional development has the project provided?The PI's goal is to become an expert in the interdisciplinary field of aerobiology and a successful educator in biological engineering. In pursuit of this goal, the PI's longterm research goal is to gain deeper understanding of the effect of aerosolization on airborne pathogens at the genome level. This project serves as the foundation to initiate a fruitful collaboration between the PI and the meat, dairy and produce industry for the development of an engineering solution that will help the industry reduce the spread of resistant airborne pathogens and improve public health. The insights gained have also been added to the class curriculum and more students are being trained in these advanced technologies for food safety. As a member of the American Association for Aerosol Research (AAAR) and the American Society of Agricultural and Biological Engineers (ASABE), the PI is committed to raising awareness of the development and spread of antibiotic resistance in aerosolized bacteria and the potential health effects of air conditioning which is an essential part of modern life. The PI's complementary educational goals are to advance the intellectual development of the students and encourage them to think out of the box. As a faculty advisor and laboratory director, the PI is also deeply committed to serving as mentor for underrepresented minorities including women in STEM (science, technology, engineering and math). This project helps the PI fulfill the goals and establish a successful, dynamic research and teaching career. How have the results been disseminated to communities of interest?The results of this project have been disseminated in the following ways: 1. Publications and manuscripts submitted to scientific journals including the ASABE journal "Applied Engineering in Agriculture", Aerosol Science and Technology and Journal of Meat Science, 2. Platform and poster presentations at research conferences including the American Association for Aerosol Research (AAAR) and American Society of Agricultural and Biological Engineers (ASABE) annual meetings, 3. Research Showcase with industry partners, 4. Meetings and discussions with collaborating industry partners, 5. The methods and results that are developed in this project have been added to the undergraduate curriculum and the graduate program. What do you plan to do during the next reporting period to accomplish the goals?Further studies will incorporate larger bioaerosol sample sizes with more facilities, specifically large abattoirs with a high cattle throughput. The blueprints of the facilities will be received first and the HVAC airflow will be modeled based on the mechanical design and layout. Then, upon inspection of the modeled flow, specific points of interest will be chosen for collection inside the facility. Temperature, humidity, airflow velocity and direction data will be taken at every collection site as well as outside of the facility. Furthermore, bioaerosol samples will be collected also at the exhaust of the beef facilities. The large air volume bioaerosol collections will be continued at the dairy and produce facilities during the processing periods. Additional microbial and molecular methods will be developed to include additional pathogens (L. monocytogenes) and their ARGs in the analysis. The bioaerosol samples will be collected at various test points at the different facilities to further validate how bacteria and antibiotic resistance genes (ARGs) are disseminated via airflow. Microbiome analysis using Illumina sequencing and the Qiime program will be conducted to delineate the microbiome composition of the samples. A new 3D ultrasonic anemometer will be used in the dairy barn and at the produce processing facility for the measurement of wind velocity, direction, temperature and relative humidity data to validate the airflow model. Based on the airflow pattern models and bioaerosol movement the most optimal air intake/exhaust design will be selected for each facility that results in the highest sanitation (i.e. the least number of pathogens) in the airflow. A "universal" model template will be selected from the airflow models that can be used to model the airflow to different facility designs and HVAC parameters. The long-term goal of this research is to develop tools for the validation and optimization of these innovative technologies designed to mitigate and reduce food safety risks.

Impacts
What was accomplished under these goals? Research Objective 1 (RO1) Meat facility bioaerosol sampling. The results show that the concentrations of bioaerosols collected with the wetted wall cyclone (WWC) collectors are not only dependent on the location of the physical procedures of processing the carcasses, but also highly dependent on the movement of air in the facility. STEC and Salmonella were found aerosolized inside of all the facilities. Contrary to many studies, a dynamic WWC unit placed near the de-hiding area in the medium-size Facility A with eight cattle processed per day collected the least amount of bioaerosols. However, the continuous WWC placed on the opposite side of the de-hiding area collected the second highest concentration of bioaerosols with a significantly different microbiome. Upon inspection of the flow trajectories, it was apparent that the dynamic WWC was "upstream" of the airflow that was directed towards the de-hiding area, while the continuous WWC unit was downstream of the airflow, and was subsequently able to collect all the particles that were becoming aerosolized. Furthermore, the concentrations of total bacteria count (TBC) and STEC significantly increased between the morning and afternoon phases of processing. The highest concentration of bioaerosols in Facility A were collected at the de-limbing area, which also had the highest local mean age of air (LMA). In the small rural Facility B with two cattle processed per day the location of sampling for the bleeding and de-limbing process was the same. The largest bacteria concentration was collected near the freezer, with a concentration of 10^5.8 OTUs/m3, where OTU denotes operational taxonomy units. The least amount of bioaerosols was collected during the washing phase, with a concentration of 10^1.0 OTUs/m3. The total bacteria collected for the entire morning session was 10^5.8 OTUs/m3. This concentration was significantly higher than the morning average of 10^4.46 OTUs/m3 collected at Facility A, and still higher than the average afternoon concentration of 10^5.4 OTUs/m3. This is likely due to the very small slaughtering area and that the only air movement was caused from an axial fan that pointed towards the carcass. STEC was also found in the facility. In the large Facility C with 3000 cattle processed per day the concentration of the collected bioaerosols was dependent on the physical location of the slaughtering process and on the airflow movement in that area. Higher bioaerosol concentration was found in the summer compared to the spring. In the spring data collection, the highest concentrations of TBC were found in the dehiding area. In the summer in the same area, TBC was found to be four magnitudes higher. Similarly, for Salmonella the spring counts were very low, while in the summer they were three magnitudes higher. Dairy facility bioaerosol sampling. Bioaerosol, manure and lagoon samples that were collected at the dairy center during two seasons (spring and summer) are currently being analyzed for total bacteria concentration, pathogen content, antibiotic resistance genes and microbiome composition. Produce processing facility bioaerosol sampling. A large produce processing facility in South Texas has been visited recently to discuss a field study that will include large air volume bioaerosol collection using the WWC during the processing of fresh produce and the microbial and molecular analysis of the samples for Salmonella, STEC and Listeria monocytogenes. Research Objective 2 (RO2) Meat facility Computational Fluid Dynamics (CFD) modeling. CFD was utilized to analyze the flow patterns and track aerosolized particles at the different facilities. Using velocity point profiles at the inlet location of the WWC in the CFD model of Facility A, it was also noticed that the dynamic WWC in the de-limbing area had the largest negative velocity in the z-direction, or vertical direction. This correlates with the continuous WWC near the freezer at Facility B, which captured the highest concentration collected at the Facility B and also had the largest negative velocity in the z-direction compared to the other collectors. Facility B was found to have a correlation between decreasing LMA and increasing TBC collected. In Facility C the model showed that the HVAC airflow moved along the processing steps, from the de-hiding step towards the clean areas. Particle trace analysis in Facility A using particles emitted from three significant locations in the beef slaughter process to determine the contamination removal effectiveness (CRE) values of the heating, ventilation, and air conditioning (HVAC) systems showed that Facility A has a large spread of bioaerosols across the entire floor and near the freezer door. The intake closer to the front of the facility was found to eject air directly on the de-hiding area resulting in loose vortices forming at the front and rear of the facility. Bioaerosols emitted from workers entering from the main hallway into the slaughter floor were found to spread throughout the facility. New HVAC systems designed for both Facility A and Facility B showed displacement ventilation was the most effective to reduce particle spread. The CRE in Facility A was improved to 0.975 and above, and limited particle movement to the rear of the facility drastically. No change was found in the CRE when particle size was increased to 8 μm. The new design in Facility B reduced the number of particles exiting the door into the hallway by utilizing a negative pressure gradient, causing air to flow from the door into the slaughter room. In Facility C, an air curtain was installed as a mitigation technique to redirect the airflow of the facility. The air curtain prevented the HVAC airflow from entering the chiller and fabrication areas from the de-hiding process. Dairy facility CFD modeling. A 3D ultrasonic anemometer will be used at a free-stall dairy farm to collect physical data on temperature, relative humidity, air (wind) velocity and direction tovalidate the model. In addition, the mechanical layout of the barn, the number axial fans and their position, as well as the number of animals, the area and heat released from their bodies are added to the model. Produce processing facility CFD modeling. Based on the facility layout and physical measurements, computational airflow models will be created to optimize the airflow to preventpathogens from being entrained in the airflow and carried to the areas where the clean produce are stored. Air curtains will be modeled at the entrance of the storage rooms that are separated by plastic strips from the washing areas. Research Objective 3 (RO3) Sampling for antibiotic resistant bacteria in the dairy barn. In the dairy center, which has 36 axial fans for evaporative cooling, a Kirby-Bauer test on different bacterial isolates from twenty manure and twenty bioaerosol samples collected along the barn revealed that bacteria located under the first row of fans were more resistant to antibiotics than those in other areas of the farm. At this area, all the manure accumulated when the floor was flooded for cleaning four times a day. Additionally, this area was shaded at the time of sample collection. The simulation of the airflow patterns revealed that the area with high antibiotic resistance was not disturbed by the fan-generated air movement. High humidity, low sunlight, and low air movement seem to promote the growth and spread of antibiotic resistant genes (ARGs), which then can be reintroduced into the dairy center. Research Objective 4 (RO4) Create and validate a universal computational model for ventilation. Based on the datasets that have been collected up-to-date in the meat and dairy facilities, universal, adaptable, modular airflow models with the least entrainment of pathogens are being created focusing on the areas that are most critical to sanitation and food safety and can easily be adapted to each facility based on their blueprints.

Publications

• Type: Journal Articles Status: Published Year Published: 2019 Citation: Beck, S.H., Castillo, A., Kinney, K.A., Zuniga, A., Mohammad, Z., Lacey, R.E., and M. D. King, M.D. (2019). Monitoring of Pathogenic Bioaerosols in Beef Slaughter Facilities Based on Air Sampling and Airflow Modeling, Applied Engineering in Agriculture, 35(6): 21 pages.
• Type: Conference Papers and Presentations Status: Published Year Published: 2019 Citation: Ramos, G., Venkateshaiah, B., Gerlich, R., Jantrania, A., and King, M.D. 2019. Development of Antimicrobial Resistance in Bioaerosols. Abstract 13BA.2 in Proc. 37th AAAR Annual Conference. 2019. Portland OR.
• Type: Conference Papers and Presentations Status: Published Year Published: 2019 Citation: Pak, H-M., and King, M.D. Tracking the Movement of Antibiotic Resistant Genes in Dairy Farms using Computational Fluid Dynamics. Abstract 9BA.23 in Proc. 37th AAAR Annual Conference. 2019. Portland OR.
• Type: Conference Papers and Presentations Status: Published Year Published: 2019 Citation: Zuniga, A., and King, M.D. Displacement Ventilation to Prevent Pathogen Spread during Meat Processing. Abstract 9BA.18 in Proc. 37th AAAR Annual Conference. 2019. Portland OR.
• Type: Conference Papers and Presentations Status: Accepted Year Published: 2019 Citation: Pak, H-M., and King, M.D. Monitoring of antimicrobial resistant bacteria using bioaerosol collection and airflow modeling in a dairy facility. Southwest Dairy Regional Dairy Center (SWRDC) Research Showcase. 2019. Stephenville, TX.
• Type: Theses/Dissertations Status: Published Year Published: 2019 Citation: Dynamic Monitoring and Displacement Ventilation to Prevent Pathogen Spread During Meat Processing. A Thesis by Alexander Zuniga, 2019. Biological and Agricultural Engineering, Texas A&M university