Highly clonal relationship among Salmonella Enteritidis isolates in a commercial chicken production chain, Brazil

Abstract

In this study, we described the comparison among pulsed-field gel electrophoresis (PFGE), random amplified polymorphic DNA (RAPD), ribotyping, and PCR-ribotyping methods for subtyping Salmonella Enteritidis isolated from an industrial chicken production chain. One hundred and eight S. Enteritidis were isolated at all stages of poultry meat processing plant. These isolates were pheno- and genotypically characterized by using antimicrobial susceptibility test, phage typing, RAPD, PFGE, ribotyping, and PCR-ribotyping. The highest antibiotic resistance rates were observed for enrofloxacin (18.5%) followed by furazolidone (15.7%), cefoxitin (1.8%), ciprofloxacin, and ampicillin with 0.9% each one, while seven isolates (6.4%) were pan-susceptible. Most strains belonged to the globally disseminated phage type PT4 (n = 74; 69.2%). Additionally, we identified strains belonging to phage types PT1 (n = 19; 17.8%) and PT7a (n = 14; 13.1%). Moreover, our results showed that these four molecular methods indicate similar results showing high similarity (≥ 90%) among S. Enteritidis strains, suggesting that these isolates appear to be from a common ancestor being spread at all stages of the poultry production chain. In summary, the combined molecular approaches of these methods remain a suitable alternative to efficiently subtyping S. Enteritidis in the absence of high-resolution genotyping methods and these results may serve as a baseline study for development of mitigation strategies.

Introduction

Nontyphoidal Salmonella enterica (NTS) is the most important foodborne pathogen worldwide causing diarrhea, vomiting, nausea, fever, and abdominal pain [1]. Among these NTS, Salmonella Enteritidis has been associated with poultry products, which are considered major reservoirs of many non-host-specific serovars of Salmonella, and often, human infections are attributed to the consumption of these products, including eggs and chicken meat [2]. S. Enteritidis is the most prevalent serotype distributed in several countries from Latin America, Europe, Africa, and Asia, and poultry products still play a key role in the spread of this serovar to humans [3].

From an epidemiological perspective, several reports have documented a limited number of genotypes among S. Enteritidis strains, reinforcing the notion that most S. Enteritidis strains are derived from a common clonal group [4–8]. In order to understand this public health concern and address mitigation strategies, many subtyping approaches, including pulsed-field gel electrophoresis (PFGE) and PCR-based methods have been developed to powerfully subtype S. Enteritidis strains and provide valuable epidemiological information [6, 8]. Despite that, little is known about the epidemiology of Salmonella serovars given that these data are still underexplored in Brazil compared with the USA and other countries with particular concern in the clonal relationship of Salmonella strains [8].

In this context, this study aimed to (i) gather information regarding the source of the S. Enteritidis strains isolated from Santa Catarina State, Brazil, (ii) verify which molecular method shows better discriminatory power, and (iii) analyze the correlation between genetic and phenotypic characteristics of the isolates.

Materials and methods

Bacterial strains and sampling

A total of 108 S. Enteritidis strains were included in this study. These isolates were obtained from different points here identified as collection points (CP) in a poultry production chain during summer and winter seasons in Santa Catarina State, second-largest poultry producer of Brazil [9]. Samples were collected with drag swabs from ground (CP1), transport chicken vehicle (CP2), rinse water from chicken crate (CP3), feather swab (CP4), chicken hanging hook swab (CP5), chicken after plucking (CP6), carcass before cooling (CP7), carcass chilling water (CP8), chilled carcass (CP9), packaging (CP10), and storage (CP11) (Table 1).

Table 1.

Source of the 108 S. Enteritidis isolated in an industrial chicken production chain in Santa Catarina, Brazil

Collection point	Processing step	Sample type	No. of strains (n = 108)
CP1	Integrated poultry houses	Drag swab	5
CP2	Reception	Crate swab	10
CP3	Reception	Crate washing water	12
CP4	Reception	Chest feather and cloaca swab	8
CP5	Reception	Chicken hanging hook	3
CP6	Defeathering	Defeathered carcass	12
CP7	Chilling	Carcass before chilling	17
CP8	Chilling	Carcass washing water	5
CP9	Chilling	Chilled carcass	22
CP10	Packaging	Table swab (equipment)	2
CP11	Storage	Frozen carcass/7 days	12

The detection of Salmonella enterica was performed according to Andrews et al. [10]. In brief, samples (25 g and/or 25 mL) were pre-enriched in 225 mL of Buffered Peptone Water (BPW) and incubated at 37 °C for 24 h. For drag swab, samples were pre-enriched in 3 mL of BPW. Subsequently, 0.1- and 1-mL aliquots were transferred to Rappaport-Vassiliadis (RV) (9.9 mL) and tetrathionate (TT) (9 mL), respectively. Samples inoculated in RV broth were placed in water bath at 42 °C for 24 h. Those inoculated in TT broth were kept at 37 °C for 24 h. Samples were then streaked onto Xylose Lysine Deoxycholate (XLD) agar and incubated at 37 °C for 24 h. Colonies presumably identified as Salmonella were transferred to Triple Sugar Iron (TSI) and Lysine Iron Agar (LIA). S. enterica was confirmed by the slide agglutination test using Salmonella O Poly Antisera (anti-O).

Phage and serological typing

Phage typing was performed at the Laboratorio de Referencia Nacional para Colera e Doencas Entericas, Instituto Oswaldo Cruz, FIOCRUZ, using the phage-typing scheme standard protocol [11]. The phage typing was performed by using a phage set for S. Enteritidis comprising PT1, PT4, PT5, and PT7a phages. Briefly, overnight cultures of S. Enteritidis grown at 37 °C in Phage broth were streaked onto Phage agar and 10 μL of each phage type was added individually to each petri dish. After 18 h of incubation period at 37 °C, the phage lysis was analyzed [11]. The phage set was provided by the World Health Organization (WHO) Reference Center for Phage Typing (Colindale, London, UK). All isolates were serotyped traditionally on the basis of somatic O and phase 1 and phase 2 of H flagellar antigens by agglutination tests with antisera, as specified in the Kauffmann-White-Le Minor scheme for Salmonella serotyping [12, 13].

Antimicrobial resistance test

Antimicrobial susceptibility profiles were determined by disk diffusion method by using 12 antimicrobials classified by the WHO [14] as critically important (amikacin (AK; 30 μg), imipenem (IPM; 10 μg), cefotaxime (CTX; 30 μg), ceftazidime (CAZ; 30 μg), ampicillin (AMP; 10 μg), ciprofloxacin (CIP; 5 μg), and enrofloxacin (ENR; 5 μg—veterinary only)); highly important (cefoxitin (FOX; 30 μg), chloramphenicol (CH; 30 μg), sulfamethoxazole/trimethoprim (STX; 25 μg), and tetracycline (TET; 30 μg)); and important (furazolidone (FUR; 15 μg)) antimicrobials. The disk diffusion test and interpretative criteria were used according to the Clinical and Laboratory Standards Institute guidelines [15, 16].

DNA extraction of S. Enteritidis isolates

A genomic prep cell and tissue DNA isolation kit (Amersham Biosciences) was used according to manufacturer’s guidelines to extract DNA from S. Enteritidis (n = 108) cultures after an 18-h incubation period at 37 °C in tryptic soy broth (TSB) (Becton, Dickinson, Franklin Lakes, NJ). Afterward, DNA quality was checked with a NanoDrop spectrophotometer (Thermo Scientific). These DNA samples were used to perform random amplified polymorphic DNA, ribotyping, and PCR-ribotyping.

Random amplified polymorphic DNA

Polymerase chain reactions (PCR) were prepared using master mix ready-to-go random amplified polymorphic DNA (RAPD) analysis beads (Amersham Biosciences), and primers OPB17 (5′AGGGAACGAG3′) and 23L (5′CCGAAGCTGC3′) (Invitrogen, USA) according to Lin et al. [17]. Concisely, a final 25-μL volume containing 10 ng of each S. Enteritidis DNA was amplified by using GeneAmp PCR System 9600 (Perkin Elmer thermal cycler). The PCR conditions were 95 °C for 5 min, followed by 45 cycles of 95 °C for 1 min, 36 °C for 1 min, and 72 °C for 2 min. The amplified products were electrophoresed in 1× Tris-acetate running buffer. The gels were stained with ethidium bromide, destained with ddH20, and the DNA visualized under UV trans-illumination (Gel Doc 2000, Bio-Rad Laboratories).

Ribotyping and PCR-ribotyping

The probe of 1.3 kb for the gene 16S rRNA was prepared by PCR, as described by Pelkonen et al. [18] using the DNA of Escherichia coli ATCC 25922 and primers PELK1 (5′TTCGAGCTCAGATTGAACGCTG3′) and PELK2 (5′ATTGGATCCACGATTACTAGCG3′). The amplification product was electrophoresed and the gel band with the probe was purified using GFX PCR DNA and Gel Band Purification kit (Amersham Biosciences) before hybridization. Salmonella DNA samples were digested by PstI, and after electrophoresis, Southern blot was conducted with a 1.3-kb probe prepared by PCR and marked directly with a non-radioactive marker included in the ECL kit (Amersham Biosciences) according to manufacturer’s instructions. Next, the detection of hybridized fragments after DNA transference was carried out using a peroxidase reagent provided in the ECL kit.

PCR-ribotyping was carried out as described by Kostman et al. [19] using KOST1 (5′TTGTACACACCGCCCGTCA 3′) and KOST2 (5′ GGTACCTTAGATGTTTCAGTTC 3′) primers. Shortly, PCR were performed with 200 ng of DNA sample, 20 mM Tris-HCl (pH 8.4), 50 mM KCl, 1.5 mM MgCl₂, 1 U of Platinum Taq DNA Polymerase (Invitrogen, Brazil), and dNTP mix (200 μM each) in a final volume of 25 μL. Amplifications were carried out by an Eppendorf 6333 Nexus MasterCycler Thermal Cycler using the following PCR conditions: 94 °C for 6 min, followed by 30 cycles of denaturation at 94 °C for 1 min, annealing at 55 °C for 1 min, and extension at 72 °C for 1 min, with a final extension at 72 °C for 3 min. Electrophoreses of amplification products were on 1% agarose gel with a 100-bp ladder (Invitrogen, Brazil) as a molecular weight marker. The gels were stained with ethidium bromide (1 μg/mL) for 20 min and visualized under UV trans-illumination (Gel Doc 2000, Bio-Rad Laboratories).

Pulsed-field gel electrophoresis

The S. Enteritidis strains were analyzed by pulsed-field gel electrophoresis (PFGE) according to the Centers for Disease Control and Prevention [20]. Briefly, S. Enteritidis isolates were grown overnight at 37 °C on Mueller-Hinton (MH) agar (Becton Dickinson, USA). The bacterial cell concentration was adjusted by diluting cells with sterile cell suspension buffer to the OD value 1.00 (range 0.8–1.0) at 610-nm wavelength. Bacterial cells were agarose-embedded and lysed, and intact genomic DNA samples were subjected to digestion with 25 U of XbaI and SpeI restriction enzymes (Amersham Bioscience, USA), individually, for at least 2 h at 37 °C. The fragments were separated by CHEF DR-III Pulsed-Field Electrophoresis System (Bio-Rad Laboratories, Hercules, CA) with the following conditions and reagents: 1% SeaKem Gold agarose (FMC BioProducts, Rockland, ME) in 0.5% Tris-borate EDTA buffer, temperature at 14 °C, voltage at 6 V/cm, run time of 19 h with an initial switch time of 2.2 and final switch time of 63.8 s. The PulseNet standard marker S. enterica serovar Braenderup H9812 strain was used as a molecular reference marker. The gels were stained with ethidium bromide and destained with ddH₂0, and the DNA visualized under UV trans-illumination (Gel Doc 2000, Bio-Rad Laboratories). Dendrograms and cluster analysis were performed using the BioNumerics software package v. 7.6 (Applied Maths, Inc., Austin, TX), including the combined analysis of fingerprint data among RAPD, ribotyping, PCR-ribotyping, and PFGE, which was carried out by using the composite data set plugin available in this software (Fig. 1). Similarity analysis was performed using the Dice coefficient (1.5% band position tolerance), and clustering was created using the unweighted pair group method with arithmetic means (UPGMA).

Fig. 1.

Genetic correlation among 13 genotypes obtained by using PFGE, RAPD, ribotyping, and PCR-ribotyping of S. Enteritidis strains isolated in an industrial chicken production chain. S. Senftenberg (SS), S. Enteritidis ATCC13076 (SE), and E. coli ATCC25922 (EC) were used as control

Results

Prevalence, antimicrobial susceptibility, and phage typing of S. Enteritidis

Overall, S. Enteritidis was isolated from all collection points and the isolation rate observed was slightly higher in winter (n = 55) than in summer (n = 53) (Table 1). Most strains belonged to the globally disseminated phage type PT4 (n = 74; 69.2%). Additionally, we identified strains belonging to phage types PT1 (n = 19; 17.8%) and PT7a (n = 14; 13.1%).

Among the 108 S. Enteritidis isolates, most were resistant to enrofloxacin (n = 20; 18.5%), followed by furazolidone (n = 17), imipenem (n = 3), cefoxitin (n = 2), ciprofloxacin (n = 1), and ampicillin (n = 1) showing 15.7%, 2.7%, 1.8%, 0.9%, and 0.9%, respectively. Interestingly, several strains displayed intermediate resistance against different drugs, particularly, enrofloxacin (n = 65; 60.1%) and furazolidone (n = 57; 52.7%), while all isolates were susceptible to amikacin, ceftazidime, and sulfamethoxazole/trimethoprim as shown in Table 2. Furthermore, twenty antimicrobial resistance patterns (R-type) were identified as summarized in Table 3.

Table 2.

Frequency of antibiotic resistance among S. Enteritidis isolates (n = 108) from an industrial chicken production chain in Santa Catarina, Brazil

Antibioticsa	Resistant (n)	Intermediate (n)	Susceptible (n)
Enrofloxacin (ENR)	20	65	23
Furazolidone (FUR)	17	57	34
Imipenem (IPM)	3	0	105
Cefoxitin (FOX)	2	4	102
Ciprofloxacin (CIP)	1	6	101
Ampicillin (AMP)	1	1	106
Tetracycline (TET)	0	3	105
Chloramphenicol (CH)	0	1	107
Cefotaxime (CTX)	0	1	107
Sulfamethoxazole/trimethoprim (STX)	0	0	108
Ceftazidime (CAZ)	0	0	108
Amikacin (AK)	0	0	108

^aAK amikacin (30 μg), AMP ampicillin (10 μg), CTX cefotaxime (30 μg), FOX cefoxitin (30 μg), CAZ ceftazidime (30 μg), CIP ciprofloxacin (5 μg), CH chloramphenicol (30 μg), ENR enrofloxacin (5 μg), IPM imipenem (10 μg), STX sulfamethoxazole/trimethoprim (25 μg), TET tetracycline (30 μg), FUR furazolidone (15 μg)

Table 3.

Antibiotic resistance pattern of the 108 S. Enteritidis isolates from an industrial chicken production chain in Santa Catarina, Brazil

Resistance pattern (R-type)a	No. of strains (n = 108)
ENR	4 (3.7%)
ENR (CIP)	1 (0.9%)
ENR (FUR)	8 (7.4%)
ENR (CIP-FUR)	2 (1.8%)
ENR (CIP-TET)	1 (0.9%)
ENR-FUR	1 (0.9%)
ENR-FUR (FOX)	1 (0.9%)
ENR-FUR (AMP-CIP-CH)	1 (0.9%)
ENR-FUR (CIP-FOX-TET)	1 (0.9%)
FOX (ENR-FUR)	1 (0.9%)
FOX-FUR (ENR)	1 (0.9%)
FUR	6 (5.5%)
FUR (ENR)	5 (4.6%)
FUR (ENR-FOX)	1 (0.9%)
CIP (ENR-FUR)	1 (0.9%)
AMP-FUR (ENR-FOX)	1 (0.9%)
(ENR)	20 (18.5%)
(ENR-FUR)	34 (31.4%)
(FUR)	10 (9.2%)
(CTX-ENR-FUR-TET)	1 (0.9%)

^aProfiles among parentheses represent strains with intermediate resistance. Each row represents a different R-type

AK amikacin (30 μg), AMP ampicillin (10 μg), CTX cefotaxime (30 μg), FOX cefoxitin (30 μg), CAZ ceftazidime (30 μg), CIP ciprofloxacin (5 μg), CH chloramphenicol (30 μg), ENR enrofloxacin (5 μg), IPM imipenem (10 μg), STX sulfamethoxazole/trimethoprim (25 μg), TET tetracycline (30 μg), FUR furazolidone (15 μg)

RAPD

Genotyping by RAPD using primer OPB17 generated only two clusters being O1 (n = 95) and O2 (n = 13) with band patterns varying between 376 and 1675 bp. The second RAPD primer 23L produced band patterns between 288 and 1943 bp, and the strains were also divided into two distinct groups, L1 (n = 107) and L2 (n = 1).

Ribotyping and PCR-ribotyping

Southern blot hybridization ribotyping (8.5 to 25.3 kb) did not work for all the S. Enteritidis strains tested, and twenty-three (21.3%) strains remained untypeable even after several attempts using Pstl. The remaining eighty-five strains were grouped into five profiles (R1, R2, R3, R4, and untypeable).

PFGE

XbaI-macrorestriction by PFGE showed high genetic similarity, clustering two pulsotypes as X1 (n = 105) and X2 (n = 3), which presented 13 fragments between 30.9 and 938.5 kb. Similarly, the SpeI enzymatic restriction revealed 17 fragments ranging between 3.6 and 885.7 kb that resulted in two clusters S1 and S2 with 92 and 16 strains, respectively.

Genotype determination

Overall, PFGE grouped the 108 S. Enteritidis strains into three profiles (PFGE1 (n = 89), PFGE2 (n = 3), and PFGE3 (n = 16)), and RAPD also generated three profiles (RAPD1 (n = 94), RAPD2 (n = 13), and RAPD3 (n = 1)) (Table 4). On the other hand, we obtained five profiles through ribotyping (R1 (n = 25), R2 (n = 48), R3 (n = 2), R4 (n = 10), and not typeable (n = 23)), while PCR-ribotyping clustered all of the strains into one profile (C1) as summarized in Table 4. The profiles obtained with PFGE, RAPD, ribotyping, and PCR-ribotyping were analyzed together by using BioNumerics software package v. 7.6 (Applied Maths, Inc., Austin, TX) to determine the genetic subtypes (genotypes) as shown in Table 4 and Fig. 1. Thus, it was possible to classify the strains into 13 genotypes (G1 to G13) (Table 4).

Table 4.

Distribution of S. Enteritidis genotypes obtained by using a combined approach of PFGE, RAPD, ribotyping, and PCR-ribotyping

Genotypes	PFGE profile	RAPD profile	Ribotyping profile	PCR-ribotyping profile	Number of strains	Distribution of genotypes by collection pointsb
G1	PFGE1	RAPD1	R1	C1	19	CP1, CP2, CP3, CP4, CP7, CP9, CP11
G2	PFGE1	RAPD1	R2	C1	39	CP1, CP2, CP3, CP4, CP6, CP7, CP8, CP9, CP11
G3	PFGE1	RAPD2	R1	C1	4	CP6, CP7, CP11
G4	PFGE1	RAPD2	R2	C1	7	CP2, CP7, CP8, CP9
G5	PFGE1	RAPD2	R3	C1	2	CP5, CP6
G6	PFGE2	RAPD1	R1	C1	1	CP11
G7	PFGE2	RAPD1	R2	C1	1	CP9
G8	PFGE1	RAPD3	R2	C1	1	CP6
G9	PFGE1	RAPD1	NTa	C1	17	CP1, CP2, CP3, CP4, CP5, CP6, CP7, CP8, CP10
G10	PFGE2	RAPD1	NTa	C1	1	CP9
G11	PFGE3	RAPD1	R1	C1	1	CP3
G12	PFGE3	RAPD1	R4	C1	10	CP2, CP3, CP4, CP5, CP9, CP10
G13	PFGE3	RAPD1	NTa	C1	5	CP1, CP2, CP6, CP7, CP8

^aNT not typeable

^bCP1 integrated poultry houses, CP2 reception, CP3 reception, CP4 reception, CP5 reception, CP6 defeathering, CP7 chilling, CP8 chilling, CP9 chilling, CP10 packaging, CP11 storage

The discriminatory power (D) obtained by combining all genetic methods was 0.804. We observed that eight of the 13 genotypes (G1, G2, G4, G5, G9, G11, G12, and G13) were found mainly, but not exclusively, in samples related to the first two stages of the production chain (integrated poultry houses (CP1) and reception (CP2, CP3, CP4, and CP5)). Four of these genotypes (G1, G2, G9, and G12) were also found in the final stages, including packaging (CP10) and storage (CP11). Moreover, genotype G3 could be found in three different stages, including defeathering (CP6), chilling (CP7), and storage (CP11). Interestingly, genotypes G6, G7, G8, G10, and G11 were related to unique profiles being associated with stages CP11 (storage), CP9 (chilling), CP6 (defeathering), CP9 (chilling), and CP3 (reception), respectively. Genotype G2 was the most frequently found, and strains belonging to this group were isolated from samples related to all stages (CP1-CP4, CP6-CP9, and CP11) of the production chain, with exception of packaging as shown in Table 4.

Discussion

Although the difference of isolation rate was not statistically significant, the similar detection of S. Enteritidis during winter and summer indicates that these strains appear to be highly adapted to the poultry production chain regardless of the weather. Similarly, Ball et al. [21] reported that the prevalence of Salmonella strains isolated from chicken farms in Uganda was higher in the rainy season. On the other hand, studies conducted in Greece [22] and South Korea [23] described the higher isolation rate in summer.

The highest frequency of enrofloxacin-resistant strains is a common trait and is in agreement with previous findings [24–26], indicating the possible overuse of this antimicrobial agent in veterinary medicine as a growth promoter or therapeutic use. This result suggests that these isolates most likely are disseminated along the poultry product chain, once there were isolated from different collection points throughout this poultry meat processing plant.

The overuse of enrofloxacin in broilers has been globally described [27]. This antibiotic is used to control severe respiratory and systemic diseases caused by bacterial infections. Chen et al. [28] demonstrated the impact of enrofloxacin therapy on the spread of resistant bacteria in chickens.

The high prevalence of furazolidone resistance found in this study is remarkable, since such resistance is not usually found in Salmonella isolates; in contrast, this broad-spectrum drug is used to eradicate Helicobacter pylori [29, 30]. Although we did not find the classical R-type (ACSSuT (Ampicillin, Chloramphenicol, Streptomycin, Sulphonamides, Tetracycline)), which is common in S. enterica isolates, twenty R-type were identified including critically important antimicrobials [31]. Moreover, a large number of strains presented intermediate resistance to various antibiotics such as ciprofloxacin, enrofloxacin, furazolidone, tetracycline, ampicillin, cefoxitin, cefotaxime, and chloramphenicol (Table 2) corroborating the overuse of antibiotics and subsequent selective pressure, which constitutes a risk factor for humans and animals to acquire multidrug-resistant (MDR) bacteria.

The high frequency of PT4 phage type reported in this study is associated with virulence factors and has been widely described in S. Enteritidis isolates from poultry, food, humans, and the environment in Brazil or elsewhere [4, 32–37]. The identification of Salmonella strains belonging to phage types PT1 and PT7a is a relevant aspect due to possible changes in the phage-type profile, since such evidence has demonstrated that acquisition of multiple antimicrobial resistance by a strain may change the phage type [37].

RAPD revealed that although being easily applied it presents low accuracy and discriminatory power even when two primers results are combined, whereas PCR-ribotyping was used as an alternative tool to genotyping of S. Enteritidis isolates that showed a non-typeable profile, but results showed only one cluster (C1) harboring all the strains. On the other hand, PFGE findings demonstrated similar discriminatory power for both enzymes; however, the results obtained with XbaI were slightly improved by SpeI for this S. Enteritidis collection.

Regarding genotypes (G1 to G13), although the discriminatory power was greater than that obtained within each method individually, this value is still below the ideal that is above 0.90 [18]. In this context, one of the main factors for the high rate of clonality among S. Enteritidis strains could be attributed to the low discriminatory power obtained with the genetic methods used in this study.

Our results showed that these four molecular methods indicate similar results showing high similarity (≥ 90%) among S. Enteritidis strains. In fact, we were able to confirm the highly clonal relationship among these strains by using this combined molecular approach and the findings described here suggest that S. Enteritidis strains might have descended from a common ancestor being spread at all stages of poultry meat processing plant in agreement with previous studies [4–6]. No correlation between phage types and genotypes, or phage types and resistance profiles could be identified. Similar results have been noted in other studies [4–7].

Conclusions

This survey provided an updated picture of the clonal dissemination of S. Enteritidis in a poultry production chain during summer and winter seasons in Santa Catarina state, which may pose a significant risk to public health. It is important to note that our study highlights that the combined molecular approach of these methods is still a suitable alternative to efficiently subtyping S. Enteritidis in the absence of high-resolution genotyping methods such as whole-genome sequence, and these results may serve as a baseline study for the implementation of mitigation strategies in that area.

Funding

The authors thank the Sao Paulo Research Foundation (FAPESP) (Grants #2013/07914-8) for the financial support.

Compliance with ethical standards

The authors declare that they have no conflict of interest.