Determination of the Prevalence and Antimicrobial Resistance of Enterococcus faecalis and Enterococcus faecium Associated with Poultry in Four Districts in Zambia

The presence of antimicrobial-resistant Enterococci in poultry is a growing public health concern worldwide due to its potential for transmission to humans. The aim of this study was to determine the prevalence and patterns of antimicrobial resistance and to detect drug-resistant genes in Enterococcus faecalis and E. faecium in poultry from four districts in Zambia. Identification of Enterococci was conducted using phenotypic methods. Antimicrobial resistance was determined using the disc diffusion method and antimicrobial resistance genes were detected using polymerase chain reaction and gene-specific primers. The overall prevalence of Enterococci was 31.1% (153/492, 95% CI: 27.1–35.4). Enterococcus faecalis had a significantly higher prevalence at 37.9% (58/153, 95% CI: 30.3–46.1) compared with E. faecium, which had a prevalence of 10.5% (16/153, 95% CI: 6.3–16.7). Most of the E. faecalis and E. faecium isolates were resistant to tetracycline (66/74, 89.2%) and ampicillin and erythromycin (51/74, 68.9%). The majority of isolates were susceptible to vancomycin (72/74, 97.3%). The results show that poultry are a potential source of multidrug-resistant E. faecalis and E. faecium strains, which can be transmitted to humans. Resistance genes in the Enterococcus species can also be transmitted to pathogenic bacteria if they colonize the same poultry, thus threatening the safety of poultry production, leading to significant public health concerns.


Introduction
Enterococcus is a genus of Gram-positive bacteria in the family Enterococcaceae, the order Lactobacillales and the phylum Firmicutes [1]. Enterococcus is part of the normal flora in the gastrointestinal tract (GIT) of mammals, fish, reptiles, insects, and birds [2,3]. Being ubiquitous in nature, it is also found in soil, plants, sewage and fresh and salt water [4,5]. Species in the genus Enterococcus (E) have emerged as pathogens of medical and public health importance [6]. This is partly due to their adaptability to the selective pressures of antimicrobials. They also have the ability to acquire, express, and transmit mobile genetic elements (MGEs) from/to pathogenic as well as non-pathogenic species in the same or PCR was run on 343 suspect Enterococcus DNA samples extracted from poultry droppings using genus-specific primers for elongation factor (tuf ) and d-alanine-d-alanine ligase (ddl) genes. PCR was subsequently run on 153 positive DNA samples using species-specific primers for Enterococcus faecalis and Enterococcus faecium. The most common Enterococcus species was E. faecalis (37.9%), followed by E. faecium (10.5%). Remarkably, 38.6% of the isolates contained more than one species, with 34.6% of the total enterococcus isolates containing both E. faecalis and E. faecium. Adding the latter to E. faecalis and E. faecium, E. faecalis would still be the most predominant species, followed by E. faecium (Figure 1). The word "Other" represents Enterococcus species-identified by PCR using genus-specific ddl and tuf gene primers-which could not be identified through PCR due to lack of additional species-specific primers, or DNA sequencing due to the unavailability of reagents. Figure 1 shows the species identified using E. faecalis and E. faecium species-specific primers.
for performing API tests on only 37 isolates was due to insufficient reagents. Particularly, the NIN, VP 1 + VP 2, ZYM A and ZYM B were enough for only 38 samples (one control Enterococcus faecalis ATCC 29212 strain and the 37 isolates).

Identification of Enterococci Using Polymerase Chain Reaction (PCR)
PCR was run on 343 suspect Enterococcus DNA samples extracted from poultry droppings using genus-specific primers for elongation factor (tuf) and d-alanine-d-alanine ligase (ddl) genes. PCR was subsequently run on 153 positive DNA samples using speciesspecific primers for Enterococcus faecalis and Enterococcus faecium. The most common Enterococcus species was E. faecalis (37.9%), followed by E. faecium (10.5%). Remarkably, 38.6% of the isolates contained more than one species, with 34.6% of the total enterococcus isolates containing both E. faecalis and E. faecium. Adding the latter to E. faecalis and E. faecium, E. faecalis would still be the most predominant species, followed by E. faecium (Fig  1). The word "Other" represents Enterococcus species-identified by PCR using genusspecific ddl and tuf gene primers-which could not be identified through PCR due to lack of additional species-specific primers, or DNA sequencing due to the unavailability of reagents. Figure 1 shows the species identified using E. faecalis and E. faecium species-specific primers.

Comparing API and PCR Identification
API and PCR results were compared to ascertain the agreement between the two methods. API correctly identified 17 (45.9%) of the 37 isolates. API could not identify isolates with more than one species and only picked one of the species in samples with two or more species (16, 43.2%). It also misidentified an isolate that contained E. faecalis and another species as E. faecium, and it was not able to identify two isolates. Additionally, API identified one isolate as E. durans1, but this could not be confirmed as the corresponding species-specific primers were not available (Table 1).

Comparing API and PCR Identification
API and PCR results were compared to ascertain the agreement between the two methods. API correctly identified 17 (45.9%) of the 37 isolates. API could not identify isolates with more than one species and only picked one of the species in samples with two or more species (16, 43.2%). It also misidentified an isolate that contained E. faecalis and another species as E. faecium, and it was not able to identify two isolates. Additionally, API identified one isolate as E. durans1, but this could not be confirmed as the corresponding species-specific primers were not available (Table 1).

Antimicrobial Susceptibility of Enterococci
All intermediate test results were considered resistant. Both Enterococcus species showed very high (97.3%) resistance to tetracycline, while 94.6% were resistant to erythromycin and 77.0% were resistant to ciprofloxacin. Remarkably, 64.9% of both Enterococcus species were susceptible to vancomycin. More than 90.0% of E. faecalis isolates were resistant to erythromycin and tetracycline and more than 50.0% were resistant to ampicillin, chloramphenicol and ciprofloxacin. Less than 20.0% of the E. faecalis isolates were resistant to vancomycin. All E. faecium isolates in this study were resistant to erythromycin. More than 80.0% of E. faecium isolates exhibited phenotypic resistance to ampicillin, ciprofloxacin and tetracycline, while less than 40.0% showed resistance to chloramphenicol and vancomycin. Susceptibility profiles of E. faecalis and E. faecium to the eight antimicrobials tested are provided in Table 4. Multidrug resistance (MDR) is defined as resistance to three or more classes of antimicrobials. The results of our study show that none of the isolates were susceptible to all antimicrobial classes tested. Of the 74 E. faecalis and E. faecium isolates tested against eight antimicrobials, only two (2.7%) were resistant to one class of antimicrobials. A total of 5 (6.8%) isolates were resistant to two classes of antimicrobials. The majority of E. faecalis and E. faecium isolates (67, 90.5%) were MDR (Table 5). The aac(6 )-Ie-aph(2")-LA resistance gene encoding resistance to gentamycin was detected in 33 and 12 E. faecalis and E. faecium isolates, respectively, representing 60.8% of the isolates. The ermB resistance gene was more common in both E. faecalis and E. faecium compared with the ermA gene. The vanA resistance gene was detected in only two E. faecalis isolates and in none of the E. faecium isolates. Table 6 shows the number of different resistance genes detected in the E. faecalis and E. faecium isolates.

Association between Antimicrobials and Resistance Genes
Differences in antimicrobial resistance patterns and resistance genes in both enterococcus species were analyzed to assess possible associations between resistance phenotypes and their corresponding genotypes. A positive association between phenotype and genotype was found for tetracycline (p = 0.047) and erythromycin (p = 0.008), but there was no association between genotype and the vancomycin resistance phenotype (p = 0.051) ( Table 8).

Discussion
The prevalence, antimicrobial susceptibility patterns and presence of resistance genes in poultry droppings from four districts in Zambia were determined. The overall prevalence of Enterococci was 31.1%. This is in agreement with other studies that reported similar results in Poland [26], Malaysia [27] and Nigeria [28]. However it was lower than that reported in a similar study conducted in Zambia, where the prevalence was 88.4% in laying hens [29]. This could be due to differences in sampling methods, farms sampled and the number of farms sampled. Another previous study [30] also reported a higher prevalence than that reported in the present study. Conversely, the prevalence rate in our study was higher than the rates reported in Ethiopia [31], Pakistan [32] and Thailand [33]. The differences in the isolation rates of Enterococci can be attributed to several factors, including antibiotic use, environmental factors and methodology. The widespread use of antibiotics has led to the selection and dissemination of antibiotic-resistant Enterococci. Enterococci are found in soil and water and can persist in the environment for long periods of time, making them more difficult to control and leading to increased isolation rates. The isolation rate can also be influenced by the type of culture method used and the presence of selective media that may preferentially isolate Enterococci [34].
Among the Enterococcus species isolated in this study, E. faecalis was the most prevalent (37.9%), followed by E. faecium (10.5%). This was in agreement with other studies [35][36][37][38] which found E. faecalis to be the most prevalent species in poultry. However, our study differed slightly from some studies that found species other than E. faecalis to be the most predominant [16,39,40]. The variations in species levels between studies might be due to differences in the type of poultry, source of chicks, sampling methods, geographical disparities, the time of study and isolation and identification procedures [40].
Although API 20 strep is considered the best identification system for bacteria [41], it does not accurately identify some species of Enterococci [42]. In the present study, we validated API 20 strep results using PCR. PCR conducted using species-specific primers identified 91.9% of samples containing both single and multiple species. API 20 strep accurately identified 45.3% of Enterococcus species, but identified only one species in isolates containing more than one species. It also misidentified 2.7% of the Enterococcus species. Our findings were in agreement with the results of previous studies [42][43][44][45].
In the present study, phenotypic resistance to critically important antimicrobials, as defined by the WHO [46], was observed and 90.5% of E. faecalis and E. faecium isolates were multidrug resistant (MDR) ( Table 4). Notably, all E. faecalis and E. faecium isolates were resistant to one or more of the tested antimicrobials (Table 5). These findings were similar to those of a study done earlier [47] in which the majority of E. faecalis and E. faecium isolates were resistant to one or more of the tested antimicrobials. Resistance to all tested antimicrobials was also observed in both E. faecalis and E. faecium isolates.
More than 50.0% of E. faecalis isolates were resistant to all tested antimicrobials, while 100.0% of the isolates were resistant to tetracycline. On average, E. faecium exhibited increased resistance to antimicrobials in comparison with E. faecalis. Our findings are in agreement with a recent study conducted in Zambia [29]. Our study also has some similarities with a study conducted in the Czech Republic [48], in which increased resistance of Enterococci to tetracycline, erythromycin and nitrofurantoin were observed, as well as a study from USA [49], in which Enterococci resistance to tetracycline, penicillin and ciprofloxacin was documented. Furthermore, our results are consistent with findings from previous studies [50][51][52][53][54][55][56] where high tetracycline resistance was reported. Our results were also comparable with those of the study by Fracalanzza et al. [57], which recorded the resistance of Enterococci to erythromycin to be at 82.0% when intermediate results were included. Nevertheless, that study noted reduced resistance to tetracycline (38.3%) and chloramphenicol (5.7%) compared with our study. The observed increase in resistance to all the antimicrobials tested indicate that poultry from these four districts in Zambia can be a source of MDR Enterococci. However, our study contrasted with other studies [58,59], which reported lower levels of resistance to antimicrobials.
Although the gene aac(6 )-Ie-aph(2")-LA, which encodes resistance to gentamicin, was detected in 60.8% of both Enterococci species tested, an association with susceptibility could not be determined, as discs containing high concentrations of gentamicin (for example 120 µg or 500 µg), which are used to detect high-level aminoglycoside resistance, were not available.
The associations between antimicrobial resistance phenotypes and genotypes in E. faecalis and E. faecium isolates were analyzed. Associations were found between genotypes and tetracycline and erythromycin resistant phenotypes. However, genotypes showed no relationship with vancomycin resistant phenotypes. The disparity observed between the phenotypes and genotypes in the case of vancomycin could be due to the fact that vancomycin resistance in Enterococci can be conferred by different gene clusters [60][61][62].

Study Design and Sites
A cross-sectional study was conducted in selected farms in Chongwe and Lusaka (Lusaka Province) and Ndola and Kitwe (Copperbelt Province) districts in Zambia ( Figure 2). The two provinces are among those that harbor most of the commercial poultry farms in Zambia.
Antibiotics 2023, 12, x FOR PEER REVIEW 9 of 16 chloramphenicol (5.7%) compared with our study. The observed increase in resistance to all the antimicrobials tested indicate that poultry from these four districts in Zambia can be a source of MDR Enterococci. However, our study contrasted with other studies [58,59], which reported lower levels of resistance to antimicrobials. Although the gene aac(6′)-Ie-aph(2″)-LA, which encodes resistance to gentamicin, was detected in 60.8% of both Enterococci species tested, an association with susceptibility could not be determined, as discs containing high concentrations of gentamicin (for example 120 µg or 500 µg), which are used to detect high-level aminoglycoside resistance, were not available.
The associations between antimicrobial resistance phenotypes and genotypes in E. faecalis and E. faecium isolates were analyzed. Associations were found between genotypes and tetracycline and erythromycin resistant phenotypes. However, genotypes showed no relationship with vancomycin resistant phenotypes. The disparity observed between the phenotypes and genotypes in the case of vancomycin could be due to the fact that vancomycin resistance in Enterococci can be conferred by different gene clusters [60][61][62].

Study Design and Sites
A cross-sectional study was conducted in selected farms in Chongwe and Lusaka (Lusaka Province) and Ndola and Kitwe (Copperbelt Province) districts in Zambia ( Figure  2). The two provinces are among those that harbor most of the commercial poultry farms in Zambia.

Sample Collection
A total of 492 freshly voided poultry droppings were collected from layers, broilers and village chickens. Five different visits were made to selected poultry farms in four districts in the Copperbelt and Lusaka Provinces in Zambia (Figure 2). Of the total samples collected, 57 were from farms in Chongwe, while 50 were from Lusaka district in Lusaka Province. Of the 385 samples from Copperbelt Province, 140 and 245 came from Ndola and Kitwe districts, respectively.

Sample Collection
A total of 492 freshly voided poultry droppings were collected from layers, broilers and village chickens. Five different visits were made to selected poultry farms in four districts in the Copperbelt and Lusaka Provinces in Zambia (Figure 2). Of the total samples collected, 57 were from farms in Chongwe, while 50 were from Lusaka district in Lusaka Province. Of the 385 samples from Copperbelt Province, 140 and 245 came from Ndola and Kitwe districts, respectively.

Isolation of Enterococci
Conventional microbiological assays were performed to detect and identify Enterococcus species as described by Facklam and Collins [63]. Briefly, 1 g of poultry droppings was suspended in 9 mL buffered peptone water (BPW) (HIMEDIA, India), mixed and incubated at 37 • C for 24 h. A loopful of the BPW suspension was streaked on Bile Esculin Agar (BEA) (HIMEDIA, India) and incubated at 37 • C for 24 h. Following this, colonial traits were noted and smears of suspect colonies (small black shiny colonies on BEA) were made and stained using Central Drug House's (CDH) Gram's color staining kit from India. Gram-positive cocci appearing in chains, doubles or singles were characteristic of enterococci. A total of 343 suspected Enterococcus isolates were recovered from the 492 samples tested. These were stored in 20% glycerol at −20 • C for subsequent experiments.

Identification of Enterococci Using Analytical Profile Index (API)
Species identification based on phenotypic characteristics and biochemical tests was conducted using BioMérieux's Analytical Profile Index (API) 20 Strep test kits. A total of 37 isolates were identified using the API 20 Strep test kits. The reasons for this are stated in Section 2.1.1.

DNA Extraction
Colonies grown overnight on a blood agar plate were placed in a test tube containing 0.5 mL of molecular grade water, vortexed and boiled at 95 • C for 10 min and then centrifuged for 5 min at 1500× g. The supernatant was pipetted into cryo-vials and stored at −20 • C for further analysis.

Molecular Identification of Enterococci
Molecular identification of the Enterococcus species was conducted using single PCR and the genus-specific and species-specific primers shown in Table 9, following the procedure described by Li et al. (2012) [64]. PCR amplification of elongation factor (tuf ) and D -Ala-D -Ala ligase (ddl) in the extracted DNA was conducted using Phusion Flash High-Fidelity PCR Master Mix (Thermofisher Scientific, USA) in a thermal cycler (Applied Biosystems, Chiba, Japan) under the following PCR conditions: initial denaturation at 98 • C for 2 min followed by 30 cycles of denaturation at 98 • C for 5 s, annealing at 56 • C for 5 s, and extension at 72 • C for 30 s. A final extension was performed at 72 • C for 1 min. PCR amplicons were run on 1.5% agarose gels. The expected bandwidths for tuf and ddl PCR products were 112 bp and 475 bp, respectively. For species identification, species-specific primers (Table 1) targeting the superoxide dismutase (sodA) gene in E. faecalis and E. faecium were used. No primers were available for other species. The PCR conditions were similar to those used for genus amplification, except for the annealing temperature, which was set to 52 • C for both species. Genes conferring resistance to aminoglycosides [aac(6 )-le-aph(2")-LA], which in this study was abbreviated as "aac", macrolides (ermA and ermB), tetracyclines (tetM, tetL, tetK, and tetX) and glycopeptides (vanA) were detected in a single PCR using the gene-specific primers shown in Table 10. One Taq Quick-load 2X Master Mix (Biolabs, Durham, NC, USA) was used for amplification in a thermal cycler (Applied Biosystems, Chiba, Japan). The following PCR conditions were employed: initial denaturation at 93 • C for 3 min followed by 35 cycles of denaturation at 93 • C for 60 s, annealing at 52 • C for 60 s and elongation at 72 • C for 60 s. The final elongation step was performed at 72 • C for 5 min. PCR amplicons were run on 1.5% agarose gels. The expected sizes of the PCR products differed for each gene (Table 10).

Data Analysis
Data were entered, cleaned and validated in a Microsoft™ excel spreadsheet (MS Office Excel ® 2016). The data were then exported to SPSS software ver. 21 (IBM Corp., Armonk, NY, USA). PCR results (positive or negative) were reference variables for descriptive analyses. Univariate analyses were conducted for descriptive statistics and data are presented as frequencies, percentages and prevalence. Funding: This research was funded by the Africa Center of Excellence for Infectious Diseases of Humans and Animals (ACEIDHA), grant number P151847 funded by the World Bank and "the APC was funded by ACEIDHA". This work was partially supported by the Japan Agency for Medical Research and Development with grant numbers JP223fa627005 and JP21wm0125008.
Institutional Review Board Statement: Ethics approval was obtained from The University of Zambia Biomedical Research Ethics Committee (UNZABREC), (Protocol code 797-2020, 16 July 2020). Final study clearance and the authority to conduct research were obtained from the National Health Research Authority.

Informed Consent Statement: Not applicable.
Data Availability Statement: All data supporting the reported results have been provided in this study. Any questions regarding data in this study or any supplementary data that may be required may be provided by the corresponding author upon request.