Prevalence of Multidrug-Resistant Pseudomonas aeruginosa Isolated from Dairy Cattle, Milk, Environment, and Workers’ Hands

Pseudomonas aeruginosa is an opportunistic pathogen causing severe infection in animals and humans. This study aimed to determine the ecological distribution and prevalence of multidrug-resistant (MDR) P. aeruginosa isolated from dairy cattle, the environment, and workers’ hand swabs. Samples (n = 440) were collected from farms and households (n = 3, each). Rectal swabs, udder skin swabs, milk, workers’ hand swabs, feed, water, water sources, and beddings were collected. Samples were subjected to the bacterial identification of P. aeruginosa via 16S rRNA. Antimicrobial resistance (AMR) was detected either phenotypically using an antibiotic susceptibility test or genotypically with AMR resistance genes (ARGs) such as drfA, sul1, and ermB. P. aeruginosa was detected on dairy farms and households (10.3–57.5%, respectively), with an average of 23.2%. The resistance of dairy farm strains was observed against sulfamethoxazole, imipenem, cefepime, piperacillin–tazobactam, and gentamycin (100%, 72.7%, 72.7%, 68.8%, and 63.3%, respectively). Meanwhile, the resistance of household strains was observed against sulfamethoxazole, imipenem, amoxicillin, gentamicin, cefepime, and erythromycin by 91.3%, 82.6%, 75.4%, 75.4%, 68.1%, and 63.8%, respectively. The susceptibility of farm strains was detected against norfloxacin, ciprofloxacin, and levofloxacin (90.9%, 84.8%, and 72.7%, respectively). Meanwhile, the susceptibility of household strains was detected against ciprofloxacin, amikacin, and norfloxacin (100%, 84.1%, and 72.5%, respectively). About 81.4% of P. aeruginosa strains were MDR. ARGs (drfA, sul1, and ermB) were detected in farm strains (48.5%, 72.7%, and 24.4%, respectively) and household strains (50.7%, 72.5%, and 47.8%, respectively). Almost all P. aeruginosa had MAR over 0.2, indicating repeated application of antibiotics. P. aeruginosa prevalence was fivefold higher in households than on farms. MDR strains were higher amongst household strains than farm strains.


Introduction
Pseudomonas is considered one of the most universally pervasive bacterial genera in the world.It is found extensively in the external environment of humans and animals.The genus has a diverse habitat, with approximately 200 species and a complex phylogeny, because of its metabolic capacity and broad potential for adaptation to diverse conditions.Pseudomonas aeruginosa is a psychotropic foodborne pathogen with high metabolic adaptability and growth capabilities at very low temperatures, supporting its prominent prevalence in several parts of the food chain [1,2].P. aeruginosa is recognised as an infectious agent transmitted via food and water [3].It is a Gram-negative bacterium with a zoonotic nature that can cause diseases transmitted from animals to humans and vice versa [4].It is considered an important cause of acute and chronic diseases in humans [5] and animals, including birds or mammals, that can act as reservoirs of bacterial pathogens.Moreover, environmental contamination via P. aeruginosa is prevalent [6].
The pathogen has been recognised in clinical and subclinical mastitis cases in dairy animals.As an environmental pathogen, it widely exists in humid areas, such as muddy bedding, dung, contaminated water, wash hoses in milking parlours, spray nozzles, the inner side of bulk tanks, cow teats, and surfaces that may be associated with the contamination of raw milk [5,7].Thus, the identification of probable contamination sources and the implementation of hygienic measures during the milking process have become a requirement in the growing dairy industry [7].In most dairy herds, the occurrence of Pseudomonas mastitis is sporadic but may cause a serious herd problem occasionally.Pseudomonas is usually regarded as an opportunist; that is, it causes disease under a certain condition of debilitation or is secondary to other infectious agents [2].Numerous risk factors contribute to Pseudomonas infection in farm animals.These factors include the use of common or contaminated teat cannulas for intramammary antibiotic infusion, which is involved in the introduction and spread of Pseudomonas mastitis; biofilm formation in milking parlours and poor environmental hygiene, which may facilitate intramammary infection [5]; and access to common sources, such as stagnant water ponds, which may account for certain herd health issues, including Pseudomonas mastitis [2].
A number of virulence factors are responsible for the pathogenicity of P. aeruginosa, namely exotoxins, proteases, elastases, and phenazine pigments, several of which are under the control of a cell-density recognition mechanism called quorum sensing [6].The bacterium is naturally resistant to numerous antimicrobial agents as a result of its outer membrane-permeability barrier.Pseudomonas also maintains antimicrobial resistance (AMR) plasmids, R-factors, and resistance transfer fragments, and it can transfer these genes with a horizontal gene transfer, mostly via transduction and conjugation [2].The treatment of P. aeruginosa infections is of great concern, owing to the organism's capacity to resist numerous existing and regularly used antimicrobials.The excessive use of antibiotics during treatment also increases the development of multidrug-resistant (MDR) strains, which renders antibiotic therapy ineffective against this microorganism.MDR-P.aeruginosa produces antibiotic-resistant genes against β-lactam, fosfomycin, fluoroquinolones, phenicol, sulphonamides, and aminoglycoside antibiotics [8].Moreover, P. aeruginosa has a prominent ability to form biofilms, making it more resistant to antimicrobial action [9].
Complex universal public hazards emerge, owing to AMR, thereby necessitating the application of new antimicrobial agents for therapeutic purposes to combat pathogenic microorganisms.However, no regulation for antibiotic use is implemented in Egypt, and antimicrobials are still applied as growth promoters and feed additives in animal feedstuffs to treat and prevent zoonotic diseases [10].The present study aimed to detect the ecological distribution and prevalence of MDR-P.aeruginosa from dairy cattle, milk, the environment, and workers' hand swabs.

Study Design
A cross-sectional study was conducted in two different housing sectors for dairy cattle.Random sampling was adopted in the examined sectors.A structured questionnaire was designed and distributed to farm owners to gain general information about each examined animal production sector and to determine the hygiene level in these houses.This study primarily aimed to detect the ecological distribution and prevalence of MDR-P.aeruginosa from dairy cattle reared in either organised farms or households, milk, the environment, or workers' hand swabs.
This work was conducted in accordance with the Declaration of Mansoura University, Egypt.Approval was granted by the Ethics Committee of the Faculty of Veterinary Medicine, Mansoura, Egypt (Ref.code No.: R/144-2022).

Origins and Processing of Samples
A total of 440 samples were collected from two different housing sectors of dairy cattle, specifically well-organised farms or households, in Dakahlia Governorate, Egypt.Amongst the samples, 320 were from three dairy cattle farms, and 120 were from three dairy cattle households.The samples were randomly selected.Samples from the dairy farm comprised rectal swabs, udder skin swabs, milk, feed, water from drinking troughs, and water sources used to wash the udder and milking utensils and bedding materials.Human samples were obtained randomly from workers' hand swabs.The examined farms were selected on the basis of the owners' agreements.All the examined dairy cattle houses depended on underground water.

Animal Samples
Milk (10 mL) was collected from each.The quarter samples were pooled per cattle as one sample in a sterile tube after cleaning and disinfecting the udders in ethyl alcohol and after discarding the first stream of the foremilk.For the rectal and udder skin swabs, moistened swabs were gently rotated either inside the rectum or over the udder skin.Ten swabs per cattle were pooled as one sample in a sterile tube.

Environmental Samples
Bedding (100 g) was collected from five different locations in each cattle farm or household and placed as one sample onto a sterile plastic bag.For water samples, 1 L of water was filtered through 0.45 µm sterile nitrocellulose filters (Millipore, Billerica, MA, USA).The filters were then vortexed in tryptone soya broth, and bacteria were allowed to grow.The water samples were collected either from water troughs in front of cattle or from the water source used in cattle drinking, washing of the udder, and washing of milking utensils.Feed samples (weighing around 100 g per sample) were collected from feed troughs in front of cattle and placed into a sterile plastic bag.

Human Samples
The moistened swabs were gently rotated over the hands of workers.Ten swabs per worker were pooled as one sample onto a sterile tube.

Animal-House Description
Dairy farms I (accommodating 704 dairy cattle) and II (accommodating 398 dairy cattle) had a history of a sharp decrease in milk production with a moderate level of adoption of hygienic measures.Farm III (accommodating 545 healthy dairy cattle) had no history of milk production drops.Good hygienic measures were adopted either in the rest area, feed storage area, or milking parlour, and the milking order was implemented properly.All dairy cattle in examined households (accommodating a few numbers of dairy cattle) had a history of subclinical mastitis.No hygienic measures were adopted for the house nor the udder.Upon asking the owners about antibiotics that were used on animals with health issues, the most frequently used antibiotics in farms and households were sulphonamides, trimethoprims, and macrolides.

Bacteriological and Chemical Identification of P. aeruginosa Strains
Dairy cattle samples, their environment, and human hand swabs were obtained aseptically and inoculated in tryptone soya broth (Oxoid, UK) before agar plating, followed by a Pseudomonas agar base (Oxoid, Basingstoke, UK) with a cetrimide nalidixic acid supplement, in accordance with the method described by Ibrahim [11].Then, the plates were aerobically incubated at 37 • C and examined after 24-48 h.Suspected colonies (3)(4)(5) characterised by pigmented colonies (either brown or green) with a musty smell were collected and subjected to further biochemical identifications.Isolation and chemical identification were conducted in the Hygiene and Zoonoses Laboratory, Faculty of Veterinary Medicine, Mansoura University, Egypt.Only one biochemically identified isolate from each pooled sample was subjected to further molecular identification and resistance checking for easy calculation of P. aeruginosa prevalence isolated from different samples.Finally, the suspected isolates were sent to the Animal Health Research Institute (Dokki, Cairo, Egypt) for further molecular characterization.

Molecular Identification of P. aeruginosa via 16S rRNA Gene Detection
DNA was extracted from all biochemically identified P. aeruginosa isolates (n = 102) incubated overnight in TSB broth using a QIAamp DNA Mini kit (Qiagen, Hilden, Germany, GmbH) in accordance with the manufacturer's recommendations.In a typical procedure, a 200 µL sample suspension was incubated with 10 µL of proteinase K and 200 µL of lysis buffer at 56 • C for 10 min.After incubation, 200 µL of 100% ethanol was added to the lysate.The sample was then washed and centrifuged following the manufacturer's recommendations.Nucleic acid was eluted with the 100 µL elution buffer provided in the kit.The primers were supplied by Metabion (Planegg, Germany) and used under certain conditions (Table 1).The primers were utilised in a 25 µL reaction containing 12.5 µL of Emerald Amp Max polymerase chain reaction (PCR) Master Mix (Takara, Osaka, Japan), 1 µL of each primer at a 20 pmol concentration, 5.5 µL of water, and 5 µL of DNA template.The reaction was performed in an Applied Biosystems 2720 Thermal Cycler (Thermo Fisher Scientific, Waltham, MA, USA).The PCR products were separated via electrophoresis on 1% agarose gel (Applichem, Darmstadt, Germany, GmbH) in a 1× TBE buffer at room temperature with 5 V/cm gradients.For gel analysis, 40 µL of products was loaded into each gel slot.A gene ruler with a 100 bp ladder (Fermentas, Thermo, Kandel, Germany) was used to determine the fragment sizes.The gel was photographed with a gel-documentation system (Alpha Innotech, Biometra, San Leandro, CA, USA), and the data were analysed using computer software.2.6.Phenotypic AMR of Farms and Households' Strains of P. aeruginosa Antimicrobial susceptibility tests (ASTs) were performed using the agar disc diffusion method on Mueller-Hinton agar (Difco, Franklin Lakes, NJ, USA) as recommended by the Clinical and Laboratory Standards Institute (CLSI).Frequently applied antibiotics for humans and animals were selected to be tested against our isolated strains of P. aeruginosa.Eleven antimicrobial discs (Oxoid, Basingstoke, Hampshire, UK) related to different classes of antibiotics were used.They were the macrolide erythromycin (E, 15 µg); the aminoglycosides gentamicin (G, 30 µg) and amikacin (AK, 30 µg); the folate pathway inhibitor trimethoprim-sulphamethoxazole (SXT, 25 µg); the fluoroquinolones norfloxacin (NOR, 10 µg), ciprofloxacin (CP, 5 µg), and levofloxacin (LEV, 5 µg); the carbapenem imipenem (IPM, 10 µg); the β-lactam amoxicillin (AX, 10 µg); and cefepime (CPM, 30 µg), and the penicillin combined with β-lactamase inhibitors piperacillin-tazobactam (TZP, 110 µg).The examined strains were assessed as susceptible or resistant in accordance with the CLSI guidelines for P. aeruginosa ATCC ® a 27853 [16,17].ASTs were performed in triplicate.To ensure data compatibility, we repeated the experiment with positive and negative controls.The positive control (quality-control organism) was P. aeruginosa ATCC ® a 27853.The negative control was 30 µL of sterile distilled water pipetted onto a blank disc (diameter = 6 mm).AMR data were accessible only when the quality-control test findings were within acceptable ranges.The multiple AMR (MAR) index was calculated by dividing the total number of AMRs for each isolate by the total number of tested antimicrobial agents [10,18].MDR-P.aeruginosa was defined as P. aeruginosa not susceptible to at least one antibiotic in at least three antibiotic classes for which P. aeruginosa susceptibility was generally expected, namely, penicillin, cephalosporins, fluoroquinolones, aminoglycosides, and carbapenems [19].

Molecular Identification of P. aeruginosa AMR-Resistance Genes (ARGs)
Given that the most frequently applied antibiotics in the examined houses were sulphonamides and macrolides, primers for drf A, sul1, and ermB were selected and supplied from Metabion (Germany) under certain conditions (Table 1).Uniplex PCR was conducted on each gene in accordance with previously described methods [13][14][15].

Statistical Analysis
The normality of the data was first tested with a one-sample Kolmogorov-Smirnov test.A Chi-square test was performed to analyse the data for comparing two or more groups of categorical variables using Statistical Analysis Software (SAS, Software version 9.4, SAS Institute, Cary, NC, USA).A comparison was conducted between the different farms examined.A comparison was conducted between the prevalence of bacteria in different animal houses, either from the animals themselves, their environment, or workers' hand swabs.The most prevalent source of P. aeruginosa contamination amongst different animal, environmental, and human samples was detected.The AMR and susceptibility between different sources were further detected at the level of the examined farms and households.The most prevalent antimicrobial genes were identified within the two examined housing sectors and between different sources of examined samples.A comparison was performed between the occurrence, phenotypic, and genotypic AMR of microorganisms amongst different sources, either animal, environmental, or human samples, as well as amongst different sectors of animal housing.The significance level was p < 0.05.

Prevalence of P. aeruginosa Isolated from Three Examined Dairy Farms and Households
The prevalence of P. aeruginosa was based on 16S rRNA detection.The prevalence of P. aeruginosa in households (57.5%) was fivefold higher than that in farms (10.6%).The prevalence of P. aeruginosa was higher in animal samples (26.4%) than environmental samples (25.4%) in all examined houses (Table 2).Human hand swabs showed a low prevalence rate (8.3%).A significant difference (p = 0.045) was observed in the prevalence of P. aeruginosa isolated from all the examined dairy farms (Table 3), in which the animal samples revealed a higher prevalence than environmental.Milk displayed the lowest prevalence amongst animal samples (10%), but rectal and udder skin swabs exhibited the highest values (20.8% and 15.6%, respectively).Meanwhile, the highest prevalence values amongst environmental samples were observed in bedding materials (10%), drinking water (9.7%), and water sources used for udder washing (8.3%).Workers' hand swab samples showed the lowest prevalence (2.2%).The prevalence of P. aeruginosa was higher in farm II (16.4%) than in farms I (13.3%) and III (5%).Regarding the examined dairy households, no significant difference was observed amongst the animal, environmental, and worker samples (Table 3).Dairy households showed a higher prevalence of P. aeruginosa in animal samples (62.2%) than the environment (61.7%) (p = 0.516).Rectal and udder skin swabs showed the highest percentage (66.7%each), followed by milk (53.3%).Feedstuffs, drinking water (73.3% each), and bedding and water sources (53.3% and 46.7%, respectively) showed higher recovery rates amongst environmental samples, with no significant difference between different sample sources (p = 0.516).The prevalence of P. aeruginosa was higher in household I (71.7%) than in households II (60%) and III (47.5%).

Discussion
P. aeruginosa had a wide-ranging prevalence in the two examined sectors (farms and households).Regarding the prevalence of P. aeruginosa recovered from examined dairy cattle farms, the results of research conducted in Malawi [20] coincided with ours.Our research detected Pseudomonas species at a rate of 10.2%, which was lower than the values previously reported by other authors (11.7% and 29.6%) [7,21].The detection rate of P. aeruginosa from rectal swabs of dairy cows in this work was lower than a previously reported one in Egypt [21], which detected the organism (34%) from faecal matter from dairy cattle farms.The recovery rate of P. aeruginosa from milk was higher than (5.4%) that of milk samples collected from bovine subclinical mastitis animals in Pengal [22] and lower than those recovered from milk and milk tank samples (70% and 24%, respectively) [7].Conversely, in another study conducted in Malawi, Pseudomonas species cannot be isolated from milk samples [20].This finding may be owing to the failed cleaning process of containers to meet standards; that is, more than 80% of the dairy farmers do not disinfect their milk-handling containers after cleaning [23].The contamination rate of water samples in our study was nearly similar to other research findings [24,25].Higher detection rates of water, feedstuffs, and environmental contamination by P. aeruginosa have been previously documented [7,20].In studies conducted in Egypt, the recovery rates of P. aeruginosa from workers' hand swabs (20% and 28%) are higher than those in the present study [7,21].The isolation of P. aeruginosa from rectal swabs from dairy cows indicated the dissemination of organisms between farm animals and their environment, as previously mentioned by Elshafiee et al. [24].P. aeruginosa was detected in water used for udder washing, consistent with the findings of Kirk and Bartlett [26].This result confirmed the persistent contamination of wash water, wash hoses, and spray nozzles in the parlour, which led to reinfection by P. aeruginosa causing clinical mastitis and chronic infections.Correspondingly, the organism was detected in udder skin swabs in the present study, possibly owing to insufficient udder hygiene before, during, and after the milking process.This result was supported by the work of Schauer et al. [5], who isolated the organism from the disinfectant solution and microfiber towels used for teat cleaning.The prevalence of P. aeruginosa was fivefold higher in households than farms, possibly owing to the lack of all hygienic measures in this sector of housing without any veterinary supervision.
A 5-year National Action Plan on AMR (2017-2022) was officially launched in numerous countries, including Egypt.Its strategic purposes are correlated with the consistent investigation of AMR and the optimisation of antimicrobial drug management in human medicine and animal health under the One Health concept released by WHO (2017).This concept aims to establish the interconnectivity of animal and human health with each other and their environments [10,18].AMR findings from dairy cattle samples in the present study agreed with previous ones that have detected 50% gentamycin resistance [25], high resistance to E, SXT [27], high resistance to AX [28], and the susceptibility of P. aeruginosa strains to CP [5].Conversely, CP resistance has been previously documented [25], as well as the high sensitivity towards G and IPM [27] and the full susceptibility of P. aeruginosa to IPM [28].Antibiotic resistance was higher in households' strains than in farm strains, probably owing to the massive use of antibiotics without veterinary supervision.The high resistance of P. aeruginosa strains to SXT was attributed to the frequent application of sulphonamides in the examined houses in the current work.The reason may be the extensive use of sulphonamides as synthetic veterinary antibiotics in numerous countries owing to their low costs [29].
The MAR index in this research ranged from 0.090 to 0.818 with a high MDR (81.4%), reflecting a great public-health hazard caused by difficulties in treating Pseudomonas infection in humans and animals.Our observation of "a MAR" of P. aeruginosa strains matched the observation of Mahmoud [25], who recorded "a MAR" index ranging from 0.44 to 0.77 and observed that six and one MDR strains originate from cows and their drinking water samples, respectively.Other researchers have recorded an elevated "a MAR" index from mastitic milk (ranging from 0.5 to 0.8) and 100% MDR in all recovered strains [28], with an MDR index of 0.8 for two strains.This multidrug resistance of recovered strains may result from the uncritical use of antibiotics in daily farm practice.The presence of AMR genes drf A, sul1, and ermB in animals and their environment can be attributed to the frequent usage of sulphonamides, trimethoprims, and macrolides in the examined farm animals.The drf A and sul1 genes in human samples collected from farm and household workers may be acquired from animals and their environment.The close contact between animal and human populations may be a high-risk factor for developing such a bacterial infection [24].Sulphonamide resistance encoded by sul1 has been previously documented (63.6%) in water samples from different sources and human ear swabs [30].Pseudomonas is one of the most prevalent sul-positive genera in soil fertilised using animal manure, suggesting a potential human-health risk [29].Studies on livestock or their products have recorded the susceptibility of P. aeruginosa isolates to SXT, which was attributed to mutations in mex gene determinants and in mutL and mutS [30].
Study limitation: multi-locus sequence typing is needed in future studies to provide more precise and valuable information on the identity amongst different sequences and to illustrate the genetic relatedness amongst the retrieved P. aeruginosa strains.

Conclusions
The household dairy sector (family farms) was contaminated by P. aeruginosa five times more than dairy farms.P. aeruginosa was found to have AMR against SXT, IPM, AX, G, and CPM.Meanwhile, P. aeruginosa was found to have antimicrobial susceptibility against amikacin, CP, and NOR in both examined sectors.The percentage of MDR strains was higher in household strains than in farm strains.All our findings indicated that the household sector, which is commonly present in Egyptian villages, represents a serious route of AMR dissemination.

Table 1 .
Target genes of P. aeruginosa with their primer sequences under certain conditions.

Table 2 .
Prevalence of P. aeruginosa in examined dairy farms (cattle, environment, and human).
* Significant difference (p < 0.05), p value: difference between sources within each farm.

Table 3 .
Prevalence of P. aeruginosa isolated from three examined dairy households (cattle, environment, and human).
* p value: difference between sources within each household.

Table 4 .
Distribution of AMR of P. aeruginosa farm strains (n = 33) based on AST results.

Table 5 .
Distribution of AMR of P. aeruginosa household strains (n = 69) based on AST results.

Table 6 .
Distribution of phenotypic and genotypic AMR profiles of P. aeruginosa farm strains (n = 33) to the tested antibiotics (n = 11).

Table 7 .
Distribution of phenotypic and genotypic AMR profiles of P. aeruginosa household strains (n = 69) to the tested antibiotics (n = 11).