Antimicrobial Resistance, Biofilm Formation, and Virulence Genes in Enterococcus Species from Small Backyard Chicken Flocks

Backyard birds are small flocks that are more common in developing countries. They are used for poultry meat and egg production. However, they are also implicated in the maintenance and transmission of several zoonotic diseases, including multidrug-resistant bacteria. Enterococci are one of the most common zoonotic bacteria. They colonize numerous body sites and cause a wide range of serious nosocomial infections in humans. Therefore, the objective of the present study was to investigate the diversity in Enterococcus spp. in healthy birds and to determine the occurrence of multidrug resistance (MDR), multi-locus sequence types, and virulence genes and biofilm formation. From March 2019 to December 2020, cloacal swabs were collected from 15 healthy backyard broiler flocks. A total of 90 enterococci strains were recovered and classified according to the 16S rRNA sequence into Enterococcus faecalis (50%); Enterococcus faecium (33.33%), Enterococcus hirae (13.33%), and Enterococcus avium (3.33%). The isolates exhibited high resistance to tetracycline (55.6%), erythromycin (31.1%), and ampicillin (30%). However, all of the isolates were susceptible to linezolid. Multidrug resistance (MDR) was identified in 30 (33.3%) isolates. The enterococci AMR-associated genes ermB, ermA, tetM, tetL, vanA, cat, and pbp5 were identified in 24 (26.6%), 11 (12.2%), 39 (43.3%), 34 (37.7%), 1 (1.1%), 4 (4.4%), and 23 (25.5%) isolates, respectively. Of the 90 enterococci, 21 (23.3%), 27 (30%), and 36 (40%) isolates showed the presence of cylA, gelE, and agg virulence-associated genes, respectively. Seventy-three (81.1%) isolates exhibited biofilm formation. A statistically significant correlation was obtained for biofilm formation versus the MAR index and MDR. Multi-locus sequence typing (MLST) identified eleven and eight different STs for E. faecalis and E. faecium, respectively. Seven different rep-family plasmid genes (rep1–2, rep3, rep5–6, rep9, and rep11) were detected in the MDR enterococci. Two-thirds (20/30; 66.6%) of the enterococci were positive for one or two rep-families. In conclusion, the results show that healthy backyard chickens could act as a reservoir for MDR and virulent Enterococcus spp. Thus, an effective antimicrobial stewardship program and further studies using a One Health approach are required to investigate the role of backyard chickens as vectors for AMR transmission to humans.


Introduction
Enterococci are Gram-positive bacteria that belong to the commensal microbiota of humans, animals, and poultry [1]. They are ubiquitous in nature and can be found in soils, freshwater, and plants [2]. Enterococcus spp. are important opportunistic human pathogens that are responsible for a wide range of serious nosocomial infections, including bacteremia, urinary tract infections, endocarditis, and intra-abdominal infections [3][4][5].
In veterinary medicine, enterococci are particularly significant as the causative agent of different infections, such as mastitis in cattle, bacteremia in dogs and pigs [6,7], and septicemia, endocarditis, amyloid arthropathy, and spondylitis in poultry [8,9].
Antibiotic resistance is an emerging world health threat. Many bacteria have developed resistance to frequently used antibiotics due to the unregulated use of antimicrobials in humans, agriculture, animals, poultry husbandry, and aquaculture in many developing countries [10,11].
One of the major concerns regarding opportunistic pathogens is their frequent antimicrobial-resistance (AMR) profile. Enterococci are intrinsically resistant to commonly used antibiotic classes, such as cephalosporins, β-lactams, sulfonamides, and are resistant at variable levels to aminoglycosides. Moreover, they are able to acquire resistance to clinically relevant drugs via horizontal transfer [12,13]. Enterococci are thought to play a key role in the acquisition, conservation, and transmission of AMR genes to other bacteria [14].
Enterococci pathogenesis is attributed to a variety of virulence factors. The most important adhesion factors that play a role in biofilm formation are Asa (aggregation substance), Esp (extracellular surface protein), EfaA (E. faecalis antigen A), Ace (adhesin of collagen from E. faecalis), and Ebp (endocarditis and biofilm-associated pili). Enterococci secrete the pathogenic factors CylA (cytolysin) and GelE (gelatinase), which are responsible for the exacerbation of infection. The expression of these factors is essential for biofilm formation, attachment, invasion, and the secretion of toxins [1,[15][16][17].
One of the major public health concerns related to enterococci is their frequent antimicrobial resistance (AMR). Human infection by AMR enterococci occurs mainly by consuming contaminated meat or meat products from poultry and other livestock [26][27][28], and contamination can occur during slaughtering and evisceration [29].
Enterococci have a high capacity to acquire antimicrobial resistance either by point mutation or by the horizontal transfer of genetic elements [30][31][32]. Conjugation is thought to be the most common way of exchanging genetic elements, either by conjugative transposons or by horizontal transfer of plasmids [33].
In general, enterococcal plasmids are classified into two groups: a conjugative group consisting of pheromone-responsive and non-pheromone-responsive plasmids and a nonconjugative group consisting of small rolling-circle replicating (RCR) and mosaic plasmids [34].
Inc18 plasmids and pMG1-type plasmids are classified as conjugative non-pheromoneresponsive plasmids, and frequently carry antibiotic resistance genes [34,35]. Nonconjugative RCR plasmids are typically small, with a high copy number, and a broad host range. Moreover, they frequently contain antibiotic resistance genes [36]. Jensen et al. [37] recently developed a classification scheme for enterococci based on the replication-initiation genes (rep) and distinguished 19 families and various unique rep genes.
Backyard birds are small flocks that are common in developing countries. They are also very popular in the USA, where birds are raised for meat production. Direct contact between poultry and humans is frequent; thus, these backyard flocks are considered a vehicle for disease transmission. Moreover, backyard poultry could be an emerging predisposing cause for MDR pathogenic bacteria, which can disseminate among humans [38]. In Saudi Arabia, many studies highlighted the prevalence of MDR enterococci in hospitals and communities [39,40]. The first vancomycin-resistant Enterococcus was detected in 1992 [41]; however, information concerning enterococci in backyard chicken is scarce.
Consequently, this study aimed to investigate the antimicrobial resistance, virulence determinants, and biofilm formation, and to characterize the plasmid content and multilocus sequence types in Enterococcus isolates from healthy chickens in backyard farms to highlight their zoonotic importance.

Antimicrobial-Resistance Genes
Nine antimicrobial-resistance genes were detected among the enterococci. Of the vancomycin-resistant isolates, 33.3% (1/3) contained vanA, while vanB was not detected in the three isolates. Twenty-four (86%) phenotypically erythromycin-resistant isolates were positive for erythromycin-resistant genes; the ermB gene was detected in all isolates, whereas the ermA gene was only detected in 11 isolates. Four tetracycline resistance genes (tetM, tetA, tetB, and tetL) were found in 94% (47/50) of tetracycline-resistant isolates. The prevalent resistance genes were tetM and tetL, accounting for 78% and 68% of the isolates, respectively; however, tetA and tetB were detected in 2% of the isolates. The cat gene was detected in 66% of the chloramphenicol-resistance isolates and the pbp5 gene for ampicillin resistance was detected in 22 enterococci isolates. Table 3 shows the distribution of antimicrobial-resistance genes among different enterococci. Table 3. Distribution of antimicrobial-resistance genes among the isolated enterococci.

Biofilm Formation
Overall, 73 Enterococcus isolates (81.1%) were biofilm producers, among which 39 were E. faecalis, 24 were E. faecium, 8 were E. hirae, and 2 were E. avium. The isolates were further classified into four categories based on the OD of the bacterial biofilm: 17 non-biofilm producers (18.9%), 16 weak biofilm producers (17.8%), 29 medium biofilm producers (32.2%), and 28 strong biofilm producers (31.1%). E. faecalis showed a significantly higher biofilm formation (p < 0.0001), and 13 E. faecalis isolates (28.9%) exhibited strong biofilm formation. Figure 2 shows the biofilm formation strength of the Enterococcus species. A statistically significant pairwise correlation (p < 0.001) was obtained for biofilm formation versus MAR  Table 4 shows the bacterial biofilm OD MAR index mean values in the four categories.

MLST of E. faecalis and E. faecium
MLST allelic profiles for E. faecalis and E. faecium are presented in Tables 5 and 6. A total of 11 STs were found among the E. faecalis isolates. The most prevalent STs were ST16 (n = 10) and ST302 (n = 8), followed by ST179 (n = 6), ST480 (n = 5), and ST752 (n = 3) ( Table 5). On the basis of the eBurst analysis and the phylogenetic analysis of the concatenated MLST sequences, seven STs were clustered into three groups: the first contained ST16, ST179, and ST302; the second ST81 and ST725; and the third ST176 and ST177; the remaining four STs (ST21, ST32, ST41, and ST480) were identified as singletons ( Figure 5).

Discussion
Backyard chickens are considered to be a vector for disseminating several zoonotic diseases, including Salmonella, Campylobacter, enteropathogenic E. coli, and several antibioticresistant microorganisms [42][43][44]. Moreover, enterococci, notably E. faecium and E. faecalis, have emerged as major multidrug-resistant zoonotic bacteria due to the widespread use of antibiotics in human and veterinary treatments [45].
In the current study, four Enterococcus species were isolated from healthy backyard chickens. E. faecalis was the most predominant species, which is consistent with the results of previous studies [43,46,47]. In contrast, E. faecium was reported as the prevalent species in poultry [48,49].
Antimicrobial resistance is one of the characteristics of enterococci, and their ability to acquire and spread antibiotic resistance presents a challenge for infection control [50].
In this study, a high proportion of resistance was identified in the isolated enterococci, the majority showing tetracycline resistance, which is accordance with various studies [43,[51][52][53]. Tetracycline-resistance genes (tetM and tetL) were detected in 78% and 68% of the isolates, respectively, while tetA and tetB genes were detected in 2% of the isolates. Tetracycline resistance is most often mediated by tetM and tetL in enterococci from humans, animals, food, and the environment [46,[54][55][56][57]. Different tetracycline-resistance genes were identified in the Enterococcus species [28,58]. Phenotypic resistance to erythromycin and ampicillin was observed in 31.1% and 30% of isolates. The ermB gene was detected in 86% of isolates, whereas the ermA gene was detected in 11 isolates, which is in accordance with Mlynarczyk et al. [59] who described erm(B) as the most prevalent gene conferring erythromycin resistance in enterococci. The pbp5 gene of ampicillin resistance was identified in 22 isolates, concordant with [43]. A high level of Enterococcus resistance to both tetracycline and erythromycin was reported in Switzerland [60], the Netherlands [61], France [62], and Portugal [63]. Macrolides, tetracyclines, and penicillins are the major antimicrobials used in integrated broiler companies [64], while in Saudi Arabia, tetracycline and erythromycin are the most frequently used antimicrobials in poultry farms [65]. Furthermore, tetracycline resistance has been described to co-select for erythromycin resistance [53]. The WHO classified macrolides as critically important (the highest priority) and tetracycline as highly important antimicrobials for human medicine [66].
Resistance to vancomycin has generated substantial research interest during the last decade since it is the drug of last resort to treat enterococci infections in humans [67]. In this work, we observed low vancomycin resistance (3.3%) among the isolated enterococci. However, the vanA gene was identified in all phenotypic-resistance isolates.
Our values are lower than those reported in other studies [68,69] and higher than the results of da Costa et al. [70]. However, Semedo-Lemsaddek et al. [43] did not detect any vancomycin resistance among their isolates. In the last decade, linezolid-resistant enterococci were detected in the USA, Europe, and Asia [50,71,72]. Remarkably, linezolidresistant enterococci were not detected in our study.
MDR was frequently detected among the isolated enterococci in this study. The emergence of MDR enterococci has also been reported worldwide, particularly in Korea (26.9%), Spain (87.5%), and Ethiopia (78.2%), and is currently regarded as a growing public health concern [72][73][74].
Biofilm formation plays a considerable role in enterococcal infections and antibiotic resistance [75,76]. In this study, biofilm formation was observed in 81.1% of the isolates, which is concordant with [75,76]. A positive pairwise correlation was observed between biofilm production and both the MAR index and MDR. Moreover, biofilm production has been linked to antibiotic resistance in enterococci [77]. In addition, the gelE gene was found in all biofilm-producing isolates but not in non-biofilm-producing isolates, suggesting its significance in biofilm development. gelE is necessary for biofilm formation because it stimulates cell aggregation in microcolonies, allowing them to construct a three-dimensional structure [78].
Despite gelE gene detection in 28% of isolates, in vitro gelatinase activity was only detected in 17% of isolates. Similarly, other investigations found that 30%, 56%, and 59% of clinical isolates generated gelatinase, while 90%, 88%, and 92% were gelE positive, respectively [79][80][81]. Together with our findings, these reports show that while gelE regulated by the fsr locus is necessary for gelatinase activity, it is insufficient since fsrA and fsrB are also required for the gelatinase phenotype.
Enterococcal cytolysin is a hemolytic virulence factor linked to human disease and increased patient mortality [82]. The cylA gene was detected in 21 isolates (23%), concordant with other studies in poultry [64,65]. However, only 57.1% of the isolates expressed hemolysin activity, which is in agreement with the results in [83][84][85]. These findings may be attributable to environmental factors such as the in vitro and in vivo conditions used to test for phenotypic characteristics, as these could have a significant impact on gene expression [86].
MLST represents an outstanding tool for global and long-term epidemiological studies. In this work, all 45 E. faecalis were divided into 11 STs. The most common STs in the backyard chickens (ST16, ST302, and ST179) have been previously found in poultry, wild birds, and pigs [63,87,88]. Furthermore, ST16, ST21, ST179, and ST480 were reported among E. faecalis hospital isolates in Saudi Arabia [89]. ST16 isolates were previously reported to display major diversity as to the source of isolation and lesions [22]. Among the ST16 E. faecalis isolates, two isolates were resistant to vancomycin. vanA E. faecalis ST116 isolates were previously isolated from turkey meat and non-hospitalized humans [67,90].
MLST genotyping of E. faecium isolates revealed eight different ST types, of which five belonged to CC9 and three belonged to CC17, suggesting an evolutionary link between backyard E. faecium isolates [91]. ST9, ST157, ST194, and CC9 in particular were previously isolated from poultry and poultry meat [92,93]. Although CC17 was reported as a nosocomial clonal complex [94], several studies reported the circulation of E. faecium CC17 in animals [67,93,95]. Backyard chickens possibly acquired the CC17 E. faecium isolates from contaminated environments, or humans visiting the farm. This suggestion is supported by a previous study that demonstrated the transmission of E. faecium of human origin to chickens [96]. Moreover, human-linked E. faecium has been isolated from various water and food sources [97,98].
Plasmids are believed to be plastic structures that change as a result of the everchanging environment in which they reside [33]. In this work, a recently published scheme for plasmid classification was utilized in order to investigate whether specific plasmid families were involved in AMR in enterococcus from backyard chickens. Ten enterococci did not carry any plasmid of the rep-families. This result is concordant with the studies of Jensen et al. and Cho et al. [37,99], in which approximately one-third of their isolates sets from humans, animals, and environments did not yield any amplicons. However, a lower percentage of negative rep-families was reported elsewhere [100,101], where 4% and 1.3% of E. faecium and E. faecalis did not yield any amplicons, respectively.
Seven different rep-family plasmid genes were identified in this study, a value concordant with several previous studies that detected five to nine rep-family plasmid genes [37,54,[102][103][104] and lower than the study of Cho et al. [99], who identified 12 repfamily plasmid genes.
A limitation of this study was the lack of environmental samples from the investigated flocks. However, in a previous study [108], a diversity of Enterococcus species was isolated from farm environments with multiple antibiotic-resistance profiles, indicating the role of chicken in environmental contamination. Furthermore, this was a small study that could not include additional samples of Enterococcus species taken from this region.

Study Area
The study was conducted in Al-Ahsa Governorate, in the eastern region of Saudi Arabia (25 •

Bacterial Isolation
Swabs were cultured on BD™ Enterococcosel™ Agar (Heidelberg, Germany) and incubated at 37 • C for 24 h. Colonies exhibiting a black hallo were selected and subcultured on 5% sheep blood agar (Oxoid, UK) for purification. Purified isolates were identified based on colony morphology, Gram staining, catalase, and oxidase tests. Further, specieslevel identification was conducted biochemically using GP identification cards and the automated Vitek 2 compact system (BioMérieux, France).

DNA Extraction and 16S rRNA Gene Amplification and Sequencing
Biochemically identified isolates were cultured in brain heart infusion broth (Oxoid, UK) at 37 • C for 48 h. Cells were harvested by centrifugation, and the bacterial DNA was extracted and purified using the QIAamp DNA mini-kit (Qiagen SA, Courtaboeuf, France) according to the manufacturer's instructions. The 16S rRNA gene was amplified and sequenced according to Weisburg et al. [109] and further analyzed using the National Center for Biological Information (NCBI) Basic Local Alignment Search Tool (https://blast. ncbi.nlm.nih.gov/Blast.cgi, accessed on 15 December 2021).

Detection of Antimicrobial-Resistance Genes
Antimicrobial-resistance genes associated with vancomycin (vanA, vanB), erythromycin (ermA, ermB), tetracycline (tetA, tetB, tetM, tetL), chloramphenicol (cat), linezolid (optrA), and ampicillin(pbp5) were determined by PCR [11,[114][115][116][117][118][119][120]. Primers and PCR conditions are presented in Supplementary Table S2. 4.6. Phenotypic Detection of Virulence Factors 4.6.1. Quantitative Biofilm Assay Antimicrobial Biofilm formation was assessed according to the methods described by Stepanović et al. [121]. Enterococci from an overnight culture were cultivated in trypticase soy broth (TSB) supplemented with 1% glucose and incubated for 24 h at 37 • C. The culture density was adjusted to an approximate 0.5 McFarland standard. Each culture was diluted in sterile TSB (1:100), and 200 µL from each was transferred to three wells of sterile 96 well polystyrene microtiter plates (Sigma-Aldrich, St. Louis, MO USA). A sterile TSB was used as a negative control, and E. faecalis (ATCC 29212) was used as a positive control. The plates were incubated at 37 • C for 48 h, washed with sterile phosphate-buffered solution, air-dried, and stained with 2% crystal violet for 30 min. Subsequently, the wells were gently washed with sterile deionized water and air-dried. The dye bound to the adherent cells was re-solubilized with absolute ethanol (150) µL. Each well's optical density (OD) was measured at 570 nm in a plate reader (BioTek-800 ST, St. Louis, MO USA). The experiment was performed in triplicate on three different days. Each Enterococcus isolate was classified as a negative, weak, moderate, or strong biofilm producer following the criteria described by Stepanović et al. [121].

Gelatinase Activity
Gelatinase activity was assessed by inoculating pure culture on agar plates containing 3% gelatin [122]. After 48 h incubation, plates were flooded with a saturated ammonium sulfate solution. A transparent halo zone surrounding the colonies was considered positive for gelatinase.

Cytolysin Activity
For screening hemolysin production, Enterococcus isolates were streaked on Columbia agar supplemented with 5% horse blood and incubated at 37 • C for 24 h. A clear (ß-hemolysis) or green (α hemolysis) zone around the colonies was defined as positive, whereas the γ-hemolysis was defined as negative activity [123].

Molecular Detection of Virulence Factor Genes
The virulence factor genes, including gelE (gelatinase), agg (aggregation substance), and cylA (activator of cytolysin), were screened by PCR, according to [124,125]. Primers and PCR conditions are tabulated in Supplementary Table S2.

Multi-Locus Sequence Typing
MLST for E. faecalis and E. faecium was performed by sequencing seven housekeeping genes described by Ruiz-Garbajosa et al. and Homan et al. [22,23]. Different sequences were assigned allele numbers, and different allelic profiles were assigned STs based on the MLST database (http://www.mlst.net/databases/, accessed: 15 December 2021).

PCR for repA Genes (Plasmid Families)
All MDR isolates were screened for rep-like sequences by PCR according to Jensen et al. [37] with primers and PCR conditions listed in Supplementary Table S2.

Conclusions
This study investigated virulence genes, antibiotic resistance, multi-locus sequence types, plasmid-associated genes, and the biofilm production of enterococci from healthy backyard chickens in Saudi Arabia to highlight the role of backyard chickens as a potential reservoir for MDR and virulent enterococci. Molecular analyses revealed the presence of nosocomial-associated CC17 and a variety of mobile genetic elements among the enterococci from backyard chickens, suggesting the possibility of their dissemination in backyard farms and their environments.
High resistance to different antimicrobial classes was identified, suggesting the overuse of antimicrobials in backyard chicken farms. The emergence of MDR and virulent enterococci in backyard chicken farms is a public health concern. Thus, regular surveillance for the occurrence of MDR and virulent enterococci in backyard poultry and its environment is recommended to prevent its spread and to minimize environmental contamination.
Furthermore, proactive antimicrobial agent control measures should be developed to limit the spread of MDR enterococci.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/ 10.3390/antibiotics11030380/s1, Figure S1: Description and categories of the collected variables. Table S1: Description and categories of the collected variables.  Institutional Review Board Statement: The Taif University Ethics Committee has approved the study protocol (TURSP-2020-262).

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