Get Complete Project Material File(s) Now! »


Formatted for submission to Applied and Environmental Microbiology


Cattle are commonly treated with antibiotics that may be excreted in their urine or feces. Application of manure or composted manure containing antibiotics or antibiotic resistant bacteria (ARB) as a soil amendment may result in transfer to plants. This study was conducted to determine the effects of antibiotic administration and soil amendment practices on microbial diversity and antibiotic resistance of bacteria recovered from the surfaces of lettuce and radishes grown in field using recommended application rates. Vegetables were planted in field plots amended with raw manure from antibiotic-treated dairy cows, composted manure from cows with different histories of antibiotic administration, or chemical fertilizer control (12 plots, n=3). Culture-based methods, 16s rDNA amplicon sequencing, qPCR and shot-gun metagenomics were utilized to acquire the effect of soil amendment on the vegetable bacterial communities and associated resistance genes. Biological amendments resulted in distinct separation of bacterial communities on both vegetables compared to no amendment. Increases in clindamycin resistant bacteria, a class of antibiotics administered to cattle, were noted on lettuce grown in biological soil amendments. Additionally, vegetables grown in manure were associated with increased abundance of specific ARG copies and resistance genes to additional classes of antibiotics. Growth in compost resulted in fewer ARGS on vegetables compared to manure amended soils. This study demonstrates that raw, antibiotic-exposed manure may alter microbiota and the antibiotic resistance genes present on vegetable surfaces. Proper composting of soil amendments as recommended by the USDA and EPA may offer a strategy to mitigate some types of ARGs.


The accelerated dissemination of antibiotic resistant bacteria (ARB) and antibiotic resistance genes (ARGs) throughout the environment is considered one of the largest public health threats of the 21st century (1,2). Every year 2 million Americans contract bacterial infections resistant to one or more antibiotics, amplifying treatment costs and resulting in death for ~23,000 people (1). One strategy to combat this growing problem is to restrict the use of select classes of medically important antibiotics to humans (2). Antimicrobial use in agriculture is broad; from therapeutic treatment of animals, prophylactic prevention of disease and sub-therapeutic growth promotion. In 2015, over 15 million kgs of antimicrobials were distributed to food-producing animals in the USA; 62% of which were considered medically important (3). American dairy cows are commonly administered antibiotics in-between lactation periods and can produce up to 80 lbs of manure per day on a 1,000 animal per unit basis (4). Animals administered antibiotics can excrete more than 70% of some parent compounds in feces and as much as 90% in urine; excretions must be managed and often end up as soil amendments on vegetable crop fields (5-8). The consistent production of antibiotic exposed manure has put pressure on the environmental resistome, selecting for bacterial resistance.
Manure, specifically from cattle, is a known reservoir of ARB and ARGs; application of manure to soil has been shown to increase ARGs detected in soil (9-14). Composting treatments of the manure are known to reduce levels of parent antibiotic compounds, but reduction of ARB and ARGs are variable (15-19). Many studies have concluded that animal waste is a significant source of bacterial contamination on produce however few studies have aimed to document an association between environmental practices of soil amendments and ARB and ARGs associated with fresh produce (21-23). A wide range of ARB and ARGs have been detected on both farm fresh and market ready produce (24-29). Additionally, organic produce, which must be grown in natural fertilizers like manure, have been found to have equal (30,31) or higher levels of ARG containing bacteria in comparison to conventionally grown vegetables (32). Composting of manure using a method validated to reduce pathogenic bacteria is required if the compost will be applied to soils used to grow fruits or vegetables (33). It is not known how composting affects the levels of ARB and ARGs transferred to the surfaces of vegetables grown in said amendments.
In this study, culture-dependent and independent analyses were conducted to evaluate the effect of soil amendment on the bacterial communities, especially quantities of ARB and ARGs detected on the surfaces of lettuce and radishes grown in a clay loam field that, prior to this study, had not been amended with animal amendments or antibiotics for a decade. Biological soil amendments included: raw manure from dairy cows administered pirlimycin and cephapirin, statically composted manure from cows with different antibiotic treatment histories (antibiotic administration or none during collection). Lettuce and radishes grown in soils with the different biological amendments were compared to those grown using a chemical fertilizer. We aimed to characterize the bacterial communities of the vegetable surfaces through sequencing of 16S rDNA amplicons, and enumerate antibiotic-tolerant bacteria using culturing and culture independent methods. Additionally, the classes of putative ARGs recovered from the vegetable surfaces were compared via shotgun metagenomic DNA sequencing. The results will help provide important information on the interactions between vegetables grown in antibiotic exposed soil amendments and the prevalence of antibiotic resistance in the farm-to-fork continuum.

Materials and Methods

Field Study Design

The land utilized in this experiment was located in Virginia Tech’s Urban Horticulture Center (UHC) in Blacksburg, Virginia. Soil was identified as clay loam and further analyzed by Waypoint Analytical (Richmond, Virginia) (Table S1). Prior to this project, animal amendments or antibiotics had not been intentionally applied to the soil within the past decade. The land was divided into 24 (3 m x 3 m) plots; each bordered with steel siding to reduce cross contamination of the soil amendments. The plots were treated with one of four soil amendments: raw dairy cow manure, static-compost with antibiotic exposure, static-compost without antibiotic exposure, and non-amended (Fig 1). Soil was pretreated with Roundup (glyphosate) and non-amended Inorganic Nitrogen-Phosphorous-Potassium (NPK) as it was determined that the soil nutrients were not high enough to support the growth of vegetables without the assistance of NPK. The trial occurred in early Spring, 2016. Rainfall total and average temperatures are available in Wind 2017 (34).

Generation of Biological Soil Amendments

Manure from the pirlimycin- and cephapirin-treated dairy cows were compiled and mixed to form dairy manure with antibiotics (35). Briefly, manure from 18 dairy cattle that had received intramammary administration of pirlimycin (2, 50 mg doses 24 hours apart) and cephapirin (1 dose, 300 mg). Cattle manure was collected over a period of peak antibiotic excretion as determined by Ray 2017 (35). Manure was also collected from cows not currently treated with antibiotics. Manure for composting was combined with alfalfa hay (4:1) and sawdust (4:3) to achieve a Carbon: Nitrogen ratio of 25-30% and a moisture content of 55-65%. The same additions and ratios were reached with generate an antibiotic origin compost (Compost AB) or antibiotic-free compost, referred to as Compost No AB. Both compost treatments were developed using a forced aeration static composting approach following the FSMA guidelines and reached an internal temperature >131°F by day 2 of composting (35, 36). The core temperature of the compost pile remained thermophilic (>131°F) for 21 days.
Manure and compost were added to the vegetable plots at a rate of 6.72 Mg/ha as described by Wind 2017 (34). Manure was stockpiled for 57 days before being applied to six vegetable plots in the raw form. NPK was added to the fertilizer control plots at rates recommended for optimal growth for radish (50 % N-50 % P-50 % K) and lettuce (125 %N -100 %P -100% K) (37). Because the manure/compost was not nutritionally sufficient alone to meet the optimal growth levels, supplemental inorganic N-P-K was also applied to the plots at rates of 50-50-20 for radish and 100-100-75 for lettuce. N-P-K, manure, and compost(s) were applied to the plots on Day 0 of the experiment as shown in Figure 1.

In Field Lettuce Production and Harvest

Lactuca sativa cv. Organic Nancy lettuce seeds (Johnny’s Selected Seeds, Fairfield, Maine) were planted in horticulture vermiculite, hand-watered, and fertilized with inorganic NPK solution. After eight weeks, the seedlings were transplanted into 12 field plots (3 replicates per amendment, Fig. 1) at a stock rate of roughly 54 plants per plot. Transplanting occurred thirty days after application of soil amendments. Lettuce plants were grown until maturity (heads of 12 inches in diameter) and harvested on two separate dates, 38 and 39 days after transplanting (Fig 2a). Soil amendments had been applied 67/68 days prior to lettuce harvest.
Lettuce was harvested on two consecutive days from 6 of 12 plots randomly chosen each day. Temperatures ranged from a high of 68-60 ˚F over the two-day period. The similar average temperatures combined with 0% rainfall created minimal variability between Day 1 and Day 2 samples. The 12 lettuce plots were assigned random individual numerical values. From each plot, six heads of lettuce were selected and harvested. Heads of lettuce with evidence of decay or disease were not selected for analysis. The lettuce heads were removed from the base just above soil level using ethanol sterilized pruning shears; to minimize cross contamination gloves and shoe covers were changed between each plot. The bottom-most leaves (4-6) in direct contact with the soil were removed from the base and discarded before placing into large collection bags which were immediately transported to the lab for analysis. Samples were processed within two hours from harvest.

Radish Production in Field and Harvest

Raphanus sativus cv. Crunchy Royal radish (Johnny’s Selected Seeds, Fairfield, Maine) seeds were sown 30d after application of soil amendments (n=3 per amendment). Seeds were planted 1/8 to 1/4-inch-deep in rows that were roughly 2-3 inches apart (Virginia Cooperative Extension 2015). Radishes were harvested when market ready; when the bulbs began to push out of the soil line; 46/47 days after planting (Fig 2b). Soil amendments had been applied 74/75 days prior to radish harvest. Radishes were harvested on two consecutive days from six of 12 plots randomly chosen each day. Temperatures ranged from a high of 70 to a low of 65 ˚F and no rainfall occurred. These similar average temperatures combined with 0% rainfall created minimal variability between Day 1 and Day 2 samples. Radishes that showed visible signs of decay or plant disease were not selected for analysis. The radishes were pulled from the ground while wearing gloves and shoe covers; to minimize cross contamination gloves and shoe covers were changed between each plot. From each plot, at least ten radishes were harvested, placed into a collection bag; these 12 bags were immediately transported to the lab for analysis. Samples were processed within two hours from harvest.

Enumeration of Aerobic Heterotrophic Bacteria

Bacteria recovered from the surface of radish taproots (75g) or lettuce leaves (25g, 2-3 inner and outer leaves) were enumerated in this experiment. The leafy green tops and fibrous root hairs of radishes were removed aseptically with scissors before processing. Bacterial cells were disassociated from the vegetables using by gently shaking at 220 rpm on a multi-purpose rotator (Fisher Scientific, Waltham, MA) for 5 minutes submersed in a solution of sterile 0.1% peptone (Difco, Becton Dickinson and Company, Franklin Lakes, NJ) with 0.1% Tween 80 (Fisher Scientific) solution. Each sample was then hand massaged for an additional 2 minutes after shaking. For both plant types, 10 ml of the suspension were serially diluted and spread-plated (100 μl) in duplicate onto 7 different types of R2A media (Difco, Becton Dickinson and Company Franklin Lakes, NJ) containing various concentrations of antibiotics (3 μg/ ml tetracycline, 10 μg/ ml ceftazidime, 25 μg/ ml, erythromycin, 25 μg/ ml clindamycin, 25 μg/ ml sulfamethoxazole, and 11 μg/ ml vancomycin) and an R2A control. Antibiotic concentrations were determined by enumeration of bacterial colonies from compost from dairy cattle treated with antibiotics on R2A of differing antibiotic concentrations. Concentrations were chosen by an observed decrease in CFU from the lowest antibiotic concentration tested. Plates were incubated at 37 °C for 24h prior to enumeration.

Nucleic acid Isolation

Immediately after enumeration, the remaining diluent was aseptically filtered through 0.22-μm 47-mm mixed cellulose esters membrane (EMD Millipore, Merck Group, Darmstadt, Germany) to collect bacterial cells. Filters were folded four times, torn, and stored in sterile, DNase-free, O-ring screw cap tubes at -80 °C until DNA extraction. Diluent from each of the 24 samples were processed independently.
The frozen filters were placed into Lysing Matrix E tubes from the FastDNA Spin Kit for Soil (MP Biomedicals, Solon, OH) with the manufacturer’s sodium phosphate buffer and MT buffer. DNA lysis using physical disruption by the FastPrep® Instrument (MP Biomedicals, Solon, OH) occurred after 40 seconds at a speed setting of 6.0. The manufacturer’s instructions were followed except for an additional bead beating step and 2 h incubation a room temperature, allowing for maximized cell lysing. The DNA was resuspended with 100 μL DNase/pyrogen-free water and the tubes were incubated at 55 °C for 5 min. The freshly eluted DNA was then applied to the OneStep PCR Inhibitor Removal Kit (Zymo Research Corporation, Irvine, CA) per manufacturer’s directions before storing at -80 °C in DNase-free, O-ring screw cap tubes. A radish sample grown in antibiotic free compost was lost during the DNA extraction process; n=2 for Compost AB samples being analyzed throughout this experiment for this reason.

Quantification of antibiotic resistance genes

Quantitative real time PCR (qPCR) was used to determine the number of copies of 16S rDNA, tet(w) and sul1 in lysates from bacterial DNA from the surface of the field grown lettuce and radishes. DNA extracts were diluted 1/10 to reduce PCR inhibition. Diluted samples were utilized in 10- μL reactions, which were created for all gene targets. 2x SsoFast EVAgreen Supermix (BioRad Laboratories, Hercules, CA), 20 ng of DNA template and 400 nM primers were combined with 2.4 μL of molecular grade water (Sigma-Aldrich, St. Louis, MO). Triplicate technical replicates of each sample were amplified along with triplicate standard curves and a negative control. The standard curve, comprised of 7, 10-fold dilutions and ranged from 108-102
gene copies/µl for 16S rRNA and 107-101 gene copies/µl for tet(W) and sul1. The negative control was comprised of molecular grade water (Sigma-Aldrich, St. Louis, MO). Samples were amplified in a CFX Connect TM Touch Real-Time PCR Detection System (BioRad Laboratories Hercules, CA). The protocol consisted of 1 cycle of 98 °C for 2 min, 40 cycles of 98 °C for 5s and annealed at various temperatures and times depending on the gene target. 16S rRNA targets were annealed at 55°C, 5 s, tet(W) at 61°C, 7 s and sul1 at 71°C, 7 s followed by a melt curve.

16S rDNA Amplicon Sequencing and Analysis

Illumina 16S rDNA amplicon sequencing was performed on lettuce and radish DNA samples following the Earth Microbiome Project 16S Amplification Protocol version 4_13 (38,39). DNA samples from vegetables grown in each plot were amplified via PCR using unique barcoded bacteria-archaeal primers 515FB and 926R. The amount of DNA used for amplification was normalized to an equivalent 16S rDNA gene copy numbers between all samples before barcoded PCR amplification. Barcoded PCR was performed in triplicate for each sample; products were pooled on an equal mass basis of 200 ng and products purified using QIAquick PCR Purification Kit (QIAGEN, Valencia, CA). The final pooled product was submitted to the Genomics Research Laboratory of the Biocomplexity Institute (BI) of Virginia Tech for paired-end 300 cycle sequencing on the Illumina Miseq. PANDAseq (40) was used to stich the paired-end reads together at a quality score of >0.80 and sequence length of 372-375 bp. The QIIME pipeline (41) was used to annotate the reads to the Greengenes 16S rRNA gene database (42), after which mitochondrial and chlorophyll sequences were filtered out of the OTU table. The samples had a minimum number of reads of 6127 and a maximum of 40750. All samples were rarefied to 6127.

Metagenomic Analysis

Metagenomic analysis was performed on the 12 lettuce and 11 radish samples. Two lanes comprised of 23 undiluted DNA samples, were submitted to the Genomics Research Laboratory of BI. DNA (3 ng) were prepared using the Accel-NGS 2S DNA kit (SwiftBio, Ann Arbor, MI) incorporating 11 cycles of PCR to prepare libraries for high throughput sequencing on Illumina HiSeq 2500 with a high output paired-end 2×100 read length protocol. The paired-end sequence files (one file per end) were transformed into fastq format and then uploaded to MetaStorm (43). MetaStorm is an online platform that allows metagenomics data to be analyzed using a variety of databases. The Comprehensive Antibiotic Resistance Database (CARD v1.0.6) was selected in MetaStorm and used as the ARG functional annotation reference database for the read matched samples (44). The gene counts derived from MetaStorm were normalized to the abundance of 16S rRNA gene to determine the relative abundances of the total detected ARGs (45). The trimmomatic default setting was used in Metastorm, providing 80% nucleotide coverage of each read.

Statistical Analysis

JMP® Pro 12 (SAS Institute, Cary, NC) was utilized for all statistical analyses; p ≤ 0.05 indicated statistical significance for all parametric and non-parametric tests. Plate counts between 25-250 CFU/plate were log-transformed to approximate normal distribution.
The overall effect of soil amendment type was compared using a one-way ANOVA analysis with a Tukey’s post-hoc analysis to test for differences in the average log CFU/g of antibiotic-tolerant bacteria recovered off lettuce and radish surfaces. The same statistical measures (one-way ANOVA, Tukey’s post-hoc analysis) were taken to determine the effect of soil amendments on the antibiotic-tolerant bacteria enumerated (log CFU/g) each individual media type.
Copies ARG (tet(w) and sul1) were normalized by dividing ARG copy numbers /16S rRNA gene copy numbers. The effect of soil amendment type on the proportion of target genes were compared using the nonparametric Wilcoxon coupled with a Steel-Dwass All Pairs test to conduct multiple comparisons. The same statistical measures were used to analyze total gene copies (tet(w) and sul1). Significance between samples was defined as p≤0.05.
The α-diversity estimates acquired from the16s rDNA Amplicon sequencing were calculated by analyzing the observed species, Shannon index and Chao1 values. Values were compared by using Wilcoxon coupled with a Steel-Dwass All Pairs test. Unweighted and Weighted Unifrac distances derived from the β-diversity estimates were plotted in Multidimensional Scaling (MDS) plots in PRIMER-E (version 6.1.13). β-diversity estimates were compared in PRIMER-E using analysis of similarities (ANOSIM) (p≤0.10); levels of separation were defined by Ramette (10). Overall rarefied bacterial compositions derived from 16s rDNA Amplicon sequencing were compared using nonparametric Wilcoxon coupled with a Steel-Dwass All Pairs test. The relative abundances of total ARGs from shotgun metagenomics were compared in PRIMER-E using analysis of similarities (ANOSIM)

Effect of Soil Amendments from Antibiotic –Treated Cows on Antibiotic Resistant Bacteria & Resistance Genes Recovered from the Surfaces of Lettuce and Radishes: Field Study

Related Posts