| Resum: |
Wastewater surveillance is based on monitoring pathogens in an aquatic environment. This technique can be useful to track potentially zoonotic diseases harbored in farm production through the slurry, in which methods like passive and grab samplings are used to collect and trace this residual waste. Therefore, the aim of this research was to conduct wastewater genomic surveillance in a swine farm using nanopore sequencing to identify possible pathogens and antibiotic resistance gens. In addition, we want to assess the effect of different sampling techniques for wastewater microorganism monitoring. The study was performed in an experimental swine farm at the Faculty of Veterinary Sciences in the Universitat Autònoma de Barcelona. In the study set-up, 26 piglets (2 pens of 13 piglets each) were sampled (rectal swabs) and the slurry was manually collected using a Moore swab (passive sampling) and a sterile immersion bottle (grab sampling) from the collection basin. Sampling was performed on four different dates. DNA was extracted and pooled by type of sampling (rectal swab, Moore swab and immersion bottle), biological replicate (pen A vs pen B) and by date. The sequencing was performed using nanopore sequencing, from Oxford Nanopore Technology. Sequencing reads were analyzed using the platform EPI2ME to classify them based on taxonomy and to detect antimicrobial resistance genes (AMR). The reads obtained were rarefied and data was analyzed with the Vegan R package. Alpha diversity was estimated in each sample with two different metrics: Richness and Shannon index. No statistical differences were found on Richness (p = 0. 243, p = 0. 126, p = 0. 741) or Shannon index (p = 0. 3135, p = 0. 1002, p = 0. 5140) between sampling techniques (rectal swab, Moore swab and immersion bottle), implying that each microbial community is similarly diverse. Regarding beta diversity, a pairwise comparison of Bray-Curtis dissimilarities was conducted between types of sampling. No statistical difference in Bray-Curtis dissimilarities was observed between Moore swabs and immersion bottles (p = 0. 65), indicating that both methods retrieve similar microbial community structures. In contraposition, the microbiome community structure of rectal swabs and Moore swab communities, and rectal swabs and immersion bottles were statistically different (p = 0. 043 and p = 0. 043), showing that the microbial community structure is different in swine rectum/feces and farm slurry. Regarding taxonomy, the most abundant genus found in rectal swab samples was Prevotella spp (10 - 11%). Contrary, the most abundant genus in Moore swab and immersion bottle samples was Arcobacter spp (25 - 54%). Even though the microbiome community structure of rectal swab and the slurry samples are statistically different, it is worth noting that several genera, relevant for farm health management, can be identified in all sampling methods. Per example, the genus Clostridium spp can be traced in the three communities with an abundance ranged between 2% - 7%, in which species like C. difficile, C. butyricum and C. perfringes were the most frequent. Regarding AMR genes, no significant difference was found between the AMR genes detected in the three sampling methods (p = 0. 6592, p = 0. 5637 and p = 0. 8845), implying that all methods recover AMR genes similarly. The most frequent gene groups detected in the three types of samplings were ANT(6)s and the groups of tetracycline resistant genes: tet40, tet44, tetQ, tetW and tetO. To conclude by using both methods of sampling it has been able to detect and trace ARM genes in farming slurry. Although, microbiome structure between rectal swab and Moore swab/immersion bottle has been statistically different, it is important to highlight the potential of long read DNA sequencing to detect several potential pathogenic organisms as C. butyricum, C. perfringes, Arcobacter cryaerophilus or Prevotella dentalis in the three types of samplings. |