Aerosols have been suspected to transport food pathogens and contaminate fruits and vegetables grown in close proximity to concentrated animal feeding operations, but studies are lacking that substantiate such transport. To monitor the potential transport of bacteria originated from fresh or dry manure through aerosols on a dairy, we identified by 16S rRNA sequencing, bacteria in aerosols collected within 2 to 3 meters from dairy cows at two dairies. Gram-positive Firmicutes were predominant in aerosols from a dairy in Sonoma, California, and surrounded by vineyards, in contrast to sequences of Gram-negative Proteobacteria predominant in aerosols from a dairy in Modesto, California, also surrounded by other dairies. Although Firmicutes represented approximately 50% of the 10 most abundant sequences, aerosols from the Sonoma dairy also contained sequences of Bacteriodetes and Actinobacteria, identified previously with animal feces. While none of the top 10 sequences from fresh or dry manure from Modesto dairy were detected in aerosols, two of the sequences from the phylum Bacteriodetes and one from class Clostridia from fresh manure were detected in aerosols from Sonoma. Interestingly, none of the sequences from dry manure were in the top 10 sequences in aerosols from both dairies. The 10 most abundant sequences in aerosols from the Modesto dairy were all from Proteobacteria and nearly half of them were from genus Massilia, which have been isolated previously from immune-compromised people and aerosols. We conclude that the predominant bacteria in aerosols are diverse among locations and that they do not reflect the predominant species of bacteria present in cow feces and/or in close proximity to cows. These results suggest that the aerosol sequences did not originate from manure. Large volumes of aerosols would be required to determine if bacterial sequences from aerosols could be used to track bacteria in manure to crops grown in proximity.
Citation: Ravva SV, Sarreal CZ, Mandrell RE (2011) Bacterial Communities in Aerosols and Manure Samples from Two Different Dairies in Central and Sonoma Valleys of California. PLoS ONE 6(2): e17281. https://doi.org/10.1371/journal.pone.0017281
Editor: Jack Gilbert, Argonne National Laboratory, United States of America
Received: October 1, 2010; Accepted: January 28, 2011; Published: February 18, 2011
This is an open-access article distributed under the terms of the Creative Commons Public Domain declaration which stipulates that, once placed in the public domain, this work may be freely reproduced, distributed, transmitted, modified, built upon, or otherwise used by anyone for any lawful purpose.
Funding: The work was funded by the U.S. Department of Agriculture, Agricultural Research Service under CRIS project 5325-42000-044. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Aerosols have been suspected to transport food pathogens and contaminate fruits and vegetables grown in close proximity to animal raising operations, but studies are lacking that identify the mechanisms of transport. Concentrated animal feeding operations (CAFO) produce large amounts of waste , thus raising concerns of pathogen transport through aerosols generated by waste processing on-site. The California Central Valley has locations with a high density of CAFOs and this region experiences occasional high winds and dust storms, thus creating the potential for pathogen transmission through aerosols. A mid-size dairy of 1000 cows will produce more than 12,000,000 kg of manure per year  and the manure is usually stored/processed on-site. This increases the risk of transporting bacterial pathogens by a number of mechanisms (water, wildlife, dust) and is of special concern where fruit and vegetable crops are grown nearby. Enteric bacterial pathogens causing much of the foodborne illnesses, including Salmonellae, Campylobacter sp., Listeria monocytogenes, and Escherichia coli O157:H7, have been reported to survive for long periods in manure and manure slurries ,  and a potential exists for their transmission through aerosols.
Aerosol transmission of E. coli O157:H7 was observed in one study with pigs penned 3 to 6 m away from pigs inoculated experimentally in an enclosed room , but transmission was not detected in sheep in a similar experiment. In a recent field study, a marked strain of E. coli K12 was detected 125 m downwind from a rain gun spray application of E. coli-treated pig slurry to grass pasture . Aerosol transport of Salmonella through contaminated dust to turkeys in a simulated holding-shed environment , and significant bacterial population changes in aerosols in poultry houses from pre-flock to late-flock stages, were observed by 16s rRNA analyses .
The viable microflora present in aerosols where food crops are grown are relevant, especially in regions where there is a high density of cattle nearby. Nevertheless, field studies of aerosols from dairy and agricultural production areas are lacking. Thus, we characterized and compared bacterial communities in aerosols collected in close proximity to dairy cows to those present in fresh and dry manure collected on-site. We compared bacterial community diversity by sampling on two different dairies selected for different manure management practices and land usage in the surrounding region (dairies vs. vineyards). We used high volume cyclonic air samplers to collect aerosols and measured the bacterial communities by16S rRNA gene sequencing. Aerobic bacteria were cultured from aerosols also to compare to the 16S community structure.
Materials and Methods
Air and manure samples were collected without unduly disturbing the dairy cows or interrupting any normal animal raising operations followed on the farms.
Bioaerosol and manure samples from dairies
Aerosols, fresh and dry manure samples (Figure 1) were collected from two medium-sized dairies (ca. 800 to 1000 milking cows) located in San Joaquin valley near Modesto, CA and a location near Sonoma, CA. The two dairies are separated by 190 km and were sampled at various intervals during March to September of 2006. The Modesto dairy is located centrally amongst numerous other dairies and orchards; the Sonoma dairy is an isolated dairy surrounded by vineyards. The Modesto dairy has an on-site manure solid-liquid separator and the liquid is collected in two holding lagoons (c.108 liters) equipped with circulating aerators whereas the Sonoma dairy did not use a solid separator or circulators in the lagoon. Dust storms that transport particulate manure are quite frequent in the San Joaquin valley (i.e., Modesto dairy) and historic records indicate wind gusts of 60 to 70 kph have occurred at both locations. Average recorded annual rainfall was 36 cm for Modesto and 60 cm for Sonoma valley.
Blue dots are aerosol sampling locations. Green dots represent the stalls from where fresh manure was collected. Red dots are on-site locations for dry manure. The same color coding is used in Figures 2, 3 and 5 to designate the sample sources for the bacterial taxa.
Aerosols were collected using a portable multi-stage wetted-wall air sampler (SASS 2000 Plus, Research International, Monroe, WA) with an air flow of 270 L/min. The units were placed on stands at 1 m above ground level and 2 to 3 m away from the dairy stalls to minimize the collection of heavy dust particles in the aerosol wash chamber and to maximize the collection of fine particles distributed in the air by cattle movement in the barns. Temperature, wind speed and relative humidity were monitored using a Kestrel 4000 Pocket Weather Tracker (Nielsen-Kellerman, Boothwyn, PA). Wind direction at the time of aerosol samplings at the Sonoma dairy was SSE 139° to 148° and SSW 200°, whereas it was NNE 13° to 26° at the Modesto dairy.
Airborne particles were collected using 5 mL sterile water to wash incoming air, aerosols were collected into a sterile 50 mL tube every 3 min with a 2 min transfer-refill cycle for a total of 30 min for each unit. Four samplers were placed at each location to yield a total collection time of 2 h per location. Samplings were done at 4 locations on the Modesto dairy and 3 locations on the Sonoma dairy (Figure 1) and the aerosol wash-water was pooled from all units and locations for characterization of aerobic bacteria and extraction of environmental DNA.
Fresh manure was collected using a sterile spatula from freshly dropped feces and combined from 5 cows; a total of 5 samples were created this way. For sequencing purposes, the 5 fresh manure samples were combined. Dried manure samples were collected from manure piles located on-site. Both manure and aerosol samples were stored on ice and transported to the laboratory where aerosol samples were processed on the same day for enumeration of aerobic bacteria. Remaining aerosol wash-water and manure samples were stored at −80°C until used for extraction of genomic DNA.
Aerosol wash-water, without further concentration, was used for the enumeration of aerobic bacteria using Petrifilm count plates (3 M, St. Paul, MN) . One-milliliter portions of 10-fold serial dilutions of wash-water in phosphate buffered saline (0.01 M, pH 7.4) were plated and incubated at 25°C for 2 days.
Extraction of DNA
Aerosol wash-waters were filtered through 0.2 µm cellulose membrane filters (13 mm Millipore, Billerica, MA). Filters with aerosol particles were pooled and treated with 200 µL DNAzol Direct (Molecular Research Center, Inc, Cincinnati, OH) in 1.5 mL centrifuge tubes and heated to 95°C for 20 min. Tubes were centrifuged at 10,000×g for 5 min and 100 µL of the supernatant containing DNA was used without any further purification for amplification of 16S rRNA.
Duplicate half-gram manure pellets (both fresh and dry) were extracted using the alternate protocol of the MoBio UltraClean soil DNA isolation kit (MoBio, Solano Beach, CA) and the wash steps designated 15 to 21 in the protocol were repeated to remove a “green tinge” from DNA extract.
DNA was quantified by the Quanti-iT Picogreen dsDNA assay (Invitrogen Corporation, Carlsbad, CA). Purified DNA from aerosol and manure samples was suspended at a concentration of 1 to 5 ng/µL in UltraClean PCR water.
Amplification of 16S rRNA
A 236-bp fragment of the 16S rRNA gene was amplified using eubacterial primers PRBA338f (5′ ACTCCTACGGGAGGCAGCAG 3′) and PRUN518r (5′ ATTACCGCGGCTGCTGG 3′) . Briefly, genomic DNA from environmental samples was amplified in 10 replicates of 50-µL reaction mixtures containing 5 to 50 ng of template DNA, 300 nM of each primer and two Illustra PuReTaq Ready-To-Go PCR Beads (GE Healthcare, Piscataway, NJ). PCR reactions were performed using DNA engine DYAD® Peltier Thermal Cycler (Bio-Rad Labs., Hercules, CA) using the protocol: one cycle of 92°C for 2 min; 30 cycles of 92°C for 1 min, 55°C for 30 s, and 72°C for 1 min; and one cycle of 72°C for 15 min. PCR products were pooled from all 10 reactions and purified with DNA Clean and Concentrator-5 kit (Zymo Research, Orange, CA) by using 5 volumes of DNA binding buffer to each volume of PCR product mixture as per manufacture's instructions.
16S rRNA clone libraries
PCR products were cloned using TOPO TA cloning kit as per manufacture's instructions and transformed into E. coli TOP10F' One Shot competent cells (Invitrogen, Carlsbad, CA). Clones were plated on LB agar plates containing kanamycin (50 µg/mL), isoprophyl-β-D-thiogalactopyranoside (IPTG, 20 mM), and 5-bromo-4-chloro-3-indolyl-β-D-galactopyranoside (X-gal, 80 µg/mL). Two PCR reactions were performed for each sample, and 96 clones were picked from each PCR to minimize potential PCR bias. Procedures for preparation of DNA templates using the Illustra Templiphi 100 amplification kit (GE Healthcare), sequencing reactions using the Big Dye Terminator v3.1 cycle sequencing kit (Applied Biosystems, Foster City, CA), cleaning sequencing reactions with DyeEx 96 kit (Qiagen, Valencia, CA) and electrophoresis with an Applied Biosystems 3730 DNA analyzer were same as described earlier , except that we used the reverse primer PRUN518r for unidirectional sequencing reactions.
DNA sequence analysis and dendrogram construction
DNA sequences were analyzed manually for bases called incorrectly, then trimmed at both 5′ and 3′ ends using Kodon (v.3.5, Applied Maths, Inc., Austin, TX). Only sequences with unambiguous reads were compiled into data sets and were aligned with the closest sequence relatives from GenBank database by using Kodon. Phylogenetic trees of the 16S rRNA sequences were constructed by using a neighbor-joining method incorporating Jukes-Cantor distance correction. Tree stability was assessed by bootstrap analysis with 1,000 iterations. The phylogenetic relationships between the characterized sequences were analyzed by Bionumerics (v.6.5, Applied Maths) to identify closely related genotypes. Taxonomic assignment of 16S rRNA sequences was performed using the RDP Naïve Bayesian rRNA Classifier Version 2.2 of the Ribosomal Database Project .
Diversity indices and library comparisons
Rarefaction curves of clone libraries from each sample were generated using operational taxonomic units (OTUs) with >97% 16S rRNA sequence similarities with GenBank accessions. Rarefaction analysis with 95% confidence levels was performed using aRarefactWin by Holland (Analytic Rarefaction v.1.3; S. Holland, Stratigraphy Lab, University of Georgia, Athens; http://uga.edu/strata/software/Software.html). Species richness and diversity were determined for each library of OTUs using Simpson's and Shannon-Wiener index of diversities (Bionumerics). Library Compare tool (http://rdp.cme.msu.edu/comparison/comp.jsp) of the Ribosomal Database Project  was used to determine if the library of OTUs from aerosols differ significantly at any hierarchial taxon level from OTU libraries created from fresh or dry manure collected from the same dairy (Figure S1). OTU libraries created from aerosols from both dairies were also compared (Figure S1).
Nucleotide sequence accession numbers
Representative DNA sequences of the top 10 bacterial sequences from aerosols, fresh and dry manure from both dairies were submitted to GenBank under accession numbers HQ340004 to HQ340063. The GenBank accession numbers corresponded to sequences S1 to S30 for sequences originated from the Sonoma dairy and M1 to M30 for sequences from the Modesto dairy.
Bacterial communities in aerosol and manure samples from Sonoma dairy
A total of 495 clones representing 6 different bacterial phyla were characterized from aerosol, fresh and dry manure samples collected from the Sonoma dairy (Table 1). Sequences of Gram-positive Firmicutes predominated in aerosol and fresh manure samples and represented more than half of all sequences, compared to Gram-negative Proteobacteria predominating in the bacterial communities in dry manure. Gram-negative Bacteriodetes were the second most abundant sequences identified for both fresh and dry manure samples. The 10 most abundant sequences comprised 18%, 35% and 28% of the total cloned sequences from aerosol, fresh and dry manure samples, respectively. Similar to the total cloned sequences, Firmicutes represented the highest percentage of the top 10 list (Table 2) of aerosol and fresh manure samples, whereas Proteobacteria topped the dry manure communities.
Phylogenetic analysis of the 10 most abundant sequences indicates that members of Clostridia followed by Bacteroidia predominated in fresh manure (Figure 2). Identical sequences of members of Bacteroidia and Clostridia that were present in fresh manure were detected also from aerosols. These Clostridia had >94% sequence similarity with Ruminococcus sp. (S11 to S12; Figure 2) and Clostridium sp. (S13 and S14). However, all other sequences were unique to the source and were not found in other samples; 59% of all the top sequences were of fecal origin (Figure 2). Sequences of Treponema sp. previously characterized from cow feces constituted 12% of the top 10 sequences from fresh manure. Members of Marinobacter and Halomonas isolated predominantly from marine and saline habitats were exclusively detected from dry manure but not in fresh manure or aerosols (P<0.001).
Identity or the original source of uncultured bacteria and the accession numbers are given. Numbers in parenthesis following the previous isolation source are number of clones characterized for the sequence. Numbers following the clones are identifiers for submissions to GenBank and sequence numbers S1 to S30 correspond with GenBank accession numbers HQ340004 to HQ340033. Bootstrap values >70% are noted at each node. Numbers following the Classifier assignment to the hierarchial taxa are percent reliability values for the classification.
Bacterial communities in aerosol and manure samples from Modesto dairy
A total of 471 clones representing 7 different bacterial phyla were characterized from aerosol, fresh and dry manure samples collected from the Modesto dairy (Table 1). Firmicutes predominated both fresh and dry manure samples, whereas Proteobacteria predominated the bacterial communities in aerosols representing 78% of all sequences from aerosols. Members of Bacteriodetes and Actinobacteria were the second most abundant sequences identified from dry and fresh manure samples, respectively. Chloroflexi and Gammatimonadetes represented 11% and 5% of all sequences from dry manure.
Chloroflexi, Actinobacteria and Proteobacteria sequences were represented equally in the top 10 abundant sequences from dry manure, compared to Firmicutes and Bacteriodetes in fresh manure; Proteobacteria was the only phylum detected in aerosols (Table 2). Members of β-Proteobacteria constituted 88% of the top 10 sequences from aerosols (Figure 3) and many of the sequences identified previously were associated with pollutant degradation. The 10 most abundant sequences from aerosol or manure samples were unique to the source and none of the top 10 for manure was detected in aerosols collected on the dairy. At least a third of the 10 abundant sequences were of fecal origin. All Firmicutes from the top 10 sequences in fresh manure were Clostridia that were previously isolated from animal feces (Figure 3).
Identity or the original source of uncultured bacteria and the accession numbers are given. Numbers in parenthesis following the source are number of clones characterized for the sequence. Numbers following the clones are identifiers for submission to GenBank and sequence numbers M1 to M30 correspond with GenBank accession numbers HQ340034 to HQ340063. Bootstrap values >70% are noted at each node. Numbers following the Classifier assignment to the hierarchial taxa are percent reliability values for the classification.
Rarefaction, species diversity and community comparisons from both dairies
Rarefaction estimates (Figure 4) using OTUs that shared >97% sequence similarity with GenBank accessions indicated that the clone libraries created from each environmental sample were not sufficient enough to fully explore the species diversity as indicated by the slope of the curves. However, species coverage estimates as high as 83.9% (Table 3) were obtained for bacterial communities in aerosols from Sonoma. Species diversity estimates indicate that aerosol communities from Sonoma were rich and even as compared to aerosol communities from the Modesto dairy (Table 3). Both Shannon-Wiener and Simpson's diversity index calculations yielded similar results. Drying the manure appeared to have reduced the species diversity at the Sonoma dairy and aerosolization further reduced the species diversity at the Modesto dairy.
A total of 410 sequences representing 229 OTUs were used for rarefaction of clone libraries from Modesto (top panel), whereas 440 sequences representing 259 OTUs from Sonoma (bottom panel) were used.
Sequences that belong to 19 different bacterial classes were present in manure and aerosol samples from both dairies (Figure 5). Comparison of 16S rRNA sequences at OTU level using Library Compare indicate that members of class β-Proteobacteria that were predominant in aerosols (P<0.001) from Modesto did not originate from either fresh or dry manure. On the contrary, Clostridia were abundant in aerosols from Sonoma and members of Lachnospiraceae and Ruminococcaceae could have originated from fresh manure but none derived from dry manure (P<0.001). Members of Clostridia followed by Bacteroidetes were predominant in fresh manure from both dairies, but none of the members of class Bacteroidia in aerosols were from fresh manure (P<0.001). Actinobacteria followed by Bacilli dominated the bacteria in dry manure from the Modesto dairy, whereas γ-Proteobacteria and Flavobacteria were favored in dry manure from the Sonoma dairy (Figure 5). Members of class Actinobacteria characterized from aerosols of Modesto could have been derived from dry manure (P = 0.8) but not from fresh manure. Again, these data indicate the differences in the bacteria detected at different sites and manure compared to detectable bacteria in aerosols.
Top panel are sequences from Modesto dairy. Bottom panel are sequences from Sonoma dairy. Phylum class designations: FB = Firmicutes-Bacilli; FC = Firmicutes-Clostridia; FE = Firmicutes-Erysipelotrichi; PA = α-Proteobacteria; PB = β-Proteobacteria; PD = δ-Proteobacteria; PG = γ-Proteobacteria; BB = Bacteroidetes-Bacteroidia; BF = Bacteroidetes-Flavobacteria; BS = Bacteroidetes-Sphingobacteria; AA = Actinobacteria-Actinobacteria; CC = Chloroflexi- Caldilineae; CD = Chloroflexi-Dehalococcoidetes; CT = Chloroflexi-Thermomicrobia; GG = Gemmatimonadetes-Gemmatimonadetes; SS = Spirochaetes-Spirochaetes; TM = Tenericutes-Mollicutes; and VO = Verrucomicrobia-Opitutae.
Comparison of aerosol bacterial communities from spatially separated dairies
Although the dairies are located 190 km from each other, we compared the aerosol bacterial communities for any similarities. The sequence similarities with GenBank sequences are significant; some of the 10 predominant sequences were assigned to >94% to genus level (Figure 6) by the Classifier software. Except for Burkholderia cepacia that was present in aerosols collected from both dairies all others were unique to each dairy (Figure 6). Taxonomic diversity of aerosol communities from Modesto was very low as 100% of the top ten were Proteobacteria and 42% of them belonged to genus Massilia; members of this genus were isolated previously from humans, polluted soils or air samples, but none of them were detected in manure from either dairy (P<0.001). In addition, 45% of all aerosol sequences were represented by the top 10 sequences. In contrast, taxonomic diversity of aerosol communities from the Sonoma dairy was very high and the clones identified representing the top 10 sequences (Figure 6) were only 18% of all sequences. Members of Bacteriodetes, Firmicutes and Actinobacteria represented 78% of the 10 abundant sequences of aerosols from Sonoma (Table 2) and almost half them were Clostridia (Figure 6).
Identity or the original source of uncultured bacteria and the accession numbers are given. Sequence numbers (Sq no.) are GenBank identifiers for sequences from Sonoma dairy (S) and Modesto dairy (M). Numbers in parenthesis following the source are number of clones characterized for the sequence. Bootstrap values >70% are noted at each node. Classifier assignment to the hierarchial taxa along with percent reliability values are given.
Culturable aerobic bacteria in aerosols and climate conditions
Aerosol-wash water samples used for bacterial community analysis from both dairies contained the same level of aerobic bacteria; the counts were 130±80 CFU/L of air per hour. In addition, aerobic counts from the Sonoma dairy remained the same from air samples collected on four different occasions during the months of May to August 2006 and the counts were 140±66 CFU/L of air per hour. Average temperature during samplings was 25±3°C and increased from 21±4°C at the beginning of samplings to 29±3°C towards the end. Wind speeds at the time of samplings averaged 3.5±1.6 kph and on one occasion during August 2006 the wind speed peaked at 13 kph. Humidity at the time of samplings was 43±12%.
We detected significant differences among bacterial communities in aerosols collected on-site in close proximity to dairy animals from two different dairies in California. This was achieved by using aerosol collection devices similar to the ones used for automated indoor pathogen detection systems , . Low levels of aerobic bacteria were enumerated from Petrifilms, so we characterized bacterial communities from 97,200 liters of air from the Sonoma dairy and 129,600 liters of air from the Modesto dairy. These volumes are small compared to 1.4×106 liters of air collected in an urban environment during a 24 h period to characterize >1 µm particles for bacteria . We also maximized for the collection of bacteria that were likely from dairy cows by placing the aerosol devices in close proximity to hundreds of dairy cows in the barns. Since meteorological conditions are known to strongly influence the community structure of aerosols , aerosol wash-water from four devices was pooled to minimize the influence of wind, humidity and temperature fluctuations on the bacterial communities detected. Furthermore, location effects were also averaged by placing the collection devices in multiple locations on the dairy (Figure 1), and pooling the samples from all locations.
The detection of Proteobacteria representing 78% of all clones and 100% of the 10 most abundant sequences in aerosols from the Modesto dairy is noteworthy, because this phylum was nearly non-detectable in samples of fresh manure. A high proportion of Proteobacteria (81%) was detected also from aerosols collected downwind from a field treatment of biosolids . The increase in proportion of Proteobacteria in aerosols (Table 2) is an indication of their non-fecal nature . Drying in manure piles appears to have eliminated the most abundant sequences of phylum Firmicutes in fresh manure from both dairies (Table 2). Similarly, a high percentage of Firmicutes (75%) identified from dairy cow waste collected within 12 h after excretion  decreased to 10% in aerobic effluents of slurries prepared from the same waste. In contrast, Firmicutes were more abundant in aerosols and fresh manure from the Sonoma dairy compared to the Modesto dairy. Firmicutes were reported to be predominant also in aerosols produced during thermophilic composting . Thus, these results indicate a dynamic nature of bacterial communities in aerosols from different dairies and are influenced by different environmental conditions.
The absence of sequences in aerosols originating from fresh or dry manure from the Modesto dairy (Figure 3) is noteworthy, especially considering our collection of aerosols in close proximity to cows on a dairy producing ∼35,000 kg of manure per day , plus the aerosols emanating presumably from the surrounding dairies. This result is consistent with data reported by Durso et al  that bacterial communities in a feed lot pen surface are separate and distinct from the feces collected from cattle in the same pen. However, 3 (uncharacterized Bacteroidia and Clostridia) out of the top 10 sequences characterized from the Sonoma dairy (Figure 2) originated from fresh manure and members of both phylogenetic classes have been proposed as surrogates for tracking fecal contamination , . We attribute this result to the fine dust created by intense cattle movement around the aerosol collectors and to a more favorable wind direction (SSE 139° to 148°, SSW200°) at the Sonoma dairy, in contrast to a lack of apparent dust in the air nearer to the collectors at the Modesto dairy where the wind direction was NNE 13° to 26°.
Foodborne pathogens of high priority, such as Salmonella, Campylobacter and Listeria, were not detected from any of the 966 cloned sequences from aerosol and manure samples from both dairies. We detected one clone matching an E. coli in aerosols from the Sonoma dairy, but this is consistent with reported high concentrations of E. coli present in environmental samples on ranches and dairies . Many of the unidentified sequences characterized from both dairies had been isolated previously from fecal sources (Figures 2 and 3). Gram-positive anaerobic Clostridia, some species of which are human pathogens, were present at a significant level (32% of top 10 sequences) in aerosols generated on the Sonoma dairy (Figures 5 and 6). These results are consistent with reports of Clostridia being predominant in aerosols from swine confinement facilities  and colonic contents of dairy cows . Sequences representing the genus Massilia , and a possible human pathogen, were detected in aerosols from both dairies. Burkholderia cepacia, found in aerosols from both dairies is a human pathogen linked with outbreaks particularly among cystic fibrosis patients and has been developed as a biopesticide and for bioremediation . It was detected also in a previous study of urban aerosols collected from a site in Albany, California . Duganella zoogloeoides, a possible biocontrol agent, was detected from the Modesto dairy aerosol communities. Fresh manure from the Sonoma dairy contained possible animal or human pathogens of genus Treponema that were also detected from fecal microbiota obtained from a cattle herd with endemic Salmonella . Dry manure from Modesto contained a sequence that matched 100% with the GenBank sequence of potentially pathogenic slow growing Mycobacterium sp.  and >96% sequence similarity with Mycobacterium avium paratuberculosis that causes debilitating Johne's disease in cattle and known to be associated with human Crohn's disease . This is of particular importance since 27% of library sequences from Modesto dry manure and 8% of sequences in aerosols from Sonoma belong to Actinobacteria and, thus, a potential exists for the aerosol transport of Mycobacterium avium paratuberculosis, which has been reported to survive in feces for more than a year , and has been isolated from two-thirds of manure lagoons . Many other potential human and animal pathogens were detected at a single clone level from aerosols and manure samples from both dairies, but the concentration and potential viability of these organisms is uncertain.
In summary, we are reporting the first study of bacterial communities in aerosols from dairies in California. The bacteria in aerosols from the Modesto dairy did not originate from either fresh or dry manure samples collected on-site, a result in contrast to at least three abundant sequences detected in Sonoma dairy aerosols originating from fresh manure. Since, most bacterial sequences detected from aerosols or manure were unique to the sample source, further studies of aerosols from more dairies in the same or different locations will be required to determine the variability in bacterial communities in aerosols, whether bacteria in aerosols have enhanced fitness for survival in that environment, and whether any sequences can be identified as surrogates for tracking transport of enteric pathogens from point sources (e.g., ranches/dairies) to raw produce crops grown in the same vicinity.
Comparison of OTU libraries from aerosols with libraries generated from fresh and dry manure samples collected from Sonoma and Modesto dairies using RDP Library Compare. OTU libraries generated from aerosols from both dairies were also compared.
Conceived and designed the experiments: SR. Performed the experiments: CS SR. Analyzed the data: CS SR. Contributed reagents/materials/analysis tools: SR RM. Wrote the manuscript: SR RM.
- 1. Cole DJ, Hill VR, Humenik FJ, Sobsey MD (1999) Health, safety, and environmental concerns of farm animal waste. Occup Med 14: 423–448.DJ ColeVR HillFJ HumenikMD Sobsey1999Health, safety, and environmental concerns of farm animal waste.Occup Med14423448
- 2. McGarvey JA, Miller WG, Sanchez S, Stanker L (2004) Identification of bacterial populations in dairy wastewaters by use of 16S rRNA gene sequences and other genetic markers. Appl Environ Microbiol 70: 4267–4275.JA McGarveyWG MillerS. SanchezL. Stanker2004Identification of bacterial populations in dairy wastewaters by use of 16S rRNA gene sequences and other genetic markers.Appl Environ Microbiol7042674275
- 3. Pillai SD, Ricke SC (2002) Bioaerosols from municipal and animal wastes: background and contemporary issues. Can J Microbiol 48: 681–696.SD PillaiSC Ricke2002Bioaerosols from municipal and animal wastes: background and contemporary issues.Can J Microbiol48681696
- 4. Kudva IT, Blanch K, Hovde CJ (1998) Analysis of Escherichia coli O157:H7 survival in ovine or bovine manure and manure slurry. Appl Environ Microbiol 64: 3166–3174.IT KudvaK. BlanchCJ Hovde1998Analysis of Escherichia coli O157:H7 survival in ovine or bovine manure and manure slurry.Appl Environ Microbiol6431663174
- 5. Cornick NA, Vukhac H (2008) Indirect transmission of Escherichia coli O157:H7 occurs readily among swine but not among sheep. Appl Environ Microbiol 74: 2488–2491.NA CornickH. Vukhac2008Indirect transmission of Escherichia coli O157:H7 occurs readily among swine but not among sheep.Appl Environ Microbiol7424882491
- 6. Hutchison ML, Avery SM, Monaghan JM (2008) The air-borne distribution of zoonotic agents from livestock waste spreading and microbiological risk to fresh produce from contaminated irrigation sources. J Appl Microbiol 105: 848–857.ML HutchisonSM AveryJM Monaghan2008The air-borne distribution of zoonotic agents from livestock waste spreading and microbiological risk to fresh produce from contaminated irrigation sources.J Appl Microbiol105848857
- 7. Harbaugh E, Trampel D, Wesley I, Hoff S, Griffith R, et al. (2006) Rapid aerosol transmission of Salmonella among turkeys in a simulated holding-shed environment. Poult Sci 85: 1693–1699.E. HarbaughD. TrampelI. WesleyS. HoffR. Griffith2006Rapid aerosol transmission of Salmonella among turkeys in a simulated holding-shed environment.Poult Sci8516931699
- 8. Brooks JP, McLaughlin MR, Scheffler B, Miles DM (2010) Microbial and antibiotic resistant constituents associated with biological aerosols and poultry litter within a commercial poultry house. Sci Total Environ 408: 4770–4777.JP BrooksMR McLaughlinB. SchefflerDM Miles2010Microbial and antibiotic resistant constituents associated with biological aerosols and poultry litter within a commercial poultry house.Sci Total Environ40847704777
- 9. Ravva SV, Sarreal CZ, Duffy B, Stanker LH (2006) Survival of Escherichia coli O157:H7 in wastewater from dairy lagoons. J Appl Microbiol 101: 891–902.SV RavvaCZ SarrealB. DuffyLH Stanker2006Survival of Escherichia coli O157:H7 in wastewater from dairy lagoons.J Appl Microbiol101891902
- 10. Ibekwe AM, Grieve CM (2004) Changes in developing plant microbial community structure as affected by contaminated water. FEMS Microbiol Ecol 48: 239–248.AM IbekweCM Grieve2004Changes in developing plant microbial community structure as affected by contaminated water.FEMS Microbiol Ecol48239248
- 11. McGarvey JA, Miller WG, Zhang R, Ma Y, Mitloehner F (2007) Bacterial population dynamics in dairy waste during aerobic and anaerobic treatment and subsequent storage. Appl Environ Microbiol 73: 193–202.JA McGarveyWG MillerR. ZhangY. MaF. Mitloehner2007Bacterial population dynamics in dairy waste during aerobic and anaerobic treatment and subsequent storage.Appl Environ Microbiol73193202
- 12. Wang Q, Garrity GM, Tiedje JM, Cole JR (2007) Naive Bayesian classifier for rapid assignment of rRNA sequences into the new bacterial taxonomy. Appl Environ Microbiol 73: 5261–5267.Q. WangGM GarrityJM TiedjeJR Cole2007Naive Bayesian classifier for rapid assignment of rRNA sequences into the new bacterial taxonomy.Appl Environ Microbiol7352615267
- 13. Bergman W, Shinn J, Lochner R, Sawyer S, Milanovich F, et al. (2005) High air flow, low pressure drop, bio-aerosol collector using a multi-slit virtual impactor. J Aerosol Sci 36: 619–638.W. BergmanJ. ShinnR. LochnerS. SawyerF. Milanovich2005High air flow, low pressure drop, bio-aerosol collector using a multi-slit virtual impactor.J Aerosol Sci36619638
- 14. McBride MT, Masquelier D, Hindson BJ, Makarewicz AJ, Brown S, et al. (2003) Autonomous detection of aerosolized Bacillus anthracis and Yersinia pestis. Anal Chem 75: 5293–5299.MT McBrideD. MasquelierBJ HindsonAJ MakarewiczS. Brown2003Autonomous detection of aerosolized Bacillus anthracis and Yersinia pestis.Anal Chem7552935299
- 15. Radosevich JL, Wilson WJ, Shinn JH, DeSantis TZ, Andersen GL (2002) Development of a high-volume aerosol collection system for the identification of air-borne micro-organisms. Lett Appl Microbiol 34: 162–167.JL RadosevichWJ WilsonJH ShinnTZ DeSantisGL Andersen2002Development of a high-volume aerosol collection system for the identification of air-borne micro-organisms.Lett Appl Microbiol34162167
- 16. Brodie EL, DeSantis TZ, Parker JP, Zubietta IX, Piceno YM, et al. (2007) Urban aerosols harbor diverse and dynamic bacterial populations. Proc Natl Acad Sci U S A 104: 299–304.EL BrodieTZ DeSantisJP ParkerIX ZubiettaYM Piceno2007Urban aerosols harbor diverse and dynamic bacterial populations.Proc Natl Acad Sci U S A104299304
- 17. Baertsch C, Paez-Rubio T, Viau E, Peccia J (2007) Source tracking aerosols released from land-applied class B biosolids during high-wind events. Appl Environ Microbiol 73: 4522–4531.C. BaertschT. Paez-RubioE. ViauJ. Peccia2007Source tracking aerosols released from land-applied class B biosolids during high-wind events.Appl Environ Microbiol7345224531
- 18. Wu CH, Sercu B, Van de Werfhorst LC, Wong J, DeSantis TZ, et al. (2010) Characterization of coastal urban watershed bacterial communities leads to alternative community-based indicators. PLoS One 5: e11285.CH WuB. SercuLC Van de WerfhorstJ. WongTZ DeSantis2010Characterization of coastal urban watershed bacterial communities leads to alternative community-based indicators.PLoS One5e11285
- 19. Le Goff O, Bru-Adan V, Bacheley H, Godon JJ, Wery N (2010) The microbial signature of aerosols produced during the thermophilic phase of composting. J Appl Microbiol 108: 325–340.O. Le GoffV. Bru-AdanH. BacheleyJJ GodonN. Wery2010The microbial signature of aerosols produced during the thermophilic phase of composting.J Appl Microbiol108325340
- 20. Durso LM, Harhay GP, Smith TP, Bono JL, Desantis TZ, et al. (2011) Bacterial community analysis of beef cattle feedlots reveals that pen surface is distinct from feces. Foodborne Pathog Dis. In press. LM DursoGP HarhayTP SmithJL BonoTZ Desantis2011Bacterial community analysis of beef cattle feedlots reveals that pen surface is distinct from feces.Foodborne Pathog DisIn press
- 21. Bernhard AE, Field KG (2000) Identification of nonpoint sources of fecal pollution in coastal waters by using host-specific 16S ribosomal DNA genetic markers from fecal anaerobes. Appl Environ Microbiol 66: 1587–1594.AE BernhardKG Field2000Identification of nonpoint sources of fecal pollution in coastal waters by using host-specific 16S ribosomal DNA genetic markers from fecal anaerobes.Appl Environ Microbiol6615871594
- 22. Savichtcheva O, Okabe S (2006) Alternative indicators of fecal pollution: relations with pathogens and conventional indicators, current methodologies for direct pathogen monitoring and future application perspectives. Water Research 40: 2463–2476.O. SavichtchevaS. Okabe2006Alternative indicators of fecal pollution: relations with pathogens and conventional indicators, current methodologies for direct pathogen monitoring and future application perspectives.Water Research4024632476
- 23. Cooley M, Carychao D, Crawford-Miksza L, Jay MT, Myers C, et al. (2007) Incidence and tracking of Escherichia coli O157:H7 in a major produce production region in California. PLoS One 2: e1159.M. CooleyD. CarychaoL. Crawford-MikszaMT JayC. Myers2007Incidence and tracking of Escherichia coli O157:H7 in a major produce production region in California.PLoS One2e1159
- 24. Nehme B, Letourneau V, Forster RJ, Veillette M, Duchaine C (2008) Culture-independent approach of the bacterial bioaerosol diversity in the standard swine confinement buildings, and assessment of the seasonal effect. Environ Microbiol 10: 665–675.B. NehmeV. LetourneauRJ ForsterM. VeilletteC. Duchaine2008Culture-independent approach of the bacterial bioaerosol diversity in the standard swine confinement buildings, and assessment of the seasonal effect.Environ Microbiol10665675
- 25. McGarvey JA, Hamilton SW, DePeters EJ, Mitloehner FM (2010) Effect of dietary monensin on the bacterial population structure of dairy cattle colonic contents. Appl Microbiol Biotechnol 85: 1947–1952.JA McGarveySW HamiltonEJ DePetersFM Mitloehner2010Effect of dietary monensin on the bacterial population structure of dairy cattle colonic contents.Appl Microbiol Biotechnol8519471952
- 26. La Scola B, Birtles RJ, Mallet M-N, Raoult D (1998) Massilia timonae gen. nov., sp. nov., Isolated from blood of an immunocompromised patient with cerebellar lesions. J Clin Microbiol 36: 2847–2852.B. La ScolaRJ BirtlesM-N MalletD. Raoult1998Massilia timonae gen. nov., sp. nov., Isolated from blood of an immunocompromised patient with cerebellar lesions.J Clin Microbiol3628472852
- 27. Holmes A, Govan J, Goldstein R (1998) Agricultural use of Burkholderia (Pseudomonas) cepacia: a threat to human health? Emerg Infect Dis 4: 221–227.A. HolmesJ. GovanR. Goldstein1998Agricultural use of Burkholderia (Pseudomonas) cepacia: a threat to human health?Emerg Infect Dis4221227
- 28. Hernlem BJ, Ravva SV (2007) Application of flow cytometry and cell sorting to the bacterial analysis of environmental aerosol samples. J Environ Monit 9: 1317–1322.BJ HernlemSV Ravva2007Application of flow cytometry and cell sorting to the bacterial analysis of environmental aerosol samples.J Environ Monit913171322
- 29. Patton TG, Scupham AJ, Bearson SM, Carlson SA (2009) Characterization of fecal microbiota from a Salmonella endemic cattle herd as determined by oligonucleotide fingerprinting of rDNA genes. Vet Microbiol 136: 285–292.TG PattonAJ ScuphamSM BearsonSA Carlson2009Characterization of fecal microbiota from a Salmonella endemic cattle herd as determined by oligonucleotide fingerprinting of rDNA genes.Vet Microbiol136285292
- 30. Leclerc MC, Haddad N, Moreau R, Thorel MF (2000) Molecular characterization of environmental Mycobacterium strains by PCR-restriction fragment length polymorphism of hsp65 and by sequencing of hsp65, and of 16S and ITS1 rDNA. Res Microbiol 151: 629–638.MC LeclercN. HaddadR. MoreauMF Thorel2000Molecular characterization of environmental Mycobacterium strains by PCR-restriction fragment length polymorphism of hsp65 and by sequencing of hsp65, and of 16S and ITS1 rDNA.Res Microbiol151629638
- 31. Ravva SV, Stanker LH (2005) Real-time quantitative PCR detection of Mycobacterium avium subsp. paratuberculosis and differentiation from other mycobacteria using SYBR Green and TaqMan assays. J Microbiol Methods 63: 305–317.SV RavvaLH Stanker2005Real-time quantitative PCR detection of Mycobacterium avium subsp. paratuberculosis and differentiation from other mycobacteria using SYBR Green and TaqMan assays.J Microbiol Methods63305317
- 32. Whittington RJ, Marshall DJ, Nicholls PJ, Marsh IB, Reddacliff LA (2004) Survival and dormancy of Mycobacterium avium subsp. paratuberculosis in the environment. Appl Environ Microbiol 70: 2989–3004.RJ WhittingtonDJ MarshallPJ NichollsIB MarshLA Reddacliff2004Survival and dormancy of Mycobacterium avium subsp. paratuberculosis in the environment.Appl Environ Microbiol7029893004
- 33. Raizman EA, Wells SJ, Godden SM, Bey RF, Oakes MJ, et al. (2004) The distribution of Mycobacterium avium ssp. paratuberculosis in the environment surrounding Minnesota dairy farms. J Dairy Sci 87: 2959–2966.EA RaizmanSJ WellsSM GoddenRF BeyMJ Oakes2004The distribution of Mycobacterium avium ssp. paratuberculosis in the environment surrounding Minnesota dairy farms.J Dairy Sci8729592966