Multidrug-Resistant Listeria Species Shows Abundance in Environmental Waters of a Key District Municipality in South Africa

The prevalence of bacteria with multidrug-resistance (MDR) is a significant threat to public health globally. Listeria spp. are naturally ubiquitous, with L. monocytogenes particularly being ranked as important foodborne disease-causing microorganisms. This study aimed to evaluate the incidence and determine the antimicrobial resistance (AMR) profiles of multidrug-resistant Listeria spp. (MDRL) isolated from different environmental samples (river and irrigation water) in the Sarah Baartman District Municipality (SBDM), Eastern Cape Province (ECP), South Africa. Molecular identification and characterization were carried out using polymerase chain reaction (PCR) and isolates that exhibited phenotypic resistance were further screened for relevant antimicrobial-resistant genes (ARGs). Findings revealed a total of 124 presumptive Listeria isolates; 69 were molecularly confirmed Listeria species. Out of the confirmed species, 41 isolates (59%) were classified as L. monocytogenes while 9 (13%) were classified as L. welshimeri. All Listeria spp. exhibited phenotypic resistance against ampicillin, penicillin, and trimethoprim-sulphamethoxazole and further screening revealed ARGs in the following proportions: sulI (71%), blaTEM (66%), tetA (63%), and blaCIT (33%). Results confirmed the occurrence of ARGs among Listeria inhabiting surface waters of ECP. The present study indicates that the river water samples collected from SBDM are highly contaminated with MDRL, hence, constituting a potential health risk.


Introduction
Genus Listeria is made up of a group of small rod-shaped, non-spore-forming, facultative anaerobic microorganisms that belong to the family Listeriaceae. These bacterial species are motile and capable of thriving at low temperatures and in harsh conditions [1,2]. The genus Listeria contains 20 major species including L. welshimeri, L. grayi, L. marthii, L. seeligeri, L. ivanovii, L. innocua, and L. monocytogenes and these microbes are present within various environmental niches including soil, vegetables, water, wastewater, and animal feces [3]. Many studies have investigated the occurrence of Listeria species in different environmental sources such as soil [4][5][6], while other studies reported on farm samples [7], livestock along with ready-to-eat (RTE) food products [8][9][10][11]. Additionally, some other studies have investigated the prevalence of Listeria in wastewater treatment plants (WWTPs) suggesting that these pathogens can thrive even after chlorination [12][13][14][15]. Among the common Listeria spp., L. ivanovii and L. monocytogenes are both considered as relevant animal and human pathogens. However, L. monocytogenes is regarded as the third most common foodborne pathogen causing infections in individuals with higher risk, for instance, elderly and immune-compromised people, as well as pregnant women [16]. These infections can lead to a potentially critical and fatal illness known as listeriosis, with (specifically irrigation water) from some farms in the SBDM. Six samples were collected on a once-off regime in the month of September. Water samples were collected accordingly in duplicates by means of sterile bottles (1-L) and conveyed in an ice-box to the laboratory for bacteriological analysis in no more than six hours of collection.

Processing of Samples
For bacteriological analysis, the standard filtration method was followed as described by Olaniran et al. [15] with modification. The water samples (100 mL) were filtered via membrane filter papers (0.45-µm pore size) (Merck, South Africa) with the aid of a vacuum pump. For enumeration of the Listeria spp., each membrane filter was aseptically picked with sterilized forceps and transferred onto the surface of Chromogenic Listeria agar (ISO) Base (Oxoid Limited, Basingstoke, UK) mixed with one vial of OCLA (ISO) selective supplement and OCLA (ISO) differential supplement (Oxoid Ltd., Basingstoke, UK). This followed the standards as described previously by Ottaviani and Agosti (ALOA) in ISO 11290-1:1997 used for enumerating Listeria spp. including L. monocytogenes [33]. The media plates were incubated for 24 h at 37 • C. After the incubation, blue/green colonies with no halo were considered as unconfirmed Listeria spp. on the ISO Base. The data were converted to log 10 CFU per 100 mL [34].

Isolation of Listeria spp. by Enrichment
Due to the slight turbidity of the river and irrigation water samples collected, enrichment was carried out to increase the detection of Listeria spp. by inserting the membrane filter into 90 mL of Tryptic Soy Broth (TSB) and afterward incubated aerobically at 37 • C for 16-18 h (Laboratorios CONDA, Madrid, Spain). Following incubation, the second enrichment was done by aseptically transferring 1 mL of the enriched TSB suspension onto 9 mL of Fraser Broth (Oxoid Ltd., Basingstoke, UK), vortexed, and then incubated at 37 • C for a period of 24 h [34]. From the second enriched suspension, 0.1 mL aliquot of the broth was spread onto Chromogenic Listeria agar (LCA) (Oxoid Ltd., UK), thereafter supplemented using OCLA (ISO) selective supplement with OCLA (ISO) differential supplement, following the standards as defined by Ottaviani and Agosti (ALOA) in ISO 11290-1:1997 and incubated aerobically for 24 h at 37 • C. After incubation, presumptive Listeria isolates (showing blue/green colonies either with or without halos) were subsequently purified onto nutrient agar (NA) (Oxoid Ltd., Basingstoke, UK) and incubation conditions of 37 • C for a period of 24 h were followed. Pure colonies were further inoculated in sterile nutrient broth (NB) following an incubation period of 18-24 h at 37 • C. The overnight culture was stored in 25% glycerol stock and kept in a freezer at −80 • C for further analysis.

DNA Extraction
The boiling method was employed to extract bacterial DNA following the methods described by Maugeri et al. [35]. The presumptive isolates were resuscitated by inoculation into NB and an incubation period of 18-24 h at 37 • C was followed. A loopful from the NB culture was suspended in 200 µL of sterile distilled water contained in 1.5 mL Eppendorf tubes and vortexed. For 15 min, the suspension was heated at 100 • C in an MS2 Dri-Block DB.2A instrument (Techne, Marshalltown, South Africa) to break down the bacterial cells. After heating, the suspension was then centrifuged (at 10,000 rpm for 10 min) in order to take away the cell debris from the DNA material confined within the supernatant. Thereafter, the lysate supernatant was employed as a DNA template in the polymerase chain reaction (PCR) technique.

Molecular Identification of Presumptive Listeria Isolates
The presumptive Listeria isolates were molecularly confirmed using the PCR technique. Table 1 summarizes the set of primers of target Listeria species, their respective PCR product sizes, and PCR conditions. represents minutes " represents seconds.

Multiple Antibiotic-Resistance Phenotypes (MARPs) and Multiple Antibiotic-Resistance Index (MARI)
The MARPs pattern of antimicrobial-resistant Listeria species (ARLS) was generated based on isolates that exhibited phenotypic antibiotic resistance (AR) to 3 or more antimicrobial agents as described by Krumperman [40], MARI was also determined for each isolate. MARI values were calculated using the mathematical formula: where "a" denotes the sum of test antibiotics the isolates displayed resistance to; "b" denotes the total sum of antimicrobial agents used.

Molecular Characterization of the Relevant Antimicrobial-Resistant Genes
The PCR profiling of antimicrobial resistance (AMR) determinants of ARLS involved the screening of 10 antimicrobial-resistant genes (ARGs) encoding for relevant β-lactamases following the method previously described by Dallenne et al. [41]. Table 2 summarizes the primers used and amplification conditions for the detection of extended-spectrum beta-lactamases (ESBL) ARGs. The β-lactamases assay comprised of one singleplex PCR for the detection of cefotaximase-Munich (CTX-M) group 8/25 and three multiplex PCR assay, which include multiplex I PCR for detection of Temoneira (TEM), sulphadryl variable (SHV) and oxacillinase (OXA) β-lactamase; multiplex II PCR is for the detection of plasmid-mediated AmpC β-lactamase types including the forkhead-box (FOX) group, Citron Rho-Interacting Serine/Threonine Kinase and the chromosomal EBC family; multiplex III PCR for the detection of carbapenemase including Klebsiella pnuemoniae carbapenemase (KPC), Verona integon-encoded metallo-beta-lactamase (VIM) and active-on-imipenem (IMI) genes. A total of 14 ARGs were screened using the PCR technique. Table 3 summarizes the list of primers for the screening of relevant ARGs.  [41] represents minutes " represents seconds b Y = T or C; R = A or G; S = G or C; D = A or G or T [41].  represents minutes " represents seconds.

Enumeration and Distribution of Presumptive Listeria spp. in Water Samples
The results showed that the standard plate count for Listeria spp. obtained from the river samples collected from Bloukrans River access point one (BRT) and Bloukrans River access point two (BRD) ranged from 3.49-3.88 log 10 CFU/100mL. The standard plate count from Duncan Farm irrigation-water (DFW) samples ranged from 0-3.68 log 10 CFU/100mL. Therefore, the dissemination of presumptive Listeria spp. was higher for the BRD sampling site in the SBDM, as compared to the other sampling sites.

Molecular Confirmation and Characterization of the Recovered Listeria Isolates
A total number of 124 unconfirmed Listeria isolates were recovered from the water samples analyzed. The results revealed that 55.7% (69/124) harbored the prs gene for the confirmation of Listeria spp. (Supplementary Figure S1). Further characterization revealed 59% (41/69) of the confirmed Listeria isolates were confirmed as L. monocytogenes (Supplementary Figure S2) and 13% (9/69) were confirmed to be L. welshimeri (Supplementary Figure S3).

Antimicrobial Susceptibility Patterns (ASP) of the Confirmed Listeria spp.
All confirmed Listeria spp. were subjected to antibiogram susceptibility testing. The result showed that besides gentamicin and amikacin which had the lowest percentage resistance of 32% and 39%, respectively, the Listeria isolates exhibited resistance (100%) against other test antibiotics including penicillin, cephalothin, ciprofloxacin, levofloxacin, trimethoprim-sulphamethoxazole, nalidixic acid, chloramphenicol, tetracycline, erythromycin, vancomycin, clindamycin, and oxacillin. Among the 15 antimicrobial agents, only amikacin and gentamicin had activity against L. monocytogenes with a low resistance percentage of 39% and 32%, respectively whereas no L. welshimeri isolates exhibited resistance against amikacin. However, L. welshimeri exhibited a resistance of 33% against gentamicin.

MAR patterns and MAR Indices
The MAR phenotype patterns and MAR indices of Listeria spp. were generated and are summarized in Table 4. Results from the phenotypic resistance profiles of L. monocytogenes show four MAR patterns against the antibiotics ( Table 4). The four MARP patterns were generated from 13 to 15 antibiotics. The most frequently observed MARP pattern among the L. monocytogenes was the AP-PG-KF-CIP-LEV-TS-NA-C-T-ERY-VA-CD-OXA pattern with 41.5% (17/41) of the isolates exhibiting the resistance pattern. The MARI values of all L. monocytogenes ranged between 0.87 and 1 (Table 4), which is greater than the 0.2 arbitrary threshold set by Krumperman [40]. MARI values ranging from 0.87 to 0.93 were observed for L. welshimeri, which is greater than the 0.2 threshold (Table 4). A total of two MARP patterns were observed for L. welshimeri isolates as shown in Table 4; with the most frequently observed MARP pattern being AP-PG-KF-CIP-LEV-TS-NA-C-T-ERY-VA-CD-OXA, observed amongst 66.7% of the isolates. Table 4. MAR patterns of L. monocytogenes isolated from the water samples. The ARGs screened in this study were chosen due to the high percentage of MAR phenotypes exhibited by the confirmed Listeria spp. (Table 4). Isolates were screened for 14 relevant ARGs that belong to five different classes of antibiotics including betalactams, phenicols, sulfonamides, tetracyclines, and aminoglycoside. Additionally, the confirmed isolates were screened for 10 relevant ESBL genes. The percentage occurrence and distribution of ARGs harbored by L. monocytogenes and L. welshimeri are summarized in Table 5.

Discussion
The emergence of MDRL poses a serious public health threat to humanity. In this study, our findings provide evidence that the environmental waters of the ECP are probable reservoirs of highly pathogenic Listeria spp. harboring ARGs. The standard plate count of the presumptive Listeria isolates recovered from the selected sample sites was high and this is in accordance with the findings reported by Bilung et al. [7] and Iwu and Okoh [47]. The South African standard guidelines for domestic and irrigation water use are not specific for bacterial concentration. Based on the potable drinking water and irrigation water quality guidelines set by the South African Department of Water Affairs (DWAF) and World Health Organization (WHO) which is ≤1-1000 CFU/100 mL of fecal coliforms [48][49][50], the standard plate counts observed in this study exceeded the acceptable limits. This may be as a result of fecal contamination due to the function of the river as a receiving watershed or run-off from nearby farms involved in animal husbandry. However, in this study, results showed no occurrence of Listeria spp. in the irrigation water samples.
Globally, several studies have also investigated the prevalence of L. monocytogenes present in river and irrigation waters. In the United States, L. monocytogenes was reported in 43% (604/1405) [51] and 31% (53/170) [52] of water samples collected from river and irrigation waters. Furthermore, the bacterium has also been isolated from water samples collected from the point of sewage discharge (into a river) in Europe [53]. The study of Manjur et al. [54] carried out in Bangladesh also investigated the distribution of Listeria spp. confirming the occurrence of the microorganisms in soil and surface water samples. In this study, the percentage occurrences (PO) observed was 55.7% for Listeria species and 59.4% for L. monocytogenes, while the PO of L. welshimeri (13%) was much lower. Similarly, other studies reported Listeria spp. with PO of 64% [7] and 33.30% [55], which were recovered from surface water bodies (in Canada) and environmental samples (in Malaysia), respectively. Contrary to our findings, another recent study by Iwu and Okoh [47] reported low PO of L. monocytogenes (14%) recovered from irrigation water in ECP, South Africa.
The patterns of phenotypic resistance exhibited by Listeria spp. in this study revealed that 86.7% of the isolates displayed MDR against the test antibiotics. In contrast, previous studies reported high susceptibility to penicillin, ampicillin, and tetracycline exhibited by L. monocytogenes recovered from pork and slaughterhouses in Brazil [8] and in retail raw food in China [56], respectively. The ineffectiveness of penicillin against L. monocytogenes observed in this study agrees with another study of Aras and Ardıç [9], which reported L. monocytogenes isolated from turkey and meat samples with resistance percentage of 75% and 66.7% to ampicillin and penicillin, respectively. In contrast, the findings of this study revealed resistance exhibited by L. monocytogenes against the beta-lactams ampicillin, penicillin, and cephalothin. Similarly, Olaniran et al. [15] reported resistance against penicillin exhibited by Listeria spp. isolated from treated wastewater (WW) final effluents and receiving watersheds.
The Listeria spp. in this study exhibited varied susceptibility to amikacin and gentamicin (aminoglycoside class). Likewise, our finding is corroborated by the study of Bilung et al. [7], who revealed effective activity of gentamicin against Listeria isolates that were recovered from environmental samples collected from farms in Malaysia. Generally, Listeria spp. have been reported to be susceptible to a wide range of antibiotics. However, many studies have reported Listeria spp. exhibiting phenotypic resistance against antibiotics used in the treatment of listeriosis such as tetracycline, chloramphenicol, and fluoroquinolones [16,17,57].
The MARP patterns observed in this study revealed L. monocytogenes and L. welshimeri isolates displaying four and two MARP patterns, respectively. Contrary to studies by Arslan and Özdemir [58] and Conter et al. [59], where Listeria spp. recovered from homemade white cheese and also food processing environments, respectively, were reported to exhibit single-drug resistance, MDR against a range of 13 to 15 test antibiotics was observed in the current study. Evaluation of the MAR index of ARLS revealed a MAR index that varied from 0.8 to 1 (Table 4) which is above the arbitrary threshold set by Krumperman [40]. In principle, a MARI value above the set threshold (0.2) suggests that the isolates originate from sources of high antibiotic usage. This may imply that the river waters in this study area are highly contaminated with antibiotics including antibiotic residues (antibiotic resistant bacteria (ARB) and ARGs), or other degradable products. These antibiotics find their way into water bodies via final wastewater effluent discharge, surface runoffs from animal husbandry, agricultural soil upon which manure is used as fertilizer and animal waste from nearby grazing livestock systems [60].
The detection of ARGs harbored by Listeria spp. and the increasing rates of Listeria infections are becoming major concerns to public health [26]. Furthermore, the selective pressure exerted by the exposure of Listeria spp. to antibiotics within an environment niche can increase the transfer of ARGs between Listeria spp. and other bacterial species by self-transferable plasmids [16,17]. The confirmed Listeria spp. were screened for 24 relevant ARGs, out of which only 7 (29.2%) ARGs (sulI, tetA, and ESBL) were detected. Our findings revealed that among the ARGs detected, sulI (which confers AR against sulphonamides) was the most dominant amongst the Listeria spp., particularly L. monocytogenes. Another dominant ARG was tetA, which confers resistance against tetracycline, followed by bla TEM among L. monocytogenes (66%) and L. welshimeri (44%) isolates (Table 5). Similarly, Srinivasan et al. [25] reported a low frequency of ARGs harbored by L. monocytogenes recovered from dairy milk farms. Among the ARGs reported included tetA (32%), and sulI (16%). However, other tetracycline resistance genes screened were not found to be harboured by L. monocytogenes [25].
In this study, most of the confirmed Listeria spp. that exhibited phenotypic resistance did not harbor ARGs. For instance, all L. monocytogenes and L. welshimeri exhibited phenotypic resistance against chloramphenicol. However, only 7% of L. monocytogenes harbored catII gene. A low percentage occurrence against aminoglycosides was observed; however, the screened Listeria spp. did not harbor ARG. Other studies of Davis and Jackson [23] and Morvan et al. [61] also reported Listeria isolates exhibiting phenotypic resistance against aminoglycosides but did not harbor ARGs.
There is insufficient information concerning the occurrence of ESBLs encoding genes harbored by Listeria spp. and this is among the few studies reporting the detection of ESBLs encoding genes harbored by Listeria spp. recovered from the aquatic environment in South Africa.

Conclusions
The findings of this current study revealed the presence of Listeria spp. in river water and irrigation water samples. All the confirmed Listeria spp. exhibited multidrug resistance (MDR) and harbored more than one ARG. The occurrence of MDR strains of L. monocytogenes in the water samples indicates that the bacteria are most likely to be transferred to other ecosystems. The ARGs harbored by the Listeria spp. are most likely to be transferred to other bacterial species within the aquatic environment. Therefore, this poses a significant public health risk and environmental crisis that involves the spread of ARB. Therefore, more comprehensive monitoring of the nature of acquisition and spread of ARGs among Listeria spp. is needed particularly in the agroecosystem. The final effluent from WWTPs should also be appropriately depurated before discharge into receiving water bodies.