Molecular Characterization of Staphylococcus aureus Isolated from Human and Food Samples in Northern Algeria

Staphylococcus aureus is a commensal resident of the skin and nasal cavities of humans and can cause various infections. Some toxigenic strains can contaminate food matrices and cause foodborne intoxications. The present study aimed to provide relevant information (clonal complex lineages, agr types, virulence and antimicrobial resistance-associated genes) based on DNA microarray analyses as well as the origins and dissemination of several circulating clones of 60 Staphylococcus aureus isolated from food matrices (n = 24), clinical samples (n = 20), and nasal carriers (n = 16) in northern Algeria. Staphylococcus aureus were genotyped into 14 different clonal complexes. Out of 60 S. aureus, 13 and 10 isolates belonged to CC1-MSSA and CC97-MSSA, respectively. The CC 80-MRSA-IV was the predominant S. aureus strain in clinical isolates. The accessory gene regulator allele agr group III was mainly found among clinical isolates (70.4%). Panton–Valentine leukocidin genes lukF/lukS-PV were detected in 13.3% of isolates that all belonged to CC80-MRSA. The lukF/S-hlg, hlgA, and hla genes encoding for hemolysins and leucocidin components were detected in all Staphylococcus aureus isolates. Clinical and food isolates harbored more often the antibiotic resistance genes markers. Seventeen (28.3%) methicillin-resistant Staphylococcus aureus carrying the mecA gene localized on a SCCmec type IV element were identified. The penicillinase operon (blaZ/I/R) was found in 71.7% (43/60) of isolates. Food isolates belonging to CC97-MSSA carried several antibiotic resistance genes (blaZ, ermB, aphA3, sat, tetM, and tetK). The results of this study showed that all clones were found in their typical host, but interestingly, some nasal carriers had isolates assigned to CC705 thought to be absent in humans. The detection of MRSA strains among food isolates should be considered as a potential public health risk. Therefore, controlling the antibiotics prescription for a rational use in human and animal infections is mandatory.


Introduction
Staphylococcus aureus is a Gram-positive bacterium, considered as a commensal resident of the skin and nasal cavities of humans and animals. This bacterium can spread from alongside the food chain. In this study, clonal lineages of MSSA and MRSA isolates collected from human and food samples in northern Algeria were characterized using DNA microarrays, and provided useful data on their virulence factor genes and antibiotic resistance-associated genes.

Sample Collection
The sampling procedure of this study was performed between January 2017 to January 2018 in two provinces ("wilayas") of Médéa and Ain-Defla in northern Algeria. The provinces of Médéa and Ain Defla are two neighboring regions located in the central region of Algeria at nearly 100 km from the capital Algiers covering an area of 8866 km 2 and 4897 km 2 , respectively. The two regions are characterized by a temperate Mediterranean climate. The region of Ain Defla and the north of Médéa are characterized by cereal cultivation and bovine farming. The south of Médéa belongs to an agro-pastoral region with animal breeding with a focus on sheep farming.
In this study, 71 volunteers (veterinarians and farmers) having contact with livestock animals and others with less exposure to animals such as administrative clerks were screened for staphylococcal nasal carriage. The nasal samples were obtained from both nares with sterile cotton swabs and were transported to the laboratory for microbiological examination.
Furthermore, 112 food samples (raw milk, beef meat, sausages, chicken meat, and creamery cake) were collected from several commercial vendors. The samples were cooled and transferred to the laboratory for microbiological examinations. Thirty-six S. aureus isolated from clinical samples and nasal swabs were kindly provided from local health-care facilities and private labs in the two provinces and were collected during the same time of food sampling.

Ethical Statement
The study protocol was approved by the Medical Ethics Research Committee of the Yahia Farès University, Urban Pole, Médéa, Algeria, and from the managers of the hospital in which the study was conducted. Informed written consent was obtained from each participant in the study. Confidentiality and personal privacy were respected at all levels of the study. Collected data will not be used for any other purpose.

Isolation of Staphylococcus spp.
The bacteriological analyses of samples from humans (nasal swabs, clinical samples) were performed using conventional methods, as described previously [41,42].
Nasal swabs and clinical samples were inoculated into Brain Heart Infusion (BHI) broth (Oxoid Ltd., Basingstoke, UK) and incubated overnight at 37 • C. Initial broth culture was streaked on Mannitol Salt Agar (MSA) (Oxoid Ltd., Basingstoke, UK) and Columbia agar supplemented with 5% sheep blood and incubated at 37 • C for 24 to 48 h. The plates were inspected for characteristic morphology of staphylococci and suspicious colonies were identified using conventional methods (Gram staining, catalase, and coagulase reactions) in accordance with standard microbiological methods [43]. The food samples were processed according to EN ISO 6888 1-2 1999 [44] using Baird-Parker medium supplemented with 5% of egg yolk emulsion (Oxoid Ltd., Basingstoke, UK) and 0.5% potassium tellurite. Colonies with typical shape were selected for presumptive conventional phenotypic and biochemical identification using catalase and coagulase tests.
Colonies were conserved on BHI broth with 50% of glycerol as well as by freeze-drying method in skimmed milk until further investigations.

MALDI-TOF MS Identification
Matrix-assisted laser desorption-ionization time-of-flight mass spectrometry (MALDI-TOF MS) was used for species identification; the procedure was performed as described by Bizzini and Greub, 2010 [45].
Briefly, all suspected isolates were subcultured on Mueller-Hinton agar (Oxoid Ltd., Basingstoke, UK) supplemented with 5% sheep blood at 37 • C for 24 h. Pure colonies were put in 300 µL of water, vortexed and then precipitated with 900 µL ethanol (96% vol/vol) (Carl Roth GmbH, Karlsruhe, Germany). After centrifugation for 5 min at 10,000× g, the pellet was resuspended with a volume of 50 µL of 70% (vol/vol) formic acid (Sigma Aldrich Chemie GmbH, Steinheim, Germany) and 50 µL of acetonitrile (Carl Roth GmbH, Karlsruhe, Germany). After centrifugation for 5 min at 10,000× g, 1.5 µL of the supernatant was spotted onto a MTP 384 Target Plate Polished Steel TF (Bruker Daltonik GmbH, Bremen, Germany) and allowed to air dry at room temperature. A droplet of 2 µL of matrix solution (α-cyano-4-hydroxycinnamic acid) (Sigma-Aldrich Chemie GmbH, Taufkirchen, Germany) was added on top of each sample spot and again the spots were completely dried. Spectra recording was performed using an Ultraflex instrument (Bruker Daltonik GmbH, Bremen, Germany) and data were automatically analyzed using the Biotyper 3.1 software (Bruker Daltonik, Bremen, Germany).

Molecular Identification of Staphylococcus spp.
DNA was extracted from suspected Staphylococcus colonies using the HighPure PCR Template Preparation Kit (Roche Diagnostics Deutschland GmbH, Mannheim, Germany) according to the manufacturer's instructions. DNA quantity and purity were determined using a NanoDrop™ 1000 spectrophotometer (Thermo Fisher Scientific, Wilmington, MA, USA).
A multiplex real-time PCR assay was performed to detect 16S rRNA, nuc and mecA genes corresponding to Staphylococcus spp., S. aureus, and MRSA, respectively [46]. The multiplex real-time PCR TaqMan assay was carried out with the LightCycler 480 Probes Master kits (Roche Diagnostic Deutschland GmbH, Mannheim, Germany) using the CFX-96 real-time PCR system (Bio-Rad, Hercules, CA, USA), which was used for thermocycling and fluorescence detection as described previously [46]. The real-time PCR amplification was performed in a total volume of 25 µL containing 10 µL of PCR master-mix (1×), 6.75 µL of primers (0.2 µM) and TaqMan probe (0.1 µM) mixture and 5 µL of template DNA (100 pg/µL); distilled water was added for a final volume of 25 µL.

DNA Microarray Analysis
The microarray carried 334 oligonucleotide probes for virulence-associated genes including genes encoding exotoxins and enterotoxins, immune evasion factors, microbial surface components recognizing adhesive matrix molecules (MSCRAMM) as well as resistance determinants and other target genes for typing markers such as species, SCCmec, capsule, and agr group typing markers. DNA microarray analysis was performed on the ArrayStrip platform using the S. aureus specific DNA microarray assay (StaphyType; Abbott (Alere Technologies GmbH), Jena, Germany) following the manufacturer's instructions. Briefly, bacterial colonies were picked from an overnight culture on blood agar. The genomic DNA was extracted after bacterial cells lysis using the EZ1 system (Qiagen, Hilden, Düsseldorf, Germany). Purified nucleic acids were subjected to a linear primer elongation. Biotin-16-dUTP was integrated into the resulting amplicons. Labeled genomic DNA was hybridized to the probes of the microarray. After washing, horseradish-peroxidasestreptavidin conjugate was added followed by precipitation with a special dye. Then, a scanned image was obtained by a microarray reader and subsequently the data were analyzed automatically using software provided by Alere Technologies GmbH. According to difference in the intensity of the colors for each spot, results interpretation was done as "positive" (present), "negative" (absent) or "ambiguous" as described previously [10,47,48].
The clonal complex recognition was based on analysis of hybridisation profiles with regard to non-motile genes, as previously reported [49].

Results
In this study, 60 S. aureus were isolated and identified using MALDI-TOF MS and real-time PCR. Twenty-four strains were isolated from 112 food samples, 16 were isolated from 71 nasal swabs and 20 clinical strains were kindly provided from hospital and private labs collection. The origin and distribution of S. aureus are shown in Table 1.
Staphylococcus aureus isolated in this study were grouped into 14 different clonal complexes using DNA microarray analysis (Table 1; Figure 1 and Table S1). Out of 24 S. aureus isolated from food, 6 different CCs were identified, while 9 different CCs were determined for 20 clinical isolates and those of nasal swabs (Table 1).
Pathogens 2021, 10, x FOR PEER REVIEW 6 of 19 Figure 1. Graphic visualization of source of the isolates (clinical, food and nasal swabs) in circular layout [50], revealing the association between the MSCRAMMs genes, Methicillin resistance and SCCmec Typing, capsule/biofilm associated genes, virulence-associated genes, toxin genes, antibiotic resistance genes, accessory gene regulator allele, and the clonal complexes detected by microarray using interconnections with a variety of colored ribbons. The circumference (track) of the circle is divided into arcs of varying lengths according to the abundance of every single factor.  [50], revealing the association between the MSCRAMMs genes, methicillin resistance and SCCmec Typing, capsule/biofilm-associated genes, virulence-associated genes, toxin genes, antibiotic resistance genes, accessory gene regulator allele, and the clonal complexes detected by microarray using interconnections with a variety of colored ribbons. The circumference (track) of the circle is divided into arcs of varying lengths according to the abundance of every single factor.
The occurrence of virulence-associated factors as detected by microarray analysis is summarized in Tables 2 and 3.
The ica operon ACD responsible for intercellular adhesion protein AB and biofilm PIA synthesis protein D was detected in all isolates, whereas the bap gene encoding surface protein involved in biofilm formation was found only in one CC15-MSSA isolate recovered from a pus sample of a patient with a hospital-acquired infection.
The gene encoding for toxic shock syndrome toxin 1 (tst1) was found in 8 human isolates assigned to CC22-MSSA and CC22-MRSA. Genes etA/B encoding for exfoliative toxins were not detected, while the etD gene was present in all CC80-MRSA isolates. The edinB gene encoding for epidermal cell differentiation inhibitor B was likewise detected in all isolates belonging to CC80-MRSA whereas, edinA and edinC genes were not detected.
Genes encoding for enterotoxins were detected in several isolates: The sea gene was found in 17 isolates (10 human isolates and 7 food isolates) belonging to CC1-MSSA, CC5-MSSA, CC6-MRSA, CC8-MSSA and CC30-MSSA. The seb and sec genes were identified in 5 and 2 isolates, respectively. The enterotoxin gene cluster egc was present in 17 isolates. The clinical isolates carrying the sea gene were predominantly collected from samples of community infections and the food isolates harboring this gene originated from milk and minced beef meat.
The genes responsible for antimicrobial resistance detected by microarray analysis are shown in Table 4.
The erm(C) gene encoding for macrolide/clindamycin resistance was found in seven MRSA isolates, while only two MSSA isolates (food samples) assigned to CC97 carried the erm(B) gene.
Fusidic acid resistance gene fusC was detected in all CC1 isolates. The tetK gene associated with tetracycline resistance was identified in 15 S. aureus isolates (7 food and 8 clinical isolates). In one food isolate belonging to CC1-MSSA, tetK and tetM were detected simultaneously.
Additionally, the aadD gene, which is responsible for tobramycin resistance, was detected in one clinical isolate.

Discussion
To estimate the potential health hazards caused by S. aureus transmission, it is imperative to study their origins, lineages and dissemination simultaneously on humans, in the environment, on animal farms and alongside the food chain.
The present study was aimed to provide relevant information (clonal complex lineages, agr types, virulence and antimicrobial resistance-associated genes) about S. aureus isolated from a variety of human clinical, nasal and food samples in northern Algeria based on DNA microarray assay. This study focused only on two provinces (Médéa and Ain-Defla)in Algeria and hence our results cannot be generalized for a larger scale.
In this study, 24 S. aureus from food sources originated predominantly from raw milk and minced beef meat and only three of them were identified as MRSA. These findings corroborated several studies reporting low percentages of MRSA in food samples [51][52][53].
The majority of clinical isolates of S. aureus were recovered from pus and suppurative infections in which 50% were identified as MRSA (7 CA-MRSA and 3 HA-MRSA). It is known that S. aureus isolates are considered as a principal pathogenic agent in skin and soft tissue infections [54].
Out of 16 S. aureus isolated from nasal carriage from individuals with contact to livestock animals, four (25%) were identified as MRSA. These findings were lower compared to previous reports conducted in Algeria on resembled populations [35,55].
Molecular genotyping of isolated S. aureus resulted in assignment to a high variety of CCs. The general trend showed that CCs in clinical or food samples were closely similar, in contrast to those detected in nasal carriage. The lineages found in food isolates mostly belonged to CC1-MSSA and CC97-MSSA. The majority of CC97 (62.5%) originated from milk samples. Previous studies indicated that S. aureus isolated from bovine milk belonged to CC97 and were mainly represented by isolates from ruminants [9,10,20,56,57]. CC97 strains are occasionally isolated from humans [58] and rarely found as MRSA [59].
Two human isolates belonged to the common bovine lineage CC97, suggesting a possible jump from cattle to humans, as reported previously [60]. A previous work carried out in the same region as the present study indicated that CC97 is associated with bovine subclinical mastitis [55], suggesting a possible origin of these milk samples from infected udders without rigorous surveillance.
The majority of CC1-MSSA isolates originated from meat samples and the remaining came from milk samples which in agreement with a previous study conducted by Wu et al., 2018 reported that most of S. aureus isolated from retail meat and meat products belonged to CC1-MSSA [61]. The CC1-MSSA is thought to be of human origin and began to emerge within bovine S. aureus populations causing mastitis [62], which could subsequently contaminate milk, as previously reported [9].
In this study, three MRSA belonging to CC6-MRSA-IV and CC80-MRSA-IV were isolated from food samples (minced beef meat). These lineages are likely to be found in human samples [25,63] so that this observation might indicate a possible contamination by humans at the slaughterhouse or during processing [64]. Another study showed that commercially distributed meat could play a role in the spreading of CA-MRSA [65]. However, there is no evidence that the transmission of such lineages (lacking enterotoxin genes) would increase the risk of foodborne infections [66].
Six MRSA strains isolated from clinical samples belonged to CC80-MRSA-IV. They were PVL positive and carried SCCmec type IV. This lineage has previously been found in clinical samples in Algeria [15,25,37,67], Tunisia [68], Egypt [16], Europe [23], North America [18], the Middle East [69] and Australia [19]. The results suggest that this clonal complex is a highly prevalent MRSA lineage that has already spread across the North African countries, Egypt and the Middle East and that also can be found sporadically beyond that region.
Seven of the clinical MSSA strains were assigned to CC1, CC5, and CC15 which in agreement with previously reported diversity of CCs in MSSA isolated from patients with infected diabetic foot ulcer in which MSSA isolates belonged to CC1 and CC15 [37].
Staphylococcus aureus from nasal swabs essentially belonging to CC22 (CC22-MSSA and CC22-MRSA); only one isolate belonged to CC80-MRSA IV. The results coincide with previous report, which found that MRSA in nasal carriers on farms belonged to CC22-MRSA-IV and CC80-MRSA-IV [55]. The prominent epidemic CC22 lineage is known as a typical widespread human-associated clone, and there are several distinct MRSA strains that are responsible for both, hospital-acquired and community-acquired MRSA infections [59,70].
One nasal carriage isolate was assigned to CC398-MSSA. The results of a former study strongly suggested that LA-MRSA CC398 originated in humans as MSSA [71]. This clone has been recovered also from food of animal origin [72]. Staphylococcus aureus of ST398 is known to have comparable low host specificity and is able to colonize or cause infections in a broad variety of animal species, including humans [59,73]. It is worth to note that isolated CC398-MSSA lacked many virulence factors such as the Panton-Valentine leukocidin toxin or enterotoxin genes; also, no antibiotic resistance determinants were detected. In fact, the transmission of some clones from animals to humans and vice versa is accompanied by the acquisition of adaptation factors and at the same time loss of certain virulence factors [73,74].
The present findings on human nasal carriage revealed that one S. aureus belonged to CC705-MSSA. It was reported that this clone was not of human origin but it is a common lineage in cattle [60,75]. In the same way, Agabou et al., 2017 found only two CC705-MSSA isolates in Algerian cattle nasal carriage without any detection of this clone in noses of farm workers [35]. Furthermore, it was reported that CC705 (ST705) has emerged as a dominant lineage derived from bovine milk worldwide [56]. Thus, we assume a transmission from an animal or food source to the human carrier.
Two isolates originating from nasal carriage were assigned to CC15-MSSA. Monecke et al. (2009) reported that CC15-MSSA are mainly isolated from healthy carriers [76]. In addition, it was suggested that MRSA from this lineage are extremely rare [59]. Two other human isolates assigned to CC30-MSSA were identified in this study. A previous study showed that CC30 is an important clonal complex from which virulent MSSA as well as HA and CA-MRSA originated [59].
In the present study, other S. aureus lineages, such as CC30 and CC45, were recovered from nasal carriers. These CCs were described in the past among other clusters of strains colonizing healthy individuals in Europe [73,76,77].
In this study, Panton-Valentine leukocidin genes lukF/lukS-PV were detected in 8 S. aureus isolated from clinical samples and belonging to CC80-MRSA. In concordance, a study described that all MRSA harboring lukF/lukS-PV belonged to CC80 [37]. Other reports confirming that PVL gene carriage associated with CC80 is widely distributed in North African countries, including Algeria [15,25].
From the three MRSA isolated from food samples, we found only one PVL-positive. Recently, Mairi et al., 2019 found also some MRSA-PVL positive isolates from food belonging to the ST80-IV CA-MRSA clone [52]. However, MRSA PVL-positive clonal lineages are rare in milk products [78] or in animals [35,79].
The leukocidin genes lukF-PV(P83) (F component from hypothetical leukocidin of ruminants) and lukM were detected in CC705 (nasal carriage isolate) and CC479 (food isolate). Schlotter et al., 2012, found that the majority of CC479 S. aureus isolates from milk of dairy herds harbored the leukocidin genes lukF-(P83)/lukM [80]. In a comparative study of 456 strains of S. aureus isolated from milk of bovine intramammary infections and bulk tanks obtained from 12 European countries, it was reported that isolates were predominantly assigned to CC705 [81]. In addition, several studies reported that CC705 is a common bovine clone among bovine mastitis S. aureus isolates [55,60,82] and from nasal carriage in cattle [35]. The detection of leukocidin genes lukF-PV (P83)/lukM in a human nasal swab isolate CC705 suggests zoonotic transmission from bovine to humans in close contacts.
In this study, sea and seh genes were notably detected in CC1 MSSA isolated from food samples. Staphylococcal food poisoning (SFP) is an intoxication caused by ingestion of contaminated foods (milk, cream filled pastries, sandwich filling, sausages, ground meat, salads, and cooked meals) containing sufficient quantity of staphylococcal enterotoxins (SEs). The most important enterotoxins involved in SFP are the classical SEA, and the SEH [3]. Monecke et al., 2011 found that all CC1-MRSA isolates harbored the enterotoxin gene seh [59]. Others found the seh gene in S. aureus isolates from milk and milk products [78]. Also, Jørgensen et al., 2005 reported that SEH was responsible for milk-associated food-poisoning outbreaks [83].
It has been reported that isolates harboring the gene encoding for toxic shock syndrome toxin 1 (tst1) in combination with other toxin genes were highly pathogenic for humans [84]. In this study, isolates possessing the tst1 gene were negative for lukFS-PV, lukF-PV (P83), luk-M, sea, seb and sec genes.
Biofilm formation is regulated by expression of polysaccharide intracellular adhesion (PIA) protein, which mediates cell-to-cell adhesion and is the product of the ica locus containing icaA/B/C/D [85]. The operon icaACD was detected in all isolates. The bap gene encoding surface protein involved in biofilm formation was found only in one CC15-MSSA isolate. All MRSA SCCmec type IV isolates possessed the icaACD genes responsible for biofilm formation and carried the MSCRAMMs genes. In concordance, Mirani et al., 2013 found that 98.3% of foodborne MRSA isolates carried the icaA/SCCmec IV profile [86], while an association of SCCmec type V and icaA gene was also reported [87,88].
The present study showed a high rate of MRSA in clinical samples (50%). The national rate in Algeria was approximately 35% for hospitalized patients (www.sante.dz/aarn/ documents/pdf/rapport16.pdf, accessed on 27 September 2021). Several Algerian studies focusing on human isolates showed that the MRSA rate has increased almost tenfold in just two decades from less than 5% in 1996-1997 [89] to 50% in 2011 [25] and 62.2% in 2014 [90].
Resistance to antibiotics can result from the imprudent use of antimicrobial agents in human and animals. In this study, 71.7% of all S. aureus and 88.2% of MRSA isolates carried the penicillinase operon blaZ. Previous Algerian studies reported penicillin resistance in 96 to 100% of clinical isolates [92] and in nasal human carriage, 91.6% of isolates carried the blaZ gene [35].
The detection of genes encoding resistance towards tetracycline (tetM and tetK) was not surprising, as this broad-spectrum antibiotic is widely used by Algerian veterinarians and farmers [93]. A resistance rate as high as 85% to tetracycline among MRSA strains isolated from poultry was recently reported [94]. In addition, tetracycline resistance could be expected in other bacterial genera as a consequence of the extensive use of tetracycline in livestock [93].

Conclusions
The DNA microarray results obtained in this study showed that CC 80-MRSA-IV is a significant pathogen among clinical isolates. This clone was found to be associated with many virulence factors (lukF/lukS-PV, edinB and sea) and antibiotics resistance genes (sat, aphA3, far1, tetK and blaZ). In addition, food isolates mostly assigned to CC1 and CC97 were found to harbor enterotoxins genes and antibiotic resistance genes. It was also observed that almost all clones were found in their typical host, but interestingly, some nasal carriers had isolates assigned to CC705 thought to be absent in humans. Our data suggest a possible jump from animals to humans and constitute a dissemination risk into the community. Staphylococcus aureus continues to be a serious human health challenge and the detection of MRSA strains among food isolates should be considered as a potential public health risk. It is necessary to monitor the health status of animals and improve hygiene conditions in the food chain to limit the dissemination of pathogenic and multidrug-resistant strains. Therefore, it is crucial to revise antibiotic prescriptions for rational use in human and animal infections. Informed Consent Statement: Informed consent was obtained from all subjects involved in the study.

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