Cattle population and management
All sampled cattle in this report resided at the USMARC in Clay Center, Nebraska, and sampling protocols were approved by the Institutional Animal Care and Use Committee at the University of Nebraska, Lincoln (IACUC, #1383). Veterinary intervention and sampling were requested by the owner soon after the abortions began. The entire cattle population was considered closed; bovine semen was the only source of new genetic material and biosecurity measures were in place to prevent direct contact with outside cattle. Staff and equipment were shared between cattle locations within USMARC.
As a part of routine management practices, all females were previously vaccinated for Brucella abortus and given an initial Campylobacter fetus and Leptospira canicola-grippotyphosa-hardjo-icterohaemorrhagiae-pomona bacterin (Spirovac VL5, Zoetis, Kalamazoo, MI), followed by a modified live IBR, BVD types 1 and 2, and PI3 vaccine in combination with C. fetus, and Leptospira spp. bacterin (PregGuard Gold FP 10, Zoetis, Kalamazoo, MI) 30 days before the 2013 breeding season. Pregnancy was diagnosed via rectal ultrasound between 45-100 days gestation. A total of 28 females (27 two-year old heifers and 1 three-year old female, herein referred to as “heifers”) aborted in late winter/early spring of 2014. Twenty-one of these cases were sampled (described below) and diagnostics performed at the UNLVDC in Lincoln, Nebraska.
All pregnant females were on pasture prior to, during, and after all sampling. During the winter months, all cattle were supplemented primarily with corn silage stored and fed from the same feed storage site. Silage was analyzed for nutritional content and combined with mineral supplementation to meet nutritional requirements. Rations were delivered by the same truck and dispensed in mobile bunk or tire feeders. These feeders were moved periodically when conditions around the bunks were deemed unsanitary. Heifers that aborted were from one of three management sites (A-C) that were managed in a similar fashion. A site map of the operation with feed delivery pathway is identified in Figure 1.
Clinical case sampling
Twenty-eight heifers and four mature cows (four years or greater) aborted by the end of the 2014 calving season, with 24 heifer abortions occurring over 12 days. All collected tissues and sampling of females that had aborted were performed by or under the direct supervision of a licensed veterinarian (lead author). Aborted fetal tissues, including fresh and fixed heart, lung, liver, spleen and kidney, as well as stomach contents and ear notches, were submitted to the UNLVDC on day one and day eight of the investigation,, the days that they were discovered and recovered by cattle managers. All fetal tissues were placed in insulated cooler with icepacks and shipped overnight to the UNLVDC.
Of the 28 heifers that were identified by cattle managers and suspected of aborting, 21 were humanely restrained in a commercial cattle chute and were subjected to individual examination and sample collection (day 8, 14, or 27). The perineum of each female was cleaned with soap and water, rinsed and dried, and the tail held away from the region. If RFM were present, a clean, gloved hand was inserted into the vagina and the membranes were extracted caudally until the hand and membranes were outside the vulva, then a section of membrane was aseptically removed and placed in a sterile collection bag. In heifers that did not have RFM, a uterine culture double-guarded swab (Jorgensen Laboratories, Loveland, CO) was used to sample near and within the cervical opening (CV swab). To facilitate the entry of the swab, a disposable lubricated vaginal speculum was used to visualize the cervical opening and the area just proximal to the opening. Approximately 1-2 cm of the distal cervix and the area just proximal to the opening were swabbed. This method prevented undue contamination from the vagina, which in many cases contained discharge and fluid as a result of metritis or pyometra. Once swabs were collected, samples were contained within the protective unit of the swab mechanism. All case samples were placed in insulated cooler with icepacks and shipped overnight to the UNLVDC. Heifers were released after sampling into communal pens for health observation, then subsequently released back onto pasture.
Feed and water sampling for L. monocytogenes and pH
Supplemental feed consisted of corn silage, earlage, and haylage separated into open concrete bunkers. Corn silage and earlage were sampled by visually dividing the piles into three stratified layers (top, middle and base) and crosswise sections (left, middle, and right) to provide targeted and documented sampling locations (Additional file 1). Samples were obtained by first brushing away loose materials on the surface that were potentially transferred or contaminated by feeding equipment. Exposed feed was then collected at a depth of approximately 10 cm. Additional samples included loose surface material from the center face and apron of the piles as well as loose material present in the drainage tube of each pile. Samples were collected from haylage in a similar fashion as silage piles but from only three of the quadrants: center-top, mid-left and bottom-right. All samples were collected with freshly gloved hands, placed in sterile bags, and transported to the laboratory where they were held at 4 °C until processing. Silage and earlage pile samples were collected a second time, ten days after the initial sampling, and were processed in the same manner. The second sample was taken after removal of at least one meter of the silage face to determine if contamination existed deeper within the silage piles, or if contamination was only present in isolated areas. Thirty-six feed samples were collected over this period.
Accessible water either from a tank or standing water in close proximity to the tank was collected where clinical cases of listeriosis had been identified. Nine water samples (three from each management site) were obtained by placing a 50 mL screw cap conical test tube into the water source. A scooping motion was used, such that the bottom of the tank or water hole was contacted by the tube to capture any sediment present. Tubes were capped and transported immediately to the laboratory for processing. Water sources were only sampled once. After sampling, water tanks were drained, disinfected, allowed to dry, and refilled.
To measure the pH of the feed, collected samples were placed in a 100 mm petri dish then wetted with approximately 10 mL sterile distilled deionized water. A surface pH meter (ExTech Instrument Corp., Nashua, NH) was used to measure the pH of the moistened surface. For each dish of silage, average pH was determined by measuring pH at three separate areas corresponding to 12, 4 and 8 o’clock positions in the petri dish.L
L. monocytogenes detection and isolation from clinical samples
All diagnostic specimens collected during the case investigation were shipped overnight in insulated coolers with icepacks to the UNLVDC. Submitted samples included fresh and fixed fetal tissues (collected and submitted on day 1 and 8) and CV swabs or RFM from the 21 sampled heifers (day 8, 14, and 27). Requested diagnostics for fetal tissues included: aerobic culture and sensitivity, C. fetus culture, Leptospira spp., BHV1, and BVDV PCR, and histopathology. Fetuses and fetal tissues, including spleen, heart, lung, kidney, and liver were fixed in 10% neutral buffered formalin, embedded in paraffin, sectioned and stained with hematoxylin and eosin, examined histopathologically by board certified pathologists with other ancillary testing at their discretion to determine a diagnosis. Testing at UNLVDC is by standard operating procedures accredited under the American Association of Veterinary Laboratory Diagnosticians (AAVLD). Fetal tissues (lung and liver) and stomach contents were subjected to culture testing. Requested diagnostics on the CV and RFM samples collected included aerobic culture and bacterial identification.
Fetal tissue samples were flame sterilized and directly plated on to TSA with 5% sheep’s blood, chocolate agar, Colombia CNA agar with naldixic acid and 5% sheep’s blood (CNA), Campy CVA agar and MacConkey’s agar (Thermo Fisher Scientific, Waltham, MA). Cervico-vaginal swabs, RFM, and fetal stomach contents were directly plated onto the same agar media as the tissues. Remaining samples were macerated (tissues) or agitated (swabs) into Fraser’s Broth (Thermo Fisher Scientific, Waltham, MA). All media except Campy CVA were incubated for 18-24 hours at 37º C with 5% CO2 supplementation, then examined by trained laboratory technicians following the UNLVDC Standard Operating Procedures. Media without pathogenic bacterial growth were re-incubated and observed following an additional 18-24 hour incubation. Campy CVA agar was incubated 48 hours at 37º C in a GasPak EZ Campy microaerophilic environment container (BD Diagnostics, Sparks, MD). Bacterial colonies with morphology consistent with members of the genus Listeria were sub-cultured onto TSA with 5% sheep’s blood for purity.
Sub-cultured, suspect Listeria colonies were subjected to gram staining, catalase testing, and were phenotypically tested using a commercial identification platform using manufacturer’s instructions for Protocol A (Biolog, Omnilog, Hayward, CA). Samples that did not have growth consistent with L. monocytogenes or L. ivanovii on primary isolation media were subjected to PCR testing specific for Listeria hemolysin (hly). Nucleic acid was extracted from culture samples in Fraser’s media using a commercial DNA extraction kit (Qiagen, DNA mini kit and QIACube) per manufacturer’s instructions for bacteria. Samples that were negative following PCR testing were placed into cold enrichment (4 ºC) for six weeks. Cultures with a lack of esculin hydrolysis in Fraser media after six weeks of cold enrichment were considered negative for L. monocytogenes. No additional diagnostics were performed on L. monocytogenes negative submissions.
L. monocytogenes detection and isolation from feed and water
Feed and water samples were screened for the presence of Listeria spp and L. monocytogenes using BAX System Real-Time PCR Assays (Dupont Qualicon, Wilmington, DE) and Atlas Detection assays (Roka Bioscience, Lake Forest Park, WA). Water samples were screened directly and after culture enrichment, while feed samples were only screened after enrichment. For direct screening, 1 mL of each water sample was placed in an Atlas G2 Sample Tube (Roka Bioscience) and subjected to testing. Feed samples were enriched for rapid screening by mixing 50 g feed into 200 mL of Listeria enrichment broth (LEB; Dupont Qualicon, Wilmington, DE) and water samples were enriched by diluting 10 mL of each water sample into 90 mL of LEB. Samples were incubated at 30 °C for 24 h. After incubation, a 1mL portion was placed into an Atlas G2 Sample Tube (Roka Bioscience) and a 20 mL portion was used to prepare a BAX template lysis. All Atlas G2 sample tubes were processed through the RokaBioscience Atlas instrument using the Atlas Listeria LSP Assay and the Atlas LmG2 Assay for Listeria spp and L. monocytogenes respectively. The BAX lysis preparations from each sample were processed through the BAX Q7 instrument using a BAX System Real-Time PCR Assay for L. monocytogenes according to the manufacturer’s instructions.
To isolate and confirm L. monocytogenes from the feed and water samples the 24 h LEB enrichments were streaked for isolation onto a Difco Oxford agar plate (Beckton Dickinson and Co., Franklin Lakes, NJ) and a CHROMagar Listeria plate (DRG International, Inc., Springfield, NJ) using a sterile cotton swab and inoculating loop. Plates were incubated at 37 °C overnight then viewed for suspect colony phenotypes; black colonies on Oxford agar for Listeria spp. and blue colonies without and with halos on chromogenic agar for Listeria spp. and L. monocytogenes respectively, were targeted. Suspect colonies were selected and placed into a 96-well block containing 1 mL per well of Fraser media containing 5% ferric ammonium citrate, and then incubated at 37 °C overnight.
The 24 h LEB enrichments of feed and water were further incubated for another 24 h (48 h total) and the above streaking for isolation onto chromogenic Listeria and Oxford agar was repeated. A secondary 48 h enrichment in Fraser media was incubated an additional 48 h at 30 °C and was streaked for isolation, incubated and viewed for suspect colonies as described above.
Suspect colonies were confirmed to the species level using a Listeria spp. specific PCR and biochemical tests. Listeria species monocytogenes, innocua, grayi, ivanovii, seeligeri and welshimeri were identified through Listeria spp. multiplex PCR [36]. The isolates that were identified as L. monocytogenes were further characterized using the serovar multiplex PCR [21]. Suspect isolates that were found to possess the phosphoribosyl pyrophosphate synthetase (prs) gene, indicative of all Listeria spp., but which could not be identified through PCR were further identified using biochemical tests. Each suspect Listeria was streaked for isolation on tryptic soy agar containing 0.6% yeast extract, incubated at 37 °C overnight and then processed using a Remel Micro-ID Listeria Kit (Thermo Fisher Scientific, Lenexa, KS) according to the manufacturer’s protocol.
L. monocytogenes strains selected for sequencing
A total of 31 L. monocytogenes strains were selected for whole-genome DNA sequencing on a MiSeq instrument (Illumina, San Diego, CA). Of those, 19 clinical strains were isolated from USMARC cattle aborted fetuses, placentas, or uterine swabs. Another eight strains were isolated from four different corn silage samples, three water tanks and one standing water source. Additionally, four other strains from bovine abortions were obtained from the UNLVDC that originated from other regions in Nebraska for use as controls and references
DNA preparation and Illumina MiSeq whole-genome sequencing
All 31 L. monocytogenes strains were passaged twice from -80 °C frozen stocks on chocolate agar plates (Hardy Diagnostics, Santa Maria, CA) at 37 °C. A single colony of each isolate was then inoculated in 1.5 mL of Brain-Heart Infusion (BHI) broth and grown overnight without shaking at 37 °C. Genomic DNA was extracted from the cultures using MO BIO microbial DNA isolation kits (MO BIO Laboratories, Carlsbad, CA) according to the manufacturer’s instructions. The extracted DNAs were quantified and checked for purity using 260/280 absorbance readings on a NanoDrop ND-1000 spectrophotometer (NanoDrop, Wilmington, DE). Individual libraries were constructed for each of the strain DNA preparations using Illumina Nextera XT DNA sample preparation kits with appropriate indices tags according to the manufacturer’s instructions (Illumina Inc., San Diego, CA). The libraries were pooled together and run on an Illumina MiSeq DNA sequencer (Illumina Inc., San Diego, CA). The genome of each strain was sequenced to a minimal depth of 10X coverage.
Assembly of L. monocytogenes chromosomes and phylogenetic trees
Adapter sequence and low-quality bases were trimmed using Trimmomatic [37]. Trimmed fastq reads were assembled using SKESA. The L. monocytogenes assembled chromosomes from 27 unique USMARC strains and 4 clinical strains from the UNLVDC were imported into parsnp [38] for genome alignments and subsequent identification of core-genome SNPs and construction of a preliminary phylogenetic tree. The strains grouped into six initial clusters. One strain from each cluster was selected for PacBio sequencing to obtain a complete closed chromosome representative of each cluster (described below). Illumina reads from strains not also sequenced with PacBio were assembled using SKESA. The SKESA, PacBio, and 20 closed L. monocytogenes chromosomes from GenBank [39] were then used in parsnp to create a new phylogenetic tree. The 20 GenBank chromosomes represented all four known L. monocytogenes lineages and were utilized to calibrate the final phylogenetic tree. Evolview was used to populate the tree with phenotypic metadata [40].
DNA preparation and PacBio whole-genome sequencing library construction
High molecular weight DNA was extracted from L. monocytogenes cultures using Qiagen Genomic-tip 100/G columns and a modified manufacturer’s protocol as previously described [41] with the addition of mutanolysin with the proteinase K step followed by incubation at 50 °C for 1 h. Ten micrograms of DNA were sheared to a targeted size of 20 kb using a g-TUBE (Corvaris, Woburn, MA) and concentrated using 0.45X volume of AMPure PB magnetic beads (Pacific Biosciences, Menlo Park, CA) following the manufacturer’s protocol. Sequencing libraries were created using 5 µg of sheared, concentrated DNA and the PacBio DNA Template Prep Kit 2.0 (3Kb - 10Kb) according to the manufacturer’s protocol. The library was bound with polymerase P5 followed by sequencing on a Pacific BioSciences (PacBio) RS II sequencing platform with chemistry C3 and the 120 min data collection protocol.
PacBio Sequence Assembly into Closed Circularized Genomes
PacBio reads were assembled using HGAP3 (SMRTanalysis Version 2.1) and the resulting contigs were imported into Geneious. Within Geneious, overlapping sequence on the ends of the contigs were removed from the 5’ and 3’ ends to generate a circularized chromosome. The chromosome was reoriented to start with a putative origin of replication with Ori-Finder [42]. The chromosome was initially polished for accuracy using the Resequencing 1.0 protocol in SMRTanalysis by mapping corrected PacBio reads to the chromosome. To correct PacBio sequencing errors (homopolymers and SNPs), Illumina reads were mapped to the initially polished chromosome using Pilon. Then, both PacBio and Illumina reads were mapped to the Pilon-generated chromosome using Geneious Mapper. Additional sequencing errors were identified and corrected by manual editing in Geneious, resulting in a finished closed circularized chromosome. Chromosome sequences were deposited into NCBI (Additional file 2) and annotated using the Prokaryotic Genome Annotation Pipeline version 4.5. MLST 2.0 [43] was used to determine the MLST allelic profiles for the L. monocytogenes strains from USMARC and UNLVDC. MLST allelic profiles were then used with the Listeria Pasteur MLST database to determine serotype, lineage and clonal complex [21, 22].