Recipient Organization
MISSISSIPPI STATE UNIV
(N/A)
MISSISSIPPI STATE,MS 39762
Performing Department
College Of Veterinary Medicine
Non Technical Summary
Infectious bovine respiratory disease (BRD) is the leading cause of sickness and death in U.S. beef cattle, and the bacteria M. haemolytica is the most common bacterial cause of BRD. For cattle at high risk for developing BRD, no management practice has been as reliable as the administration of a long-acting antimicrobial (AM) to the entire group of cattle when they are first purchased (i.e., at arrival). A rule of thumb is that mass medication will decrease expected BRD morbidity by 50%; mortality is often also reduced. Although mass medication has been effective for decreasing BRD in high-risk cattle, recent research by our group and others indicates that the practice may lead to rapid dissemination of M. haemolytica that are resistant to multiple classes of AM, as measured by culture and AM susceptibility testing of nasopharyngeal swabs (NPS) collected from cattle receiving mass medication. The mechanism by which the prevalence of MDR M. haemolytica can increase rapidly in high-risk cattle treated with AM for BRD is not known. One possibility we cannot yet rule out is that our sampling strategy failed to identify some MDR M. haemolytica that were present on NPS collected on d. 0, simply because we did not select enough colonies for characterization. Our approach has been to use standard diagnostic laboratory methodology to isolate and characterize M. haemolytica from NPS. In this approach, a single colony, which may be one of many that are present, is selected from the primary plate for determination of AM susceptibility. When only one colony is selected for characterization, we may fail to identify some MDR isolates. It is likely necessary to select more than one colony to have a high level of confidence that any MDR isolates will be found, but the number that must be selected is not known. A computer program to estimate this number has been developed and described by Döpfer et al (2008). However, in order to use this program, an estimate of the number of different AM susceptibility patterns must be used. Unfortunately, the number of different antimicrobial resistance phenotypes that can be found among M. haemolytica colonies cultured from bovine NPS has not been described or even estimatedTherefore, the objective of the proposed research is to describe the number of different AM susceptibility patterns (phenotypes) that can be identified among M. haemolytica isolates collected from NPS from cattle at high risk for BRD. The rationale for this is that there are no published data to support estimation of the number of colonies that need to be selected from a NPS culture to identify all AMR phenotypes present. This knowledge gap leads to uncertainty that limits our research and the information the research yields. In order to address this knowledge gap, in the spring of 2018 we collected NPS from 20 cattle at high risk for BRD. These NPS were cultured using standard methods and all M. haemolytica isolates (n = 437) identified on the plates were isolated and stored. In this project, we will assess the antimicrobial susceptibility patterns for all 437 isolates using the Sensititre system to determine the minimum inhibitory concentration (MIC) for several antibiotics used to treat BRD. The number of different susceptibility profiles identified from each NPS will be determined, then that information will be used in the program of Döpfer et al to estimate the number of colonies that need to be selected in order to have 95% certainty that all AM susceptibility profiles in the sample have been identified. The work described in this proposal will allow us to address a significant weakness in our research methodology, improving the reliability of research to determine factors that impact AM resistance in cattle managed to prevent BRD. Thus this proposal directly strengthens our ongoing research to determine the factors that impact the dissemination of AM resistance in cattle populations, and to develop validated methods to limit AM resistance and related negative impacts on cattle health, human health, and food safety. This work directly supports the long-range improvement and sustainability of U.S. agriculture by enabling validation of evidence-based strategies to ensure health and welfare of livestock while limiting antimicrobial resistance.
Animal Health Component
(N/A)
Research Effort Categories
Basic
100%
Applied
(N/A)
Developmental
(N/A)
Goals / Objectives
1. To confirm the identity and minimum inhibitory concentration (MIC) for several antimicrobials for multiple Mannheimia haemolytica isolates isolated from individual nasopharyngeal swabs collected from high-risk stocker cattle.2. Using the information obtained for goal #1, to determine the number of unique AMR phenotypes that can be isolated from individual nasopharyngeal swabs collected from high-risk stocker cattle.
Project Methods
The rationale for this project is that there are no published data to support estimation of the number of colonies that need to be selected from a NPS culture to identify all AMR phenotypes present. This knowledge gap leads to uncertainty that limits our research and the information the research yields. To address this knowledge gap, in the spring of 2018 we collected NPS from 20 stocker cattle. Guarded swabs were used to sample calves; briefly, cattle were restrained in a chute and an 83.4 cm (33 inch) double guarded swab (Jorgensen) was inserted into the nose and advanced to 2.5 cm below the medial canthus of the eye; the inner swab was then advanced 2.5 cm further, rotated 3 - 4 times, and withdrawn. One swab was collected through each nostril; the 2 swabs were inserted together into Cary-Blair transport material and transferred to a cooler with ice. Within 6 hours of collection both swabs together were rolled across the first quadrant of 5 sequential blood agar plates. For each plate a sterile loop was used to spread material from the first to the second quadrant, from the second to third quadrant, and from the third to fourth quadrant. Five plates were swabbed because we wished to assess whether AMR M. haemolytica colonies not identified on the first plate (perhaps due to overgrowth by contaminants) would be identified on subsequent plates.The 5 plates swabbed with NPS from each stocker calf sampled (i.e. the primary culture plates) were incubated at 37° C in 5% CO2 for 24 hours. Each plate was examined; if three or more colonies consistent in appearance with M. haemolytica (small white glossy colonies with pale edges) were identified, one colony was subjected to biochemical testing to confirm reactions consistent with M. haemolytica (catalase positive, oxidase positive, and indole negative), and each of the remaining colonies was transferred to and streaked across an individual blood agar plate (subculture). If two or fewer colonies consistent with M. haemolytica were present, because we did not want to lose one of the two or fewer colonies by using it for biochemical testing, they were each subcultured without biochemical testing, and a note was made that biochemical testing was not done. This was considered acceptable because the identity of all isolates will be confirmed with more testing (Sensititre System, ThermoFisher) when the antimicrobial sensitivity of each isolate is determined. Each individual colony that could be picked off each of the 5 primary plates from NPS from each calf sampled was subcultured once. The subculture plates inoculated with individual colonies were incubated at 37° C in 5% CO2 for 24 hours. After 24 hours, for plates with the appearance of pure culture of M. haemolytica, all growth was swabbed off the plate with a sterile swab, transferred to a cryovial containing 1 ml of 50% glycerol in phosphate buffered saline, and stored at - 80° C. For subculture plates that appeared to include growth of a contaminant, the region of the plate with only growth consistent with M. haemolytica was swabbed and transferred to a cryovial of 50% glycerol, and that isolate was noted to be possibly contaminated. Only a small number (14 out of 437) of subculture plates included contaminants. Each subcultured isolate was given a unique identifying number based on the identification number of the stocker calf that yielded the isolate, the number (1 through 5) of the primary plate the isolate came from, and the number (1 through 20) of the colony selected from the respective plate.Of the 20 stocker cattle sampled by NPS, at least one colony with the appearance of M. haemolytica was identified on at least 1 of the 5 primary plates for 13 of the cattle. One of these 13 cattle had been treated with AM before sampling and the other 12 had not. This rate of M. haemolytica identification (13 of 20 = 65%) is consistent with our previous work, in which we isolated M. haemolytica from between 10% and 88% of stocker cattle sampled.7 Of the NPS from which M. haemolytica was isolated, the number of colonies isolated on the first plate ranged from 0 to 16 (mean: 6.2 colonies; median: 6 colonies), and the total number of colonies isolated from all 5 primary plates ranged from 2 to 95 (mean: 33.6 colonies; median: 31 colonies). A total of 437 isolates resulting from this sampling are currently stored at - 80° C in the Dept. of Pathobiology and Population Medicine (DPPM) in the MSU CVM.We will confirm the identity and minimum inhibitory concentration (MIC) for several antimicrobials for each of the 437 stored isolates, using the Sensititre System (ThermoFisher) in the MSU CVM Diagnostic Bacteriology Lab in the DPPM. The Sensititre System determines MIC by broth microdilution. We have confirmed with Dr. Frank Austin (faculty member) and Dr. Bill Epperson (DPPM Head) that the isolates can be tested in the CVM Sensititre system as long as we provide all materials needed for testing, and testing does not interfere with processing isolates from clinic cases submitted to the lab. We estimate that it will take 6 to 12 months to complete the characterization of all 437 isolates, depending on the number of clinical samples the laboratory needs to process during this time. The identity of each isolate will be determined on the Sensititre Gram negative identification plate (ThermoFisher GNIP), and the MIC will be determine using the Bovine/Porcine Plate (BOPO7F). The identity of 3 isolates can be confirmed on one GNIP plate; 1 isolate can be tested on each BOPO7F plate.?Analysis of results: Each isolate confirmed by Sensititre to be M. haemolytica will be identified as sensitive, resistant, or intermediate based the measured MIC and the Clinical Laboratory Standards Institute (CLSI) breakpoints for M. haemolytica in BRD for 5 antimicrobials: ceftiofur, enrofloxacin, florfenicol, tetracycline, and tulathromycin. These antimicrobials are commonly used for the treatment or control of respiratory disease in stocker cattle, and each represents a different antimicrobial class (respectively: beta lactam, quinolone, -amphenicol, tetracycline, and macrolide). For each isolate, the "antibiogram" will be defined by whether the isolate is susceptible or resistant to each of the 5 antimicrobials. For the NPS from each stocker calf, we will determine the number of unique antibiograms identified on the first primary plate, on the second, third, fourth, and fifth primary plates, and on the sum of all 5 primary plates. This will help us determine whether we are able to identify all unique antibiograms on the first plate, or whether in future research to characterize factors that impact AMR we will need to swab more than one plate to identify all unique AMR phenotypes.Once we know the number of unique AMR phenotypes obtained from the NPS collected, we can use this information in the program developed by Döpfer et al. (Dopfer D, Buist W, Soyer Y, et al. Assessing genetic heterogeneity within bacterial species isolated from gastrointestinal and environmental samples: how many isolates does it take? Appl Environ Microbiol 2008;74:3490-3496.) to estimate the number of colonies that need to be selected from NPS culture plates provide 95% certainty that we have identified all the antibiograms (i.e. phenotypes) on the NPS. The program of Döpfer et al. uses Bayesian statistical inference in a WinBUGS code to calculate the number of isolates that need to be selected. The program requires an estimate of the number of different phenotypes likely to be identified ("prior probabilities"); the data collected in the work proposed here will provide the prior probabilities required by the program. Although CRIS funds are not adequate to support genotyping the isolates, we are actively looking for support genotype all the isolates by PFGE. The number of unique genotypes may not be the same as the number of unique phenotypes; we hope to assess this in the future.