Free access
Spotlight Selection
Research Article
15 July 2016

Influence of Host Ecology and Behavior on Campylobacter jejuni Prevalence and Environmental Contamination Risk in a Synanthropic Wild Bird Species

This article has a companion.
VIEW THE COMPANION

ABSTRACT

Campylobacter jejuni is a foodborne pathogen that often leads to human infections through the consumption of contaminated poultry. Wild birds may play a role in the transmission of C. jejuni by acting as reservoir hosts. Despite ample evidence that wild birds harbor C. jejuni, few studies have addressed the role of host ecology in transmission to domestic animals or humans. We tested the hypothesis that host social behavior and habitat play a major role in driving transmission risk. C. jejuni infection and host ecology were studied simultaneously in wild American crows (Corvus brachyrhynchos) in Davis, CA, over 3 years. We found that 178 of 337 samples tested were culture positive (53%), with infection varying by season and host age. Among adult crows, infection rates were highest during the winter, when migrants return and crows form large communal roosts. Nestlings had the highest risk of infection, and whole-genome sequencing supports the observation of direct transmission between nestlings. We deployed global positioning system (GPS) receivers to quantify habitat use by crows; space use was nonrandom, with crows preferentially occupying some habitats while avoiding others. This behavior drastically amplified the risk of environmental contamination from feces in specific locations. This study demonstrates that social behavior contributes to infection within species and that habitat use leads to a heterogeneous risk of cross-species transmission.
IMPORTANCECampylobacter jejuni is the most common cause of gastroenteritis in industrialized countries. Despite efforts to reduce the colonization of poultry flocks and eventual infection of humans, the incidence of human C. jejuni infection has remained high. Because wild birds can harbor strains of C. jejuni that eventually infect humans, there has long been speculation that wild birds might act as an important reservoir in the C. jejuni infection cycle. We simultaneously studied infection prevalence, social behavior, and movement ecology in wild American crows (Corvus brachyrhynchos). We found that social behavior contributed to patterns of infection and that movement behavior resulted in some areas having a high risk of transmission while others had a low risk. The incorporation of ecological data into studies of C. jejuni in wild birds has the potential to resolve when and how wild birds contribute to domestic animal and human C. jejuni infection, leading to better control of initial poultry contamination.

INTRODUCTION

Campylobacter jejuni is the leading cause of gastroenteritis in industrialized countries (1, 2), with most infections in humans resulting from the consumption of contaminated and improperly cooked poultry (3). Generally, infected individuals recover in a few days, but in some cases, infection can lead to hospitalization (∼15% of culture-positive cases in the United States [1]), chronic autoimmune disorders (∼2% of cases [4, 5]), and even death (∼0.06% of cases [1]). These health concerns demand that food producers manage flock infection and disinfect suspect meat at a considerable cost, yet some surveys suggest that even with the precautions taken, up to 70% of poultry sold in U.S. and United Kingdom grocery stores is contaminated with C. jejuni (6, 7). Given the human health and financial costs caused by C. jejuni, it is unsurprising that major research efforts have been devoted to reducing flock infection (811), developing genetically resistant lines of poultry (12), and limiting transmission from poultry to humans (13, 14). Despite these efforts, human infection rates in the United States have failed to match CDC targets and have actually increased 14% over the past 10 years (1). Although most research is carried out in the laboratory or on poultry farms, scientists have long known that C. jejuni can also be carried by wild birds (e.g., see reference 15), and there has been extensive speculation that wild bird populations might act as reservoirs of C. jejuni, thereby contributing to domestic animal infection or even to direct human infection via environmental contamination from feces (reviewed in reference 16).
Studies documenting C. jejuni prevalence in wild animals have increased rapidly in the past 15 years (1720). Many studies now show that diverse families of wild birds harbor C. jejuni (reviewed in reference 16) and that, at least in some cases, these strains are similar to those that infect poultry, livestock, and humans (9, 21). The majority of this work has been conducted in Europe, but a smaller number of studies in the United States demonstrate similar patterns (22, 23). Relatively few studies, however, go beyond a simple culture test to detect the presence of C. jejuni. In rare cases, direct transmission from wild birds to humans has been demonstrated beyond any doubt (24, 25), but direct transmission is difficult to verify, and the actual transmission rates are probably underestimated. One recent study concluded that 2.1 to 3.5% of annual human C. jejuni cases in Oxfordshire, United Kingdom, are directly attributable to wild birds (26). Despite ample evidence that C. jejuni infection in birds is common, the importance of wild birds in the human C. jejuni infection cycle is still unclear.
Studies that survey wild birds for C. jejuni often implicitly assume that high prevalence is a good proxy for high risk of transmission to humans or domestic animals (22). Carriage of human-pathogenic strains by wild birds is a necessary antecedent for undesirable cross-species transmission, but host ecology and behavior are also likely to be major determinants of the actual risk of transmission posed by specific wild birds. In a recent review, Waldenström and Griekspoor (16) argue that the lack of information on the ecology of wild hosts infected by C. jejuni has limited our ability to make informed risk assessments. Even basic host ecology studies of wild bird species that carry C. jejuni might greatly increase our understanding of factors that contribute to transmission. For example, Ramos et al. (27) demonstrated that C. jejuni infection rates vary among populations of yellow-legged gulls (Larus michahellis), and that these variations are attributable to the amount of human refuse in the diet at both the individual and population level. How pervasive such patterns are and how strongly they influence the risk of cross-species C. jejuni transmission are open questions.
American crows (Corvus brachyrhynchos), members of the corvid family, are excellent candidates for studying links between host ecology and C. jejuni association. Corvids are a globally distributed clade of birds known for their close association with domestic animals, agricultural fields, and urban areas. Corvids have the highest rates of C. jejuni infection of any sampled clade (16, 28, 29) and, among those families with relatively high infection rates, corvids arguably have the most direct contact with humans and domestic animals. We established previously that C. jejuni infection is common in American crows in the suburban town of Davis, CA, and that at least some of the isolates found have characteristics of strains that are pathogenic to humans (28). In addition to the high prevalence of C. jejuni infection, American crows exhibit social behavior and movement patterns that may influence the risk of transmission from crows to domestic animals or directly to humans (30). Each winter, crows form large communal roosts that may facilitate both within-species transmission and amplify cross-species transmission risks in areas surrounding the roost (31). Crows are also strong flyers and social foragers; thus, the risk of contamination may be spread far from the roost but may also be highly concentrated in areas that attract crows (e.g., feedlots or particular crops) rather than be equally distributed across the landscape.
Here, we build on previous work on the molecular characteristics of C. jejuni isolates from American crows (28) by integrating those results with host ecology. Specifically, we conducted extensive year-round sampling of crow feces to look for evidence of seasonal prevalence patterns associated with social behavior and differences in C. jejuni prevalence in feces between adults and nestlings. We hypothesized that C. jejuni prevalence would be higher in nestlings than adults, due to a less-developed immune system and high exposure to social sources of infection from parents and nest mates (32). We also predicted that overall C. jejuni prevalence in feces would be highest at the peak of the communal roosting season, when thousands of crows spend each night in close proximity and up to 58% of crows have visible fecal staining on their feathers from roost mates perched on higher branches (31). For nestlings, we predicted that the shared nest environment would drive transmission, with nest mates tending to be either all uncolonized or all colonized with a similar strain of C. jejuni. Finally, we combined our information on prevalence with movement data from global positioning system (GPS) receivers deployed on adult crows to describe winter foraging behavior and patterns of landscape use during the roosting season.

MATERIALS AND METHODS

Study population and general field methods.

We studied C. jejuni infection in wild American crows in Davis, CA, from May 2012 until June 2015. During that time, year-round fecal samples were collected for C. jejuni testing from adults and nestlings. Crow density varied seasonally, because migratory crows overwintered in Davis from October to March each year. During this period, migrants and year-round residents formed a large communal roost each night. Each morning before dawn, crows dispersed from the roost to daytime feeding locations. In the winters of 2014 and 2015, periodic roost counts were conducted before dawn to estimate the onset and duration of roosting along with the peak winter roost attendance (as described in reference 31). Resident crows also formed much smaller roosts in the summer between May and August. These roosts were unpredictable in size and location and generally only included 100 to 400 individuals. Samples were collected from summer roosts when possible, but data from summer roosts were sparse, and no samples were collected in August or September when the birds did not roost in large numbers.
After migrants left the winter roost, local crows began their breeding season, which typically lasted from early April until late June. During the 2012, 2013, and 2014 breeding seasons, breeding activity was monitored closely at crow territories on and around the University of California, Davis campus (described in reference 33). Once a nest was located, a regular census was conducted until hatching, and then each nest was visited to collect samples for C. jejuni testing. In 2012, fecal samples were collected from most nestlings that were visited, but in 2013 and 2014, samples were collected only opportunistically when a nestling defecated during handling.
Similar samples were collected from adult birds captured during the winter roosting season in 2014 and 2015. Adults were captured using a drop-in trap that was baited and set before dawn so that crows would encounter it when leaving the roost. After capture, crows were held in plastic carriers until processing; a fresh fecal sample was collected from each carrier as crows were removed. Some of the captured adults were fitted with GPS receivers to quantify winter habitat use (see below). All trapping, banding, and sampling procedures were conducted with approval from the United States Geological Survey (USGS) Bird Banding Laboratory permit no. 23777, the California Department of Fish and Wildlife permit no. 12065, and University of California Davis IACUC protocol no. 16897.

Campylobacter testing.

All fecal samples were collected by swabbing fresh feces and storing the swab in Amies clear gel transport medium (Remel BactiSwab; Fisher Scientific) in a cooler until submission for testing (within 6 h of collection) (28). In addition to the captured nestling and adult samples described above, feces were collected periodically from directly under the large winter roost or the much smaller summer roost. In order to collect the freshest samples, fecal samples were always collected before sunrise. To prevent ground contamination, only the top portion of the feces was collected. In total, 119 samples were tested from the winter roost between 31 October and 3 April, 71 samples from the small and less-predictable summer roost between 23 April and 31 July, 102 samples from nestlings between 6 May and 29 June, and 45 samples from captured adults between 8 January and 14 March. All samples were submitted for testing to the U.C. Davis Veterinary Medicine Teaching Hospital Clinical Laboratory Services Facility. Samples were cultured and, if candidate colonies were identified, subjected to biochemical testing to confirm the presence of C. jejuni (28). For a subset of samples (n = 16), duplicate swabs were collected and submitted for independent testing to evaluate the repeatability of the lab results. We calculated 95% binomial confidence intervals of infection prevalence for each type of sample using the ‘binconf’ function of the ‘Hmisc’ package in R; we report the Wilson score interval, as recommended by Agresti and Coull for binomial proportions (34).

GPS deployment.

In January 2015, eight GPS receivers were deployed on adult crows captured in the drop-in trap. The receivers were 15-g FLR-II units produced by Telemetry Solutions (Concord, CA, USA), which recorded locations that were accurate to within about five meters. Crows fitted with receivers weighed 353 to 398 g, so that the weight of the receiver was always <4.5% of the bird's body weight. Receivers were secured by a backpack harness that looped around each wing to attach at the breast; harnesses were made from 1.5-mm nylon cord and fastened with copper crimps, superglue, and stitches applied by needle and thread at each connection point. After the harnesses were attached, crows were released into a holding aviary for at least 1 h to ensure that the GPS unit was functioning properly and that the harness was safely attached. Crows were released following this holding period. Receivers were programmed to attempt to collect nine location points each day at the following fixed times: 1:00, 3:00, 7:00, 9:00, 11:00, 13:00, 15:00, 17:00, 19:00, and 23:00. To preserve battery life, the units were programmed to time out if a location fix could not be established within 65 s, so the complete data set for each bird did not always contain all scheduled locations.
The receivers were equipped for remote transfer so that locations could be downloaded to a base station held within ∼30 m of the receiver without needing to recapture the bird. Surveys of the roost occurred well before dawn 2 to 3 times per week after the receivers were deployed to download locations. At these visits, observers systematically walked under each tree and held the base station aloft to establish connections. An unmanned base station was also deployed in the field at areas of known crow activity to maximize chances of recovering location data. Location recovery efforts were continued until the roost had completely dispersed during the first week of April 2015.

Spatial analysis.

The Quantum Geographic Information Systems (QGIS) software (Quantum GIS Development Team, 2015, http://qgis.osgeo.org) was used to match our GPS locations with landscape characteristics using the USDA CropScape cropland 2014 data layer (USDA 2014; http://nassgeodata.gmu.edu/CropScape/). The cropland data layer draws basic land use data from the National Land Cover Database 2011 layer (35) but further subdivides agricultural areas into specific crop types. Nearly all nocturnal (19:00 to 7:00) locations were at or near the roost in developed areas, while diurnal (7:00 to 19:00) locations were more variable. Thus, diurnal and nocturnal time budgets were analyzed separately. Locations were pooled across individual birds before calculating the amount of time spent in each land use or crop type as a simple average of the number of locations in each area divided by the total number of locations collected. To provide context for interpreting habitat use, the total percentage of land in the study area for each habitat or crop type was also calculated. The study area was defined as a circle with a 12.5-km radius from the roost, which contained ∼95% of the observed daytime crow locations. Although the individual tracked birds did not utilize the entire area of this circle, this boundary was chosen because it best represented the choice of habitat that crows could have exploited given the distance that they traveled from the roost each day. Calculating habitat use with a 95% minimum convex polygon rather than a circle yielded qualitatively similar conclusions, and only the results using a 12.5-km buffer are presented for simplicity. We calculated the approximate home range of each individual crow using minimum convex polygons from all observed locations.
The predicted amount of crow feces was calculated per square kilometer per day in each habitat type at the peak of the roosting season by multiplying the percentage of time spent in each habit by the maximum number of crows at the roost (∼6,000) and by the estimated number of fecal samples per day per crow (Table 1). We recorded the number of defecations by nine adult crows held in aviaries for 6.3 ± 2.8 h (mean ± standard deviation). These individuals defecated 7.9 ± 3.5 times per hour (mean ± standard deviation), but this rate was almost certainly elevated by the stress of capture, especially immediately after placement in the aviary. For the purposes of illustration, a very conservative daytime defecation rate of 18 fecal samples per day per crow was adopted (1.5 feces per hour). This rate should not be considered definitive but provides a minimum estimate of crow defecation across the landscape.
A randomization test was used to determine if daytime habitat use was biased toward or against certain habitat types. First, the habitat type was identified at 50,000 regularly spaced locations within the 12.5-km buffer surrounding the roost. Using this set of locations as a starting point, 1,430 points were randomly sampled (the number of daytime crow observations) 10,000 times, and for each set of points, the percentage of locations in each habitat type was calculated. Repeated sampling of the same point was allowed to account for the fact that crows could be observed in the same location multiple times. The observed crow usage of each habitat was then compared to the distribution generated by random sampling. Crows were considered to significantly favor a habitat if the observed value was greater than the 97.5th percentile of the random distribution or significantly disfavor a habitat if the observed value was less than the 2.5th percentile of the random distribution. No similar test was performed for nighttime locations, because these locations were focused solely on the roost.

Genomic analysis.

To assess the similarity of C. jejuni isolates obtained from nest mates, we analyzed whole-genome sequences (WGS) of isolates collected from 23 individually marked nestlings from 9 nests (see collection methods above). For this analysis, isolates were included when at least two positive samples were available from separate nestlings in the same nest. High-molecular-weight genomic DNA (gDNA) was obtained from colonies isolated from blood agar plates after growth at 37°C under microaerophilic gas conditions, as previously described by the 100K Pathogen Genome Project (3638). Briefly, the cells were lysed with an enzyme cocktail, mixed by vortexing, and purified gDNA was isolated using whole-genome kits, according to the manufacturer's instructions (Qiagen, Valencia, CA, USA). gDNA purity and integrity were confirmed on a 2200 TapeStation with the genomic DNA ScreenTape (Agilent Technologies, Santa Clara, CA, USA) (3941). gDNA with A260/280 and A260/230 of >1.8 was used for library construction.
Following isolation, gDNA was sheared using the Covaris E220 with the 96 microTUBE plate (Covaris, Inc., Woburn, MA, USA) (42). Libraries were made using Kapa high-throughput (HTP) library preparation kit (KR0426 version 3.13; Kapa Biosystems, Wilmington, MA, USA) with dual-SPRI size selection (43). Libraries were constructed using the Agilent Bravo (Agilent Technologies, Santa Clara, CA). Library quantitation was done using Kapa SYBR Fast qPCR kits (Kapa Biosystems) to ensure a starting concentration of 400 ng and a fragment insert size between 350 and 450 bp (43). Libraries were indexed using Bioo Scientific NEXTflex-96 DNA barcodes version 13.05 (Bioo Scientific Corp., Austin, TX) and Integrated DNA Technologies Weimer 384 TS-LT DNA barcodes. Sequencing was done by [email protected] (Sacramento, CA, USA) on an Illumina HiSeq 2000 platform using paired-end 100 bp (PE100) reads (Illumina, Inc., San Diego, CA, USA) (44, 45).
Sequences were aligned using progressiveMauve (46, 47), and contigs were reordered using the reorder contigs option in Mauve under standard parameters using C. jejuni subsp. jejuni NCTC 11168 as the reference genome (see Fig. S1 in the supplemental material). Genomic distances were determined using the Genome-to-Genome Distance Calculator (GGDC) (http://ggdc.dsmz.de/distcalc2.php) (48, 49). Distances were calculated using formula 2, as recommended for draft genomes. Distance matrices were translated into Newick tree format in the T-Rex Web server software using the neighbor-joining method (50, 51). Trees were edited using Dendroscope 3.0 (52) and Geneious version 6.1.8 (53).
To assess whether nest mate isolates were more similar than expected by chance, we conducted a randomization test using our strain GGDCs. For this test, the GGDC matrix was held constant while randomly reassigning nestlings to different C. jejuni strains and then calculating the average nest mate similarity based on the matrix of GGDC values. This procedure was iterated 100,000 times to generate an expected nest mate similarity distribution assuming random infection; the empirical isolate similarity that was observed was then compared to the random distribution. Given the a priori prediction that nest mates would have more similar strains than expected by chance, a P value was calculated as the number of random permutations with more similar nestling isolates divided by the total number of permutations (100,000). Randomization tests and all other statistical procedures were conducted in R version 3.2.2 (R Core Development Team, Vienna, Austria).

Accession number(s).

Data from isolates that were sequenced as part of this study are accessible through NCBI (SRA accession numbers SRR1815854 to SRR1815856, SRR1815858, SRR1815860, SRR1816035, SRR1816037 to SRR1816040, SRR1816044, SRR1816046 to SRR1816049, SRR1816118, SRR1816122, SRR1816123, SRR1816125, SRR1816131, and SRR1816132; see Table S2 in the supplemental material).

RESULTS

Campylobacter prevalence by date, subject age, and social behavior.

Overall, 337 individual fecal samples were examined for C. jejuni. Of those, 178 (53%) tested positive for C. jejuni. Prevalence varied by crow age and season. Nestlings had the highest prevalence, with 73 out of 102 (72%) samples testing positive. Samples collected in the winter from captured adults or swabs under the roost had intermediate prevalence, with 51% of samples testing positive (22 of 45 captured adults and 61 of 119 roost swabs). The summer roost had the lowest prevalence, with 22 of 71 (31%) fecal samples testing positive. Differences in prevalence among samples collected during winter roosting, during summer roosting, or from nestlings all were significant (Fig. 1; Fisher's exact test, P < 0.01). Sixteen of these samples were submitted in duplicate, and all produced the same test result (9 positive, 7 negative). The difference in prevalence for adult samples collected during winter or summer roosting was consistent with the hypothesis that roost size drives transmission (Fig. 2A and B), but our data did not allow us to test for a causal link between roost size and transmission per se.
FIG 1
FIG 1 Overall prevalence of Campylobacter jejuni in fecal samples collected from the winter roost (including adults captured during the roosting season, n = 164), from nestlings (n = 102), or from the summer roost (n = 71). The infection prevalence for each pair of sample types was significantly different by Fisher's exact test. Error bars illustrate the 95% binomial confidence interval for prevalence estimates. Asterisks (*) indicate significant differences with a P value of <0.05.
FIG 2
FIG 2 Crow roost size (A) and year-round prevalence of C. jejuni in crow feces for adults and nestlings (B). Nestlings were sampled only in April and May, because crows do not breed at other times of the year. No fecal samples were collected in August and September, because crows did not roost communally in those months, making sample collection difficult. Roost point counts are plotted with a smoothed localized regression line to illustrate seasonal patterns. Prevalence estimates show the percentage of isolates that tested positive for C. jejuni in each month, pooling all samples across years. Error bars illustrate the 95% binomial confidence interval for each prevalence estimate. The total number of samples tested varied by month.
We also found that social proximity in the nesting environment was a strong predictor of C. jejuni infection (likelihood ratio test comparing a full and reduced logistic mixed model with nestling infection status as the binary response, presence of a colonized nest mate as a predictor, and nest identity as a random effect: n = 69 nestlings from 26 nests, χ2 = 21.4, P < 0.0001). In general, nest mates were either all colonized with C. jejuni or all uncolonized, with only a few nests (4 out of 26) containing a mix of colonized and uncolonized individuals.
GGDC calculations for each pair of sequenced isolates from these nestlings indicated that many nest mates were colonized by very similar strains of C. jejuni (GGDC for nest mates, 0.012 ± 0.003; GGDC for non-nest mates, 0.021 ± 0.001; Fig. 3). The randomization test demonstrated that the GGDC for strains for C. jejuni isolates from the same nest was significantly (P = 0.006) smaller than expected by chance. In a few cases, strains from the same nest were genomically divergent, suggesting that multiple strains may be infecting individuals in a single nest (i.e., acquired from both parents and/or from adult helpers-at-the-nest [54]).
FIG 3
FIG 3 Genomic distance relationship of Campylobacter jejuni strains collected from nests where at least two nestlings sampled were positive (n = 9 nests) and where full-genome sequences were available. Tips with the same letter indicate isolates that came from different nestlings in the same nest. Nest mates with GGDC values of <0.003 are shown in bold. Isolates from the same nest had smaller GGDC values than expected by chance (P = 0.006). The isolate identification numbers can be used to find NCBI accession numbers from Table S2 in the supplemental material.

Movement data and habitat use.

Location data were collected from seven of the eight GPS receivers deployed. The bird carrying the eighth unit was never seen after initial deployment, and we assume that this bird did not return to the communal roost that we were monitoring, although the lack of data could also have been caused by a hardware malfunction. From the seven birds we reencountered, we collected a total of 2,147 locations between 9 January and 27 March 2015 (Fig. 4). Of these locations, 717 were collected at night and 1,430 during the day. The average home range size (and standard deviation) for these seven birds based on minimum convex polygons was 235 ± 207 km2 (range, 78 to 684 km2).
FIG 4
FIG 4 (A) Map of the study site displaying 2,147 GPS locations collected from American crows. The large yellow circles indicate the nests included in Fig. 3. The remaining symbols indicate the 7 individual birds from which locations were obtained. Insets show zoomed-in views of the nests included in Fig. 3, with the letters corresponding to those included on the phylogeny (B) and the primate research center (C). This map uses satellite imagery available from the U.S. Geological Survey; the map was constructed using the QGIS software.
On average, crows forayed 7.7 ± 4.1 km (mean ± standard deviation) from the roost each day, with a maximum observed distance of 29 km. During the night, 96.7% of the collected locations were in developed areas, with the vast majority occurring at the communal roost. Daytime locations were recorded from both developed (25.9%) and agricultural (71.5%) areas, with the majority of daytime foraging occurring in the agricultural areas surrounding the town. Overall, daytime habitat usage differed from random expectations in every habitat category that we assessed except for areas of low-intensity development (Table 1). After accounting for the amount of each habitat type that was available, crows preferentially spent time in fallow fields, grasslands, hay fields, almond trees, walnut trees, and areas of medium- or high-intensity development (Table 1; randomization test, P < 0.04) and avoided tomatoes, winter wheat, sunflower, alfalfa, rice, and other crops (Table 1; randomization test, P < 0.001). We also noted heavy daytime use of two particular locations outside the roost. First, 3.2% of the daytime locations were at a local dairy barn, which encompassed <0.00001% of the available land area in the study site. Second, 19.4% of the daytime locations were at a local primate research center, which encompassed <0.001% of the available land area (Table 2).
TABLE 1
TABLE 1 Winter habitat use by American crows based on locations from GPS receivers compared to baseline percentage of each habitat
Habitat typeNight use (19:00 to 7:00) (%)Day use (7:00 to 19:00) (%)Baseline use %Day vs baseline P valueaCrows/km2/dayEstimated feces/km2/day
Developed96.725.911.3 27.3491
    Open or low intensity10.57.76.50.0814.0252
    Medium intensity80.012.64.1+<0.000132.1578
    High intensity6.22.30.7+<0.000135.1632
Agricultural3.371.587.7 10.0180
    Alfalfa0.77.314.7−<0.00016.2112
    Almonds0.18.95.1+<0.000121.8392
    Fallow/idle cropland0.419.516.2+0.00214.3257
    Grassland/pasture0.14.81.8+0.000134.5621
    Other hay/nonalfalfa0.04.93.8+0.03416.6299
    Sunflowers0.42.27.8−<0.00013.359
    Tomatoes0.84.915.3−<0.00013.970
    Walnuts0.19.04.5+<0.000124.4439
    Winter wheat0.05.78.3−<0.00018.6155
    Rice0.00.03.2−<0.00010.00
    Other crops0.74.37.0−<0.00018.0144
Other (e.g., wetlands, barren)0.02.61.0+<0.000131.8572
a
From randomization test.
TABLE 2
TABLE 2 Sample collection data by specific location
Specific locationCollected at night (%)Collected during day (%)Total area (km2)No. of crows/dayNo. of fecal samples/day
Primate center0.019.40.251,16420,952
Local dairy barn<0.13.20.0011923,456

DISCUSSION

Although many studies have suggested that wild birds play a role in C. jejuni transmission, the contributions of host ecology, movement, and social behavior to the risk of cross-species transmission have been largely unexplored. We documented high year-round prevalence of C. jejuni in American crows inhabiting an urban-to-rural landscape in Davis, CA. Further, we demonstrated that patterns of infection and environmental contamination are dependent on social behavior and space use; thus, studies focused solely on infection prevalence may serve as a poor proxy for actual transmission risk at a given location and time. Overall C. jejuni prevalence in feces varied between seasons with high and low social aggregation, while patterns of movement and habitat use served to amplify fecal contamination at certain times of the year and in certain locations. As a result, the risk of cross-species transmission of C. jejuni appeared to be concentrated in a predictable pattern across the landscape. Our study is among the first to combine extensive sampling and whole-genome sequencing of C. jejuni with relevant information on host ecology, movement, and social behavior.
Even with very high rates of infection, it is still unclear how important crows are in the overall C. jejuni transmission cycle that includes domestic animals and humans. At the peak of the winter roosting season, approximately 6,000 crows congregated at our study site. We estimated that 51% of those birds were infected with C. jejuni; however, many of the collected isolates may be crow specific (28) and thus might not be a threat for human infection. Furthermore, even with high rates of infection, most infected feces will not result in transmission because they may not contact another host, or the bacteria may die before another host is contacted. Still, with more than 6,000 birds occupying a small area at one time of the year, the cumulative risk of transmission might be high, especially considering the way that nonrandom space use can amplify the risk of transmission in specific areas. For example, the strong habitat associations with a local dairy barn and primate research center likely resulted in extensive exposure to crow feces at those locations during the peak of the roosting season. Even if each fecal deposition event represents a very low risk of cross-species transmission, the collective effect might be large. Although a primate facility is clearly not a typical landscape feature that crows encounter, we expect that similar patterns would be found at any animal facility where crows can access easy and abundant food resources.
Given the high prevalence that we observed in this study, it is likely that crows are continually reexposed to C. jejuni and that these exposures result in repeated infection, or that infection is long-lasting. Unfortunately, the methods used in this study did not allow collection of longitudinal data from the same individuals, so it is not clear how long crows shed C. jejuni in feces after the initial infection. In other wild birds, very little is known about the length of infection or the possibility of cross-strain immunity (16). In one aviary-based study, Waldenström et al. (55) found that experimental infection of European robins (Erithacus rubecula) with C. jejuni isolated from a song thrush (Turdus philomelos) resulted in detectable bacteria shedding for 6.8 days, but infection of the same species with a human-pathogenic strain only resulted in 0 to 1 day of detectable shedding. In contrast, broiler chickens typically remain colonized throughout their lifetime (up to 7 weeks [56]), although experimental administration of different C. jejuni strains can produce variation in the persistence of infection (57). In one case, 3-week-old herring gulls (Larus argentatus) that were naturally infected with C. jejuni were taken into an aviary, and all had cleared their infection by 4 weeks after capture, indicating that the total infection length was less than 7 weeks (58). Although there is some indication that infection may result in acquired immunity, relatively little is known about the effectiveness or strain specificity of immunity, and reinfection by identical or novel strains is common (32, 56). Assuming that infection persists for 1 to 7 weeks in crows, the average individual would need to be exposed to C. jejuni frequently to maintain the 51% prevalence that we observed in the winter. At present, it is uncertain how this high rate of exposure is maintained, but there are a few potentially contributing factors.
Foraging ecology and habitat use may contribute to high rates of crow infection. Foraging guild has been implicated as a driver of C. jejuni infection, and opportunistic ground feeders, such as crows, are among the most colonized groups (20). Unsurprisingly, then, a high prevalence of C. jejuni infection has been found in corvids around the world (15, 25, 28, 29, 59). Our study suggests that social behavior may also be an important consideration in understanding the high rates of corvid infection. For adults, we found that infection was highest in the winter months and lowest in the summer. Although we could not test directly for a causal relationship between roost size and infection prevalence, the seasonal change in prevalence broadly parallels population-level changes in social aggregation. Whether this seasonal pattern is driven by roosting behavior or some other factor is unknown at present. For example, during the winter roosting season, migrants intermingle with local birds, and it is possible that an infusion of novel C. jejuni strains, rather than roosting density per se, drives seasonal differences. We could not evaluate this possibility in our study because samples collected under the winter roost came from birds of unknown migratory origin. Regardless of the mechanism driving seasonal changes in prevalence, the highest infection rates occurred when crows roosted communally in one densely packed area each night and foraged together in large flocks during the day. During these months, up to 58% of individuals had fecal staining on their feathers (31); because birds preen feathers regularly with their beaks, the potential for fecal-oral bacterial transmission is high. In addition to preening, crows often forage socially on highly concentrated food resources, which may result in frequent fecal-oral transmission during foraging. In general, C. jejuni does not survive well in the environment outside the wet and warm conditions of its host (60), but the social behaviors of crows could permit transmission even if C. jejuni survival in feces is brief. For nestlings, we found strong evidence that the infection status of nest mates influenced the risk of infection and that infected nest mates had more similar isolates than expected by chance.
Whether crows represent a major source of domestic animal and, ultimately, human C. jejuni infection remains uncertain, but our study indicates that data on infection prevalence and molecular characteristics of isolates alone will be insufficient for understanding C. jejuni transmission dynamics. Rather, more studies are needed that combine laboratory techniques and sampling of wildlife for C. jejuni with rigorous field work characterizing the ecology, movement, and behavior of potential wild bird vectors. Interestingly, some data suggest that the size of crow roosts has increased and that their locations have shifted from rural to increasingly urban areas over the last 50 years, with unknown impacts on disease dynamics (61). As human-altered landscapes continue to bring wildlife, humans, and domestic animals into increasingly close contact, understanding the shifting ecological interactions and patterns of cross-species transmission will only become more important.

ACKNOWLEDGMENTS

We thank our many U.C. Davis undergraduate assistants for help with field data collection and lab work. Members of the Townsend and Eadie labs at U.C. Davis provided valuable feedback when planning our study. The staff at the U.C. Davis VMTH were instrumental in facilitating the testing of our samples, and the 100K Foodborne Pathogen Genome Project team was facilitated full-genome sequencing and analysis of our isolates.
The project was supported by the Agriculture and Food Research Initiative competitive grant 2014-67012-21622 to C.C.T. from the USDA National Institute of Food and Agriculture, by the 100K Foodborne Pathogen Genome Project, and by the Division of Agriculture and Natural Resources at the University of California at Davis.

Supplemental Material

File (zam999117311so1.pdf)
ASM does not own the copyrights to Supplemental Material that may be linked to, or accessed through, an article. The authors have granted ASM a non-exclusive, world-wide license to publish the Supplemental Material files. Please contact the corresponding author directly for reuse.

REFERENCES

1.
CDC. 2012. Foodborne Diseases Active Surveillance Network: FoodNet surveillance report for 2011 (final report). US Department of Health and Human Services, Centers for Disease Control and Prevention, Atlanta, GA. http://www.cdc.gov/foodnet/PDFs/2011_annual_report_508c.pdf.
2.
Westrell T, Ciampa N, Boelaert F, Helwigh B, Korsgaard H, Chriel M, Ammon A, Makela P. 2009. Zoonotic infections in Europe in 2007: a summary of the EFSA-ECDC annual report. Euro Surveill 14:pii=19100. http://www.eurosurveillance.org/ViewArticle.aspx?ArticleId=19100.
3.
Gras LM, Smid JH, Wagenaar JA, de Boer AG, Havelaar AH, Friesema IHM, French NP, Busani L, van Pelt W. 2012. Risk factors for campylobacteriosis of chicken, ruminant, and environmental origin: a combined case-control and source attribution analysis. PLoS One 7:e42599.
4.
Peterson MC. 1994. Rheumatic manifestations of Campylobacter jejuni and C. fetus infections in adults. Scand J Rheumatol 23:167–170.
5.
Allos BM. 1997. Association between Campylobacter infection and Guillain-Barré syndrome. J Infect Dis 176:S125–S128.
6.
Meldrum RJ, Wilson IG. 2007. Salmonella and Campylobacter in United Kingdom retail raw chicken in 2005. J Food Prot 70:1937–1939.
7.
Zhao C, Ge B, De Villena J, Sudler R, Yeh E, Zhao S, White DG, Wagner D, Meng J. 2001. Prevalence of Campylobacter spp., Escherichia coli, and Salmonella serovars in retail chicken, turkey, pork, and beef from the greater Washington, D.C., area. Appl Environ Microbiol 67:5431–5436.
8.
Ghareeb K, Awad WA, Mohnl M, Porta R, Biarnés M, Böhm J, Schatzmayr G. 2012. Evaluating the efficacy of an avian-specific probiotic to reduce the colonization of Campylobacter jejuni in broiler chickens. Poultry Sci 91:1825–1832.
9.
Sippy R, Sandoval-Green CMJ, Sahin O, Plummer P, Fairbanks WS, Zhang QJ, Blanchong JA. 2012. Occurrence and molecular analysis of Campylobacter in wildlife on livestock farms. Vet Microbiol 157:369–375.
10.
Tsubokura K, Berndtson E, Bogstedt A, Kaijser B, Kim M, Ozeki M, Hammarstrom L. 1997. Oral administration of antibodies as prophylaxis and therapy in Campylobacter jejuni-infected chickens. Clin Exp Immunol 108:451–455.
11.
Yaman H, Ulukanli Z, Elmali M, Unal Y. 2006. The effect of a fermented probiotic, the kefir, on intestinal flora of poultry domesticated geese (Anser anser). Rev Med Vet 157:379–386.
12.
Li X, Swaggerty CL, Kogut MH, Chiang H-I, Wang Y, Genovese KJ, He H, Zhou H. 2010. Gene expression profiling of the local cecal response of genetic chicken lines that differ in their susceptibility to Campylobacter jejuni colonization. PLoS One 5:e11827.
13.
Haughton PN, Lyng JG, Morgan DJ, Cronin DA, Fanning S, Whyte P. 2011. Efficacy of high-intensity pulsed light for the microbiological decontamination of chicken, associated packaging, and contact surfaces. Foodborne Pathog Dis 8:109–117.
14.
Rajkovic A, Tomic N, Smigic N, Uyttendaele M, Ragaert P, Devlieghere F. 2010. Survival of Campylobacter jejuni on raw chicken legs packed in high-oxygen or high-carbon dioxide atmosphere after the decontamination with lactic acid/sodium lactate buffer. Int J Food Microbiol 140:201–206.
15.
Kapperud G, Rosef O. 1983. Avian wildlife reservoir of Campylobacter fetus subsp. jejuni, Yersinia spp., and Salmonella spp. in Norway. Appl Environ Microbiol 45:375–380.
16.
Waldenström J, Griekspoor P. 2014. Ecology and host association in Campylobacter in wild birds, p 265–284. In Sheppard SK (ed), Campylobacter ecology and evolution. Caister Academic Press, Norfolk, United Kingdom.
17.
Benskin CMH, Wilson K, Jones K, Hartley IR. 2009. Bacterial pathogens in wild birds: a review of the frequency and effects of infection. Biol Rev Camb Philos Soc 84:349–373.
18.
Colles FM, Dingle KE, Cody AJ, Maiden MCJ. 2008. Comparison of Campylobacter populations in wild geese with those in starlings and free-range poultry on the same farm. Appl Environ Microbiol 74:3583–3590.
19.
Colles FM, Ali JS, Sheppard SK, McCarthy ND, Maiden MCJ. 2011. Campylobacter populations in wild and domesticated Mallard ducks (Anas platyrhynchos). Environ Microbiol Rep 3:574–580.
20.
Waldenström J, Broman T, Carlsson I, Hasselquist D, Achterberg RP, Wagenaar JA, Olsen B. 2002. Prevalence of Campylobacter jejuni, Campylobacter lari, and Campylobacter coli in different ecological guilds and taxa of migrating birds. Appl Environ Microbiol 68:5911–5917.
21.
Bessell PR, Rotariu O, Innocent GT, Smith-Palmer A, Strachan NJC, Forbes KJ, Cowden JM, Reid SWJ, Matthews L. 2012. Using sequence data to identify alternative routes and risk of infection: a case-study of Campylobacter in Scotland. BMC Infect Dis 12:1–6.
22.
Keller JI, Shriver WG, Waldenstrom J, Griekspoor P, Olsen B. 2011. Prevalence of Campylobacter in wild birds of the mid-Atlantic region, USA. J Wildl Dis 47:750–754.
23.
Pacha RE, Clark GW, Williams EA, Carter AM. 1988. Migratory birds of central Washington as reservoirs of Campylobacter jejuni. Can J Microbiol 34:80–82.
24.
Gardner TJ, Fitzgerald C, Xavier C, Klein R, Pruckler J, Stroika S, McLaughlin JB. 2011. Outbreak of campylobacteriosis associated with consumption of raw peas. Clin Infect Dis 53:26–32.
25.
Hudson SJ, Lightfoot NF, Coulson JC, Russell K, Sisson PR, Sobo AO. 1991. Jackdaws and magpies as vectors of milkborne human Campylobacter infection. Epidemiol Infect 107:363–372.
26.
Cody AJ, McCarthy ND, Bray JE, Wimalarathna HM, Colles FM, Jansen van Rensburg MJ, Dingle KE, Waldenstöm J, Maiden MC. 2015. Wild bird-associated Campylobacter jejuni isolates are a consistent source of human disease, in Oxfordshire, United Kingdom. Environ Microbiol Rep 7:782–788.
27.
Ramos R, Cerdà-Cuéllar M, Ramírez F, Jover L, Ruiz X. 2010. Influence of refuse sites on the prevalence of Campylobacter spp. and Salmonella serovars in seagulls. Appl Environ Microbiol 76:3052–3056.
28.
Weis AM, Miller WA, Byrne BA, Chouicha N, Boyce WM, Townsend AK. 2014. Prevalence and pathogenic potential of Campylobacter isolates from free-living, human-commensal American crows. Appl Environ Microbiol 80:1639–1644.
29.
Ramonaite S, Kudirkiene E, Tamuleviciene E, Leviniene G, Malakauskas A, Gölz G, Alter T, Malakauskas M. 2014. Prevalence and genotypes of Campylobacter jejuni from urban environmental sources in comparison with clinical isolates from children. J Med Microbiol 63:1205–1213.
30.
Muzaffar SB, Hill NJ, Takekawa JY, Perry WM, Smith LM, Boyce WM. 2012. Role of bird movements in the epidemiology of West Nile and avian influenza virus. Hum Wildl Interact 6:72–88.
31.
Hinton MG, Reisen WK, Wheeler SS, Townsend AK. 2015. West Nile Virus activity in a winter roost of American crows (Corvus brachyrhynchos): is bird-to-bird transmission important in persistence and amplification? J Med Entomol 52:683–692.
32.
Colles FM, McCarthy ND, Howe JC, Devereux CL, Gosler AG, Maiden MCJ. 2009. Dynamics of Campylobacter colonization of a natural host, Sturnus vulgaris (European starling). Environ Microbiol 11:258–267.
33.
Townsend AK, Barker CM. 2014. Plastic and the nest entanglement of urban and agricultural crows. PLoS One 9:e88006.
34.
Agresti A, Coull B. 1998. Approximate is better than ‘exact’ for interval estimation of binomial proportions. Am Stat 52:119–126.
35.
Homer CG, Dewitz JA, Yang L, Jin S, Danielson P, Xian G, Coulston J, Herold ND, Wickham JD, Megown K. 2015. Completion of the 2011 National Land Cover Database for the conterminous United States–representing a decade of land cover change information. Photogramm Eng Remote Sens 81:345–354.
36.
Ganesan B, Dobrowolski P, Weimer BC. 2006. Identification of the leucine-to-2-methylbutyric acid catabolic pathway of Lactococcus lactis. Appl Environ Microbiol 72:4264–4273.
37.
Heithoff DM, Shimp WR, House JK, Xie Y, Weimer BC, Sinsheimer RL, Mahan MJ. 2012. Intraspecies variation in the emergence of hyperinfectious bacterial strains in nature. PLoS Pathog 8:e1002647.
38.
Lüdeke CH, Kong N, Weimer BC, Fischer M, Jones JL. 2015. Complete genome sequences of a clinical isolate and an environmental isolate of Vibrio parahaemolyticus. Genome Announc 3(2):e00216-15.
39.
Kong N, Ng W, Cai L, Leonardo A, Kelly L, Weimer BC. 2014. Integrating the DNA integrity number (DIN) to assess genomic DNA (gDNA) quality control using the Agilent 2200 TapeStation system. Agilent Technologies, Santa Clara, CA. http://www.agilent.com/cs/library/applications/5991-5442EN.pdf.
40.
Jeannotte R, Lee E, Kong N, Ng W, Kelly L, Weimer B. 2014. High-throughput analysis of foodborne bacterial genomic DNA using Agilent 2200 TapeStation and genomic DNA ScreenTape system. Agilent Technologies, Santa Clara, CA. http://www.agilent.com/cs/library/applications/5991-4003EN.pdf.
41.
Kong N, Ng W, Lee V, Kelly L, Weimer B. 2014. Production and analysis of high molecular weight genomic DNA for NGS pipelines using Agilent DNA extraction kit (p/n 200600). Agilent Technologies, Santa Clara, CA. https://www.agilent.com/cs/library/applications/5991-3722EN.pdf.
42.
Jeannotte R, Lee E, Arabyan N, Kong N, Thao K, Huang C, Kelly L, Weimer BC. 2014. Optimization of Covaris settings for shearing bacterial genomic DNA by focused ultrasonication and analysis using Agilent 2200 TapeStation. Agilent Technologies, Santa Clara, CA. https://doi.org/10.13140/RG.2.1.2830.4082.
43.
Kong N, Ng W, Foutouhi A, Huang BC, Kelly L, Weimer BC. 2014. Quality control of high-throughput library construction pipeline for KAPA HTP library using an Agilent 2200 TapeStation. Agilent Technologies, Santa Clara, CA. http://www.agilent.com/cs/library/applications/5991-5141EN.pdf.
44.
Kong N, Thao K, Huang C, Appel M, Lappin S, Knapp L, Kelly L, Weimer B. 2014. Automated library construction using Kapa library preparation kits on the Agilent NGS workstation yields high-quality libraries for whole-genome sequencing on the Illumina platform. Kapa Biosystems, Wilmington, MA. https://www.kapabiosystems.com/document/automated-library-construction-using-kapa-library-preparation-kits-agilent-ngs-workstation/?dl=1.
45.
Miller B, van Kets V, van Rooyen B, Whitehorn H, Jones P, Ranik M, Geldart A, van der Walt E, Appel M, Kong N, Huang C, Storey D, Weimer BC. 2015. A novel, single-tube enzymatic fragmentation and library construction method enables fast turnaround times and improved data quality for microbial whole-genome sequencing. Kapa Biosystems, Wilmington, MA. https://www.kapabiosystems.com/assets/App_Note_HyperPlus_APP109001-v1.15_low.pdf.
46.
Darling AE, Mau B, Perna NT. 2010. progressiveMauve: multiple genome alignment with gene gain, loss and rearrangement. PLoS One 5:e11147.
47.
Darling AC, Mau B, Blattner FR, Perna NT. 2004. Mauve: multiple alignment of conserved genomic sequence with rearrangements. Genome Res 14:1394–1403.
48.
Meier-Kolthoff JP, Auch AF, Klenk H-P, Göker M. 2013. Genome sequence-based species delimitation with confidence intervals and improved distance functions. BMC Bioinformatics 14:60.
49.
Auch AF, von Jan M, Klenk H-P, Göker M. 2010. Digital DNA-DNA hybridization for microbial species delineation by means of genome-to-genome sequence comparison. Stand Genomic Sci 2:117.
50.
Saitou N, Nei M. 1987. The neighbor-joining method: a new method for reconstructing phylogenetic trees. Mol Biol Evol 4:406–425.
51.
Boc A, Diallo AB, Makarenkov V. 2012. T-REX: a Web server for inferring, validating and visualizing phylogenetic trees and networks. Nucleic Acids Res 40:W573–W579.
52.
Huson DH, Scornavacca C. 2012. Dendroscope 3: an interactive tool for rooted phylogenetic trees and networks. Syst Biol 61:1061–1067.
53.
Kearse M, Moir R, Wilson A, Stones-Havas S, Cheung M, Sturrock S, Buxton S, Cooper A, Markowitz S, Duran C, Thierer T, Ashton B, Meintjes P, Drummond A. 2012. Geneious Basic: an integrated and extendable desktop software platform for the organization and analysis of sequence data. Bioinformatics 28:1647–1649.
54.
Townsend AK, Clark AB, McGowan KJ, Lovette IJ. 2009. Reproductive partitioning and the assumptions of reproductive skew models in the cooperatively breeding American crow. Anim Behav 77:503–512.
55.
Waldenström J, Axelsson-Olsson D, Olsen B, Hasselquist D, Griekspoor P, Jansson L, Teneberg S, Svensson L, Ellström P. 2010. Campylobacter jejuni colonization in wild birds: results from an infection experiment. PLoS One 5:e9082.
56.
Altekruse SF, Stern NJ, Fields PI, Swerdlow DL. 1999. Campylobacter jejuni–an emerging foodborne pathogen. Emerg Infect Dis 5:28–35.
57.
Ringoir DD, Korolik V. 2003. Colonisation phenotype and colonisation potential differences in Campylobacter jejuni strains in chickens before and after passage in vivo. Vet Microbiol 92:225–235.
58.
Glünder G, Neumann U, Braune S. 1992. Occurrence of Campylobacter spp. in young gulls, duration of Campylobacter infection and reinfection by contact. J Vet Med 39:119–122.
59.
Ganapathy K, Saleha AA, Jaganathan M, Tan CG, Chong CT, Tang SC, Ideris A, Dare CM, Bradbury JM. 2007. Survey of Campylobacter, Salmonella and mycoplasmas in house crows (Corvus splendens) in Malaysia. Vet Rec 160:622–624.
60.
Skelly C, Weinstein P. 2002. Pathogen survival trajectories: an eco-environmental approach to the modeling of human campylobacteriosis ecology. Environ Health Perspect 111:19–28.
61.
Gorenzel WP, Salmon TP, Simmons GD, Barkhouse B, Quisenberry MP. 2000. Urban crow roosts—a nationwide phenomenon?, p 158–170. In Brittingham MC, Kays J, McPeak R (ed), Proceedings of the Ninth Wildlife Damage Management Conference, State College, PA.

Information & Contributors

Information

Published In

cover image Applied and Environmental Microbiology
Applied and Environmental Microbiology
Volume 82Number 151 August 2016
Pages: 4811 - 4820
Editor: T. E. Besser, Washington State University
PubMed: 27260356

History

Received: 13 May 2016
Accepted: 15 May 2016
Published online: 15 July 2016

Permissions

Request permissions for this article.

Contributors

Authors

Department of Wildlife, Fish, and Conservation Biology, University of California, Davis, Davis, California, USA
Lab of Ornithology and Department of Ecology & Evolutionary Biology, Cornell University, Ithaca, New York, USA
Allison M. Weis
Department of Population Health and Reproduction, School of Veterinary Medicine, 100K Pathogen Genome Project, University of California, Davis, Davis, California, USA
Sarah Wheeler
Sacramento-Yolo Mosquito and Vector Control District, Elk Grove, California, USA
Mitchell G. Hinton
Department of Wildlife, Fish, and Conservation Biology, University of California, Davis, Davis, California, USA
Bart C. Weimer
Department of Population Health and Reproduction, School of Veterinary Medicine, 100K Pathogen Genome Project, University of California, Davis, Davis, California, USA
Christopher M. Barker
Department of Pathology, Microbiology, and Immunology, School of Veterinary Medicine, University of California, Davis, Davis, California, USA
Melissa Jones
Department of Wildlife, Fish, and Conservation Biology, University of California, Davis, Davis, California, USA
Ryane Logsdon
Department of Evolution & Ecology, University of California, Davis, Davis, California, USA
Woutrina A. Smith
Department of Veterinary Medicine and Epidemiology, School of Veterinary Medicine, University of California, Davis, Davis, California, USA
Walter M. Boyce
Department of Pathology, Microbiology, and Immunology, School of Veterinary Medicine, University of California, Davis, Davis, California, USA
Andrea K. Townsend
Department of Biology, Hamilton College, Clinton, New York, USA

Editor

T. E. Besser
Editor
Washington State University

Notes

Address correspondence to Conor C. Taff, [email protected].

Metrics & Citations

Metrics

Note: There is a 3- to 4-day delay in article usage, so article usage will not appear immediately after publication.

Citation counts come from the Crossref Cited by service.

Citations

If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download.

View Options

View options

PDF/ePub

PDF/ePub

Get Access

Buy Article
Applied and Environmental Microbiology Vol.82 • Issue 15 • ASM Journals Pay Per View, PPV 25
Journal Subscription
Applied and Environmental Microbiology
ASM members can purchase subscriptions to journals.
Join or renew

Figures and Media

Figures

Media

Tables

Share

Share

Share the article link

Share with email

Email a colleague

Share on social media

American Society for Microbiology ("ASM") is committed to maintaining your confidence and trust with respect to the information we collect from you on websites owned and operated by ASM ("ASM Web Sites") and other sources. This Privacy Policy sets forth the information we collect about you, how we use this information and the choices you have about how we use such information.
FIND OUT MORE about the privacy policy