Molecular Evaluation of Traditional Chicken Farm-Associated Bioaerosols for Methicillin-Resistant Staphylococcus aureus Shedding

The outbreak of airborne pathogens, such as methicillin-resistant Staphylococcus aureus (MRSA) through bioaerosol, and their molecular characterization around domestic poultry farming areas, was not completely understood. This imposes risk of a MRSA-associated health threat for the relevant livestock food production units. To address this issue, the present study investigated the role of bioaerosol in transmitting MRSA strains in poultry house settings by combining molecular typing, phylogenetic classification, antibiotic susceptibility, and virulence gene distribution patterns. The present study highlights that all 18 bioaerosol and stool samples collected were MRSA positive, with a unique set of virulence factors. Out of 57 isolated MRSA isolates, 68.4% and 19.3% consisted of SCCmec I and IV elements, respectively, which are commonly linked with hospital-acquired and livestock-associated MRSA strains. It is worth noting that the exfoliative toxin eta and etb genes were carried by 100% and 70.2% of all isolates, respectively. Only 17.5% of strains showed the presence of enterotoxin entC. These MRSA isolates were resistant to chloramphenicol (C), ciprofloxacin (CIP), clindamycin (DA), erythromycin (E), and tetracycline (T), signifying their multi-drug resistance traits. A cluster of phylogenetic analysis described that 80.7% and 15.8% of total isolates belonged to Staphylococcus aureus protein A (spa) type t002 and t548. Whereas 3.5% were reflected as a new spa type. Additionally, as per the chi-squared test score value, these two spa types (t002 and t548) have a distribution correlation with HA-MRSA and LA-MRSA in all the samples (p < 0.005, chi-squared test; degree of freedom = 1). Ultimately, this study highlights the prevalence of MRSA colonization in the conventional poultry farm environment, showing the risk of bioaerosol transmission, which needs epidemiological attention and prevention strategies.


Introduction
Methicillin-resistant Staphylococcus aureus (MRSA) is a Gram-positive pathogen that could proliferate to a virulent level in the air, and the utmost opportunistic ESKAPE epidemiology of MRSA strains in the bioaerosol samples around poultry farm area in South-Taiwan. We used SCCmec and spa typing including MDR pattern, and PCR-based virulence genes identification tools to identify unique molecular distribution as well as prevalence of HA-type in the poultry settings of Taiwan.

Odorous Compounds and MRSA Prevalence in the Environment
The anemometer data showed that the frequent direction of wind in pre-and postwinter seasons was from the southeast at the sampling sites (Table 1). Whereas in winter, the direction was from the northwest. The wind speed range around the sampling period was 0.4-2.2 m/s. The concentration range of ammonia and methylamine gas in the chicken shed 1 and 2 (indoor environment) was 2-7 ppm and 5-7 ppm, respectively, and outer exposure area had low concentrations (Supplementary Table S1). The concentration of hydrogen sulfide and mercaptan was below the detection limit at both sites. Furthermore, the total airborne bacteria load range in the chicken shed 1 and 2 indoor areas differed from 1.53 × 10 3 to 2.63 × 10 3 CFU/m 3 and 1.63 × 10 3 to 2.65 × 10 3 CFU/m 3 , while at the exposure environment, the range was found as 3.04 × 10 2 to 7.67 × 10 2 CFU/m 3 . The positive results of MRSA detection revealed that all collected bioaerosol and fecal samples from the chicken farm area were severely contaminated throughout different time point.

Distribution of SCCmec Elements and MRSA Grouping
SCCmec typing was performed on 57 MRSA isolates obtained from sampling (Table 2). Among them, 68.4% of isolates possessed SCCmec I, designated as the most dominant SCCmec element. Out of the total, 19.3% and 12.3% of isolates contained SCCmec IV and SCCmec VIII. A total of 39 isolates were linked with HA-MRSA, whereas 11 and 7 strains belonged to LA-MRSA and other MRSA groups, respectively. Out of the 16 isolates from shed 1, 10 were HA-MRSA, carrying only SCCmec elements I. Only one belonged to LA-MRSA having SCCmec IV, and five strains were from other MRSA classes. Chicken shed 2 possessed a total of 12 isolates that were linked with SCCmec I containing HA-MRSA strains, and four isolates had SCCmec elements IV, considered LA-MRSA. Exposure plaza environment samples led to a total of 11 strains having SCCmec I, which were connected with HA-MRSA; four isolates had SCCmec IV of LA-MRSA. For stool samples, a total of 10 MRSA isolates were obtained, out of which six isolated were SCCmec I-linked HA-MRSA, 2-SCCmec IV of LA-MRSA, and 2-SCCmec VII, as other MRSA strains.

Antimicrobial Susceptibility and MRD Pattern of MRSA Isolates
Antimicrobial susceptibility was assessed using the disk diffusion method. A total of eight antimicrobial agents were checked for all of the isolates. The results are shown in Table 4. All isolated strains were resistant to chloramphenicol (C), ciprofloxacin (CIP), clindamycin (DA), erythromycin (E), and tetracycline (T) (100%). Only 33.3% (19/57) and 12.3% (7/57) were resistant to sulfamethoxazole-trimethoprim (S/T) and rifampicin (RA, 12.3%), respectively. As per Magiorakos et al. [32], multidrug-resistant bacteria (MDRB) is defined as being non-susceptible to ≥ 1 agent in ≥ 3 antimicrobial categories; therefore, all 57 MRSA isolates were MDR strains. As shown in Table 5, there are 37 isolates had five antimicrobial drugs resistance pattern denoted as C-CIP-DA-E-T. Moreover, 13 isolates with C-CIP-DA-E-T-S/T and one isolate with a C-CIP-DA-E-RA-T resistance profile. Critically, six strains showed seven antimicrobial drugs resistance denoted as C-CIP-DA-E-RA-T-S/T pattern.

Spa Typing and Phylogenetic Clustering of MRSA Isolates
Spa typing was performed on 57 MRSA isolates; the results are shown in Table 6. Out of 57 MRSA isolates, 46 MRSA strains were nominated as Spa type t002 (80.7%), nine (15.8%) isolates belonged to Spa type t548 strains, and two strains belonged to Spa new type (3.5%). Figure 1 shows a comprehensive phylogenetic analysis of 57 isolated MRSA strains based on Spa typing, sampling site and types, SCCmec typing, antibiotic resistance, and toxin genes profiles. Furthermore, the phylogenetic classification of all isolates was divided into three main sub-clusters: t002, t548, and new type. The Spa t002 type group is typically related to HA-MRSA (SCCmec I and VIII), while the Spa t548 type group is primarily related to LA-MRSA (SCCmec IV). According to the chi-squared test of independence, t002, HA-MRSA vs. t548, LA-MRSA has a distribution correlation (p < 0.005, chi-squared test; degree of freedom = 1).

Discussion
All nine bioaerosol and nine chicken stool samples were collected from three sampling points (chicken 1, 2, and exposure plaza). The concentration of odor-forming gases (such as ammonia and methylamine) were low in outdoor (exposure plaza) air compared to the chicken sheds' indoor air. Similarly, the total bacteria count in exposure plaza open air was lower than both chicken sheds' indoor air counts. The previous study detected ammonia (16.8-66.7 mg/m 3 ) and methylamine (up to 0.82 mg/m 3 ) concentration in the air samples, and 3.2 × 10 9 CFU/g of total bacteria load in settled dust samples from poultry farm environment [32]. In the present study, the ammonia and methylamine concentrations were in the range of 7-2 ppm and 7-2.5 ppm, respectively. While airborne bacteria were in range of 2.65 × 10 3 -3.04 × 10 2 CFU/m 3 , which is comparable to other studies from Asia and Europe [33,34]. Ammonia deteriorates the air quality of animals and poultry farms, where it was detected in the range of 0.7-20 ppm [35]. High ammonia concertation could negatively impact livestock production by increasing the severity of disease [36][37][38][39]. Out of the total bacterial load in the farming air, Liu et al. showed that 5.37% of the total 149 S. aureus isolates from indoor and outdoor bioaerosol samples were identified to be MRSA [29]. Nasal swab sampling in poultry found that 56.8% of flocks were positive for S. aureus, among them 30% were harboring MRSA strains [31]. Likewise, Zhong et al. showed that S. aureus can exist in chicken feces and indoor air samples of chicken farm areas at a significant level, and MRSA isolates were also detected in all air and stool samples, which corroborates with our observations [38].
For instance, throughout Europe and North America, the CC398 is a dominant clonal complex of a MRSA strain carrying predominantly SCCmec element types IV, V, and occasionally NT [40,41]. Especially in poultry farms, CC9 is regularly detected in the LA-MRSA strain, which is found in pig and poultry houses of Asian countries [40,42]. The sequencing data of strain CC9 indicated that highly versatile types of SCCmec elements (III, IV, V, including novel and NT) could exist in their whole genome compared to CC398 [40]. Such genetic diversity of MRSA strains obtained from poultry in Taiwan, which is also uniquely linked to bioaerosol transmission, is studied for the first time in this investigation. SCCmec typing results of this study showed a co-occurrence between SCCmec type I and IV with HA-MRSA and LA-MRSA. The typical genotype classification of S. aureus, such as SCCmec I, is linked with HA-MRSA and SCCmec IV elements linked to CA-MRSA [43]. However, SCCmec type IV and V carrying LA-MRSA lineage are more dominant in the East Asian livestock units [23,40,44]. Recent studies also showed a high prevalence of SCCmec IV in the MRSA isolates from poultry and livestock foods, which supports the present study [44,45]. Our previous report about the Chiayi's river basin and nearby livestock areas revealed that SCCmec IV and I were distributed as 64.1% and 15.4%, respectively, wherein LA and HA-MRSA were predominant isolates from water samples, whereas CA and HA-MRSA were predominant isolates from hospital and long-term care facilities environments [46,47]. In the present case, SCCmec type I (HA-MRSA) was present in the bioaerosol and stool samples, which could be a future threat to the human population. Likewise, the presence of SCCmec IV containing LA-MRSA could colonize livestock animals. Poultry-associated MRSA strain predominantly belonged to CC398 spa types, other types of clones were also detected in the diverse geographic region [11]. Here, the spa typing and its cluster analysis highlighted that t002 and t548 type strains were linked to SCCmec type I and IV elements, respectively. Such distribution of spa types was reported from swine farm, indicating t002 and t548 spa-type MRSA strains reported from 106 nasal, swabs, and environmental samples [48].
Virulence factors, such as exfoliative toxins (ETs), are epidermolytic in nature. They are serine proteases secreted by S. aureus [49]. Around 10% of the MRSA strains possess eta; however, a report by Marek et al. showed a limited presence [50,51]. ETs could contribute to the exfoliative epidermitis in pig and Staphylococcal scaled-skin syndrome in humans; however, ETs A, B, and D, originating the human S. aureus strain when inoculated in chickens, showed limited or no exfoliative activity [52][53][54]. Our detection results exhibited that 100% of total isolates in this study carried exfoliative toxin eta and 70.2% possess etb. On contrary, for chicken and duck fecal swab and livestock foods, isolated strains were found negative for exfoliative toxin genes [55,56]. Exfoliative toxin genes were reported in mastitis-infected cow milk samples from Bangladesh [57]. Although a limited risk to poultry, since eta and etb were linked to human pathogenic MRSA strains, this study highlights a potential occupational and community health challenge [49,58,59]. Therefore, bioaerosol could play a significant role in the transmission of MRSA from livestock farms to adjacent community settings. A total of 17.5% isolates possess entC that might increase the risk of Staphylococcal scalded skin or toxic shock syndrome or food poisoning issues [60].
It is noteworthy that all MRSA isolates were MRD. They can typically resist chloramphenicol, ciprofloxacin, clindamycin, erythromycin, and tetracycline. Lu et al. showed that the MRSA isolates from the poultry farm environment can also be multi-drug resistant, which supports present study results [24]. A bioaerosol from livestock farms possess tetracycline and erythromycin resistance S. aureus [30]. Taiwan has amended the veterinary drugs control act that restricts the use of multiple antibiotics in food-producing animals. Most of the antibiotics tested in this study are used for human treatment or only for disease control in livestock; therefore, any resistance to these specific antibiotics can be considered a critical epidemiological risk indicator [61]. Strikingly, all 57 MRSA isolates of this study showed multidrug-resistance ability and their virulent profiling suggest that MRSA colonization in the poultry farms have substantial potential to pose a health risk for both human and nearby livestock via bioaerosol transmission.

Study Area and Sampling Information
The geographical coordinates of the sampling area were 23 • 35 11.7 N 120 • 29 27.3 E (Figure 2). Here, two chicken sheds, and in between one open exposure plaza, were targeted for environmental bioaerosol collection via a BioStage air sampler. Three bioaerosol and three chicken stool samples were collected at three different times from each targeted sampling point (between June 2019 and March 2020). A total of 18 environmental and fecal samples were collected for further analysis.

Sampling Procedure and Environmental Parameters Analysis
For bioaerosol sampling, the BioStage single-stage cascade impactor (SKC Inc., Blandford Forum Dorset, UK) was placed on a 1.2 m high platform to collect air that could simulate the height of average human breathing. The air sampling flow rate was 28.3 L/min and the sampling time was 10 min. A total 283 L volume of air was used at a time to collect bioaerosol samples. Tryptic soy agar (TSA) with 100 mg/mL cycloheximide was inserted into a BioStage sampler to detect the total number of bacteria count in the environment per volume of air sample (meter cube), and a selective CHROMagar™ MRSA (Paris, France) was also placed in the sampler to screen and isolate MRSA from these environments [62]. Simultaneously, for fecal samples, the stool was collected into sterile specimen bottles. The gas detector tube system (Gastec Inc., Fukayanaka, Japan) was used as per the standard operating procedure to analyze the concentration of odor-producing gases, such as ammonia, methylamine, hydrogen sulfide, and mercaptan (https://www.gastec.co.jp/en/instructionmanual/, accessed on 1 June 2019). Wind speed and direction were also measured using an anemoscope (Puxicoo P6-8232, Shenzhen, China).

Isolation and Selective Cultivation of MRSA Isolates
A total of 1 g of stool sample was taken and added into 9 mL of trypticase soy broth (TSB) with 6.5% NaCl. It was incubated at 37 • C for 16 h. Using a loopful of the TSB enrichment medium, MRSA colonies were isolated on CHROMagar™ MRSA plates (incubated at 37 • C for 24 h). Simultaneously, CHROMagar™ MRSA plates placed in the BioStage sampler system were incubated at 37 • C for 24 h. A single pure mauve color colony from CHROMagar™ MRSA was picked for further analysis by transferring it to Brain-Heart Infusion Broth (BHIB) and incubating at 37°C for 24 h. Subsequently, for confirmation, they were transferred to a moderately selective and differential medium of Baird-Parker Agar and incubated at 37 • C for 24 h. Finally, these pure isolates were grown in BHIB and preserved in 33% glycerol at −20°C for further use. Staphylococcus aureus (ATCC25923, ATCC29213) was used as a positive control in this study.

DNA Extraction and PCR Detection of MRSA Isolates
For DNA extraction, first, the freshly grown bacterial cells in the BHIB medium were harvested by centrifugation at 10,000× g for 5 min. DNA was extracted by a commercial kit (MagPurix Bacterial DNA Extraction Kit, ZP02006, Taipei, Taiwan) as per their standard protocol. For PCR experiments, the primers and master mix (Fast-Run TM Taq Master Mix with Dye, Springwood, Australia) were mixed with genomic DNA (100-300 µg). The total reaction volume for PCR was 25 µL, and the PCR program conditions of specific detecting genes are described in Supplementary Table S2. Nuc and mecA genes were utilized to confirmation of MRSA strains. Various molecular elements, such as mec, virulencegenes including SCCmec, Panton-Valentine leukocidin (PVL), enterotoxins (entA~E), toxic shock syndrome toxin-1 (tsst-1), and exfoliative toxin (eta and etb) were confirmed by PCR amplification. The PCR-based Staphylococcus aureus Protein A typing (Spa typing) data were analyzed by commercial software (BioNumerics, Sint-Martens-Latem, Belgium) for the phylogenetic analysis. the amplicons were confirmed by electrophoresis using 1.5% agarose gel at 110 V for 30 min.

Antibiotic Susceptibility Tests
The disk diffusion method was used for the antibiotic sensitivity test according to the standard protocol by the Clinical and Laboratory Standards Institute (CLSI) [63].

Statistical Analysis
The chi-squared test was performed to prove the statistical significance of the distributional relationship between different isolated MRSA strain spa types with their sampling site, sampling type, SCCmec typing, MRD pattern, and existing toxin gene profile groups.

Conclusions
This pilot study was carried out with different time-specific sampling at two traditional chicken farm sheds and an exposure plaza in South Taiwan, highlighting that all of the bioaerosols collected contained MRSA strains, wherein the total airborne bacterial load was comparatively higher inside the chicken shed environment. Isolated MRSA strains were able to resist multiple antibiotics, such as chloramphenicol, ciprofloxacin (CIP), clindamycin (DA), erythromycin (E), and tetracycline (T), underpinning their MDR characteristics. This study found a high prevalence of exfoliative toxin genes, eta and etb, in the MRSA isolated strains. The SCCmec element profiling showed the predominant occurrence of SCCmec I-associated among 68.4% HA-MRSA strains, while only SCCmec IV elements were prevalent in 19.3% LA-MRSA isolates. Phytogenic classification by spa typing revealed that 46 and 9 MRSA isolates were Spa type t002 and t548 strains, respectively. Two isolates were categorized as the new spa-type. Additionally, t002 and t548 spa types were positively correlated with hospital and livestock allied MRSA infections. Ultimately, multidrug resistant HA-MRSA and LA-MRSA dominance in these samples have enough potential to impose epidemiological risk via bioaerosol transmission through unhygienic poultry practices.

Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.

Data Availability Statement:
The data presented in this study are available upon request from the corresponding author.

Conflicts of Interest:
The authors declare no conflict of interest.