Characterization of the impacts of water line cleaning and disinfection on swine wean-to-finish water line ecology

Abstract

Water is an essential yet poorly understood component of swine production, with direct impacts on swine health, performance and biosecurity. The formation of biofilms inside swine water lines creates challenges for maintaining optimum drinking water quality, available water access, biosecurity and water-administered antimicrobial treatment outcomes. Biofilms and mineral scale in water lines can decrease pipe diameter, reducing the amount of water available to pigs and could impact growth performance. Biofilms can harbor primary and opportunistic pathogens, antimicrobial resistance genes, antimicrobial resistant bacteria, and decrease water quality through dispersion into the bulk drinking water. Despite the recognition that water quality is essential to multiple components of swine production and health, little is known about water line ecology in swine farms and how to best manage it. This dissertation’s objectives were to evaluate swine water line ecology with focus on quantifying biofilm regrowth dynamics, evaluate water quality dynamics and detect antimicrobial resistance genes, and finally, identify antimicrobial resistant bacteria and characterize pathogenicity while evaluating the impacts of a water line cleaner and disinfectant, peracetic acid. One central sample population was utilized to analyze several components of baseline and post-water line treatment swine water line ecology parameters to meet each of the chapter’s study objectives. This work is highly relevant as it occurred in-situ, to accurately assess on-farm conditions versus an in-vitro assessment. These studies were observational, longitudinal studies that evaluate a terminal (one-time) application of 0.78% peracetic acid (PAA) on water line ecology in six wean-to-finish sites in Iowa on well water. This dosing rate was off-label as most swine farms are equipped with dosing equipment at 0.78% dosing rate instead of the on-label dose of 2%. The PAA was applied into the water line and allowed a contact time of 24 hours sites in between groups of pigs. Water and water line biofilm samples were collected at 10 sample collection days over a 78-day period. Pre-treatment (0) samples were collected prior to administration of 0.78% PAA, then samples were collected 24 hours post-treatment and fresh water line flush (1), and days 3, 5, 7, 14, 21, 42, 56, and 77 post-treatment to assess the varying components of water line ecology. Samples were collected from May 2023- November 2023 and aseptically in field conditions. Water samples were collected from the source water (well), and from two flush valves located on coupon side streams in the rooms of the site. Water line biofilm samples were collected from removable polyvinyl chloride (PVC) pipe sections installed in each room in a “coupon side stream”. At each sample collection date in each room one 7.62 cm section of PVC pipe was cut out to be submitted for next generation sequencing to identify antimicrobial resistance genes (ARGs) and integron genes. One 2.54 cm PVC section was collected for plating organisms on chlortetracycline (CTC) drug-infused culture plates, and one 2.54 cm PVC section was collected for lincomycin (Linco) drug-infused culture plates. One 2.54 cm sample was collected for aerobic standard plate counts (ASPC) and one 2.54 cm section was collected for anaerobic standard plate counts. In total, at each time point from one site, three water samples were collected per site (total for entire study n=180), two 7.62 cm sections of PVC pipe (total for entire study n=119, one sample became contaminated during sample collection and was excluded from analysis), two 2.54 cm sections of PVC pipe for CTC drug-infused plates (total for entire study n=120), two 2.54 cm section of PVC pipe for Linco drug-infused plates (total for entire study n=120), two 2.54 cm sections of PVC pipe for ASPC (total for entire study n=120), and two 2.54 cm sections of PCV pipe for ANSPC analysis (total for entire study n=120). Biofilm regrowth dynamics were compared to pre-treatment biofilm quantities. There was significant biofilm reduction (adjusted p-value = 0.0000) post-treatment (1) with an average 3.38 log reduction in colony forming units/ mL (CFU/mL). This impact was temporary, as biofilm quantities 3 days post-treatment were not significantly different than pre-treatment levels. Biofilm quantities actually peaked and had higher estimates 7 days-post treatment than pre-treatment quantities. This demonstrates that 0.78% will not completely eliminate biofilms from the water distribution system, suggesting that continuous water line disinfection may be necessary for the continuous management of biofilms on swine farms long-term. Chemical and microbiological water quality parameters were assessed for each of the six sites which demonstrated that generally there was no significant difference in water sample location or impact on the chemical components of water quality over sample collection dates and between water sample collection locations (well vs. rooms). Significant findings were confined to total coliforms, zinc, TDS, pH, and water temperature. Total coliforms were significantly different pending the water sample location, as rooms had significantly higher coliforms than well samples. Total coliforms had no significant differences between pre-and post-treatment sample collections, and quantities numerically peaked at 7 days post-treatment compared to pre-treatment. Total dissolved solids demonstrated significant differences between wells and rooms, and the interaction between water sample location and the sample collection date. Zinc demonstrated a significant difference for the interaction term of water sample location and the sample collection date and was approaching significance for the sample collection date. Water pH was significantly different depending on the location at which the water was sampled, wells having a higher pH than rooms, and water temperature was significantly different over sample collection dates, with temperatures decreasing over time. Bacteria in biofilms are naturally more resistant to antimicrobials than planktonic bacteria. They can also harbor and transfer antimicrobial resistance genes (ARGs). Applying 0.78% PAA may reduce biofilms in water lines and reduce the presence and spread of ARGs. ARGs and integron genes were identified through next generation sequencing from biofilm samples. This study demonstrated that there is a high level of prevalence of ARGs in swine water line biofilms over all samples with 115 samples detecting presence of ARGs and four samples with no detections. A total of 3904 ARGs were reported (n=115), with 184 unique ARGs identified. These ARGs can confer resistance medically important antimicrobials both to humans and animals. The most frequently reported ARGs in terms of drug class were the aminoglycosides, followed by beta-lactams, and tetracyclines. Integron genes, a mechanism for the spread of multi-drug resistance, were also detected 151 times with representation from three integron classes. ARGs and itegron genes post-treatment (1) were significantly lower on average by 10 unique ARGs/integrons from pre-treatment counts (p-value=0.01). ARGs and integron genes over time increased in frequency post-treatment with peak ARG/ integron gene counts at day 14. Sample collection day 14 was significantly different than pre-treatment (0) counts (adjusted p-value = 0.0127). This study demonstrated that water line biofilms are a reservoir for ARGs and the potential spread of multi-drug resistance through the presence of integron genes. Although 0.78% PAA treatment improved ARGs/ integrons immediately post-treatment, long term strategies for reduction of ARGs/ integrons are needed. Swine water line biofilms could be a potential source for opportunistic and primary pathogens for swine and have demonstrated genetic resistance. Clinically, it is important to understand if there are organisms that are displaying phenotypic resistance (PR) demonstrated from water line biofilms to frequently used water medications, chlortetracycline (CTC) and lincomycin (Linco), at concentrations at which we apply through the water lines (much higher than an MIC). Application of 0.78% PAA in water lines may remove antimicrobial resistant bacteria in water line biofilms. In this study, we extracted and plated biofilm organisms onto drug-infused culture plates with CTC or Linco infused into 5% sheep’s blood agar plates. Organisms were classified as resistant if they were demonstrating acquired resistance, and were categorized as pathogenic, opportunistic or unknown pathogenicity/normal flora. This study demonstrated high prevalence of phenotypic resistance with 71% of bacterial isolates grown on the plates demonstrating PR. Swine pathogens, opportunistic pathogens and normal flora were identified within the samples. One hundred and twelve unique organisms containing bacteria, fungi and yeasts were identified in the isolations. Pathogenic F18 Escherichia coli was isolated twice from biofilm samples, which demonstrated resistance to CTC and Linco. Application of 0.78% PAA significantly increased the probability of having zero resistant isolates (p-value =0.008) post-treatment (1). However, long term effects demonstrated similar results to other studies in this dissertation that significant elevations in phenotypically resistant isolates were appreciated on days 5, 14 and 21 (adjusted p-value = <0.05). Results from this study demonstrate that biofilms contain a diverse microbial community which include pathogens, opportunistic pathogens, and normal flora of swine that can demonstrate PR to commonly used water medications for pigs. This research demonstrates that water line biofilms in swine farms can grow back quickly after a one-time application of 0.78% PAA, degrade water quality, pose a biosecurity risk, and can be an area to improve antimicrobial stewardship and management. More research is needed to determine the best practices and interventions for managing water line biofilms in swine farms.