Assessment of the Microbiological Acceptability of White Cheese (Akkawi) in Lebanon and the Antimicrobial Resistance Profiles of Associated Escherichia coli

Dairy foods are a staple in Lebanon, a low- and middle-income country that has been experiencing serious challenges to food safety and antimicrobial stewardship among other issues. The microbiological acceptability of dairy products has been of increasing concern. This is partially due to the failing economy and prolonged power outages that affect the quality of raw material and disrupt the dairy cold chain, respectively. Therefore, we assessed the microbiological acceptability of Akkawi, a popular white-brined cheese in Lebanon. For this purpose, we quantified the densities of Escherichia coli (a fecal indicator) and Staphylococcus aureus in cheeses collected from Lebanese retail stores. Additionally, we evaluated the antibiotic resistance profiles of the E. coli isolated from the cheese. E. coli and S. aureus were detected in 40 (80%) and 16 (32%) of the 50 cheese samples, respectively. Notably, 40 (80%) and 16 (32%) of the samples exceeded the maximum permissible limit of E. coli and S. aureus, respectively. A high percentage of the 118 E. coli isolated from the cheeses showed resistance to clinically and agriculturally important antibiotics, while 89 (75%) isolates were classified as multidrug-resistant (MDR). Given that Akkawi can be consumed without cooking, our findings highlight serious food safety and antimicrobial resistance problems that require immediate interventions.


Introduction
Dairy products are important components of a healthy and well-balanced diet [1]. White cheeses are particularly popular in many Mediterranean countries and are considered a nutritious staple food, providing proteins and micronutrients like calcium and phosphorous [2]. However, cheeses are high-risk foods that can be consumed without cooking, and contaminated cheeses can result in serious foodborne diseases. For example, several foodborne outbreaks were linked to cheese contaminated with Escherichia coli, Salmonella serovars, Listeria monocytogenes, and Staphylococcus aureus in Europe [3][4][5][6][7]. Additionally, several studies have reported the detection of E. coli, S. aureus, and L. monocytogenes in unpasteurized soft cheeses in the USA [2,8]. These contaminants pose a problem, because contamination of the food with enteric pathogens [48,49]. Additionally, the antibiotic resistance profiles of E. coli have been used as indicators for the emergence and spread of AMR in foods [50]. Taken together, we aimed to evaluate the microbiological safety of Akkawi, a highly popular white cheese, which we collected from the retail markets in Lebanon. Specifically, we quantified the prevalence and densities of E. coli and S. aureus in Akkawi cheese and compared them to the official cheese standards adopted by LIBNOR (Lebanese Standards Institute) [51,52]. Additionally, we assessed the antibiotic resistance phenotypes of the E. coli isolated from the cheese in order to evaluate the dissemination of AMR in this food product.

Prevalence and Densities of E. coli and S. aureus
E. coli was detected in 40 (80%) of the 50 cheese samples. The average densities of E. coli ranged between 1 × 10 3 and 2 × 10 7 CFU/g. The highest average density of E. coli was in Akkawi samples [AA] at 2 × 10 7 CFU/g ( Figure 1). S. aureus was detected in 16 (32%) of the cheese samples, with densities ranging between 1 × 10 2 and 2 × 10 5 CFU/g. The highest S. aureus density (2 x10 5 CFU/g) was detected in sample [SA] ( Figure 1). There was no statistically significant (p value > 0.05) correlation between E. coli and S. aureus densities. The identities of the E. coli and S. aureus isolates selected from positive cheese samples were further confirmed by PCR analysis.
The contamination of cheeses can occur during any step of processing, from milk production to cheese handling and storage. The densities of bacteria like E. coli and S. aureus are used as microbiological indicators to determine the acceptability and safety of cheese [9]. High bacterial densities of fecal indicators, like E. coli, also suggest the possible contamination of the food with enteric pathogens [48,49]. Additionally, the antibiotic resistance profiles of E. coli have been used as indicators for the emergence and spread of AMR in foods [50]. Taken together, we aimed to evaluate the microbiological safety of Akkawi, a highly popular white cheese, which we collected from the retail markets in Lebanon. Specifically, we quantified the prevalence and densities of E. coli and S. aureus in Akkawi cheese and compared them to the official cheese standards adopted by LIBNOR (Lebanese Standards Institute) [51,52]. Additionally, we assessed the antibiotic resistance phenotypes of the E. coli isolated from the cheese in order to evaluate the dissemination of AMR in this food product.

Prevalence and Densities of E. coli and S. aureus
E. coli was detected in 40 (80%) of the 50 cheese samples. The average densities of E. coli ranged between 1 × 10 3 and 2 × 10 7 CFU/g. The highest average density of E. coli was in Akkawi samples [AA] at 2 × 10 7 CFU/g ( Figure 1). S. aureus was detected in 16 (32%) of the cheese samples, with densities ranging between 1 × 10 2 and 2 × 10 5 CFU/g. The highest S. aureus density (2 x10 5 CFU/g) was detected in sample [SA] ( Figure 1). There was no statistically significant (p value > 0.05) correlation between E. coli and S. aureus densities. The identities of the E. coli and S. aureus isolates selected from positive cheese samples were further confirmed by PCR analysis.  The prevalence and densities (colony-forming units per gram; CFU/g) of Escherichia coli (grey dot) and Staphylococcus aureus (orange triangle) in Akkawi cheese in Lebanon. The red line indicates the maximum allowable limit for E. coli and S. aureus (1000 CFU/g) according to the Lebanese Standards Institution (LIBNOR, standard 2002:223). The black line indicates the permissible limit of E. coli and S. aureus (10 CFU/g) according to the LIBNOR standard 2003:495.

Evaluation of Cheese Accessibility by Comparing E. coli and S. aureus Densities to LIBNOR Standards
According to LIBNOR 2002:223, the maximum allowable limit of E. coli and S. aureus in cheese is 1000 CFU/g; however, this limit was decreased to 10 CFU/g in the updated LIBNOR 2003:495. Subsequently, using E. coli densities, 40 (80%) of the cheese samples were found to be microbiologically unacceptable (unsafe) and would be rejected. Furthermore, 8 (16%) of the cheese samples exceeded the maximum allowable limit of 1000 CFU/g of

Discussion
Access to safe and nutritious food is necessary in order to maintain a healthy and productive community. Food safety is also a key to food security, which is especially critical in a country like Lebanon that has been experiencing increasing poverty due to an economic collapse. The latter has exacerbated pollution, the debilitation of infrastructure, and the proliferation of infectious diseases and antimicrobial resistance in the country [14,36,53]. Consequently, monitoring the safety of important foods is a critical need in Lebanon. Here, we focused on the safety of Akkawi, a white-brined cheese that is widely consumed (with or without cooking) in Lebanon and neighboring countries [54].
Our results revealed relatively high densities of E. coli and S. aureus in 80% and 32% of the cheese samples, respectively (Figure 1). We also found that 80% of the cheese samples would be rejected (deemed unacceptable for human consumption) based on the E. coli densities and LIBNOR standards (2002:223 and 2003:495), while up to 32% of the samples would be rejected based on S. aureus densities (Figure 1). Therefore, this study reported higher non-conformity rates in comparison to the limited studies available on dairy products in Lebanon [14,21]. Beyond the use of bacterial indicators, it is known that the enterotoxins secreted by S. aureus can cause gastrointestinal disease [9]. Given that a considerable number of people (~25% in some populations) carry S. aureus on their skin and/or in their nasal cavities, unhygienic practices while handling the cheese can lead to contamination [55,56]. This highlights the potential role of food handlers in the contamination of the cheese with important pathogens. Notably, it was estimated that S. aureus causes around 241,000 foodborne illnesses in the USA annually [55].
The high densities of E. coli, a fecal indicator, suggest the potential contamination of these cheeses with other enteric pathogens associated with fecal contamination [24,50]. Taken together, these results corroborated our previous analysis that showed that dairy foods, including cheeses, were at high risk of contamination with coliforms, L. monocytogenes, and E. coli [14]. Furthermore, fecal contamination can potentially occur at any stage of Akkawi cheese production. Although Akkawi is commonly produced from pasteurized milk, contamination can result from deficiencies in pasteurization as well as post-pasteurization processes that include hand-packing the cheese in draining hoops and curing it in brine solutions. Therefore, contamination can be attributed to the weak implementation of good hygienic practices among food handlers and food producers and/ or deficiencies in production and storage due to economic challenges that include unsustainable access to sources of electricity, clean water, and high-quality raw material.
Akkawi cheeses are not always sold in vacuum-sealed packages in Lebanon, especially those that are sourced from traditional and semi-modern dairies that usually provide the cheese in containers (buckets) filled with brine solution, and food handlers have to partition the cheese according to customers' needs at the retail store. During sampling, we observed that the food handlers were wearing gloves, but that the gloves were used for extended periods of time and were not changed regularly. Moreover, the same cutting boards and knives were used to cut different cheese types and were not cleaned regularly, which increase the risk of cross-contamination. However, we found that even vacuum-sealed Akkawi cheese samples were contaminated with high densities of E. coli and S. aureus, which also suggested that the contamination might have occurred at the processing facilities prior to reaching the market, as explained earlier [25]. Beyond this, it is important to highlight that environmental pollution (e.g., contaminated water sources and pastures) can have major impacts on the safety and quality of dairy products. For example, a report in 2013 showed that the farming environment has a significant effect on the microbiological quality of milk [26], which would persist under deficient cheese production controls and processes. Taken together, strengthening food safety knowledge, endorsing the application of good personal hygiene, ensuring the availability of necessary factors (power, clean water and environment, and good quality raw material among others), establishing a sustainable contaminant-monitoring programs, and adherence to food safety standards are needed in order to maintain the safety of cheese and decrease the burden of foodborne illnesses.
The high prevalence of MDR E. coli in Akkawi cheese was not surprising (Figures 2 and 3; Table 1), because several studies have reported a widespread antibiotic resistance in the Lebanese community, as well as in environmental and food matrices [29,34,35,[37][38][39]46]. A recent study on dairy products in Lebanon screened 29 E. coli isolates that were found to be resistant to AMC (69% of isolates), CFM (11%), CHL (31%), and STR (100%), and all the tested isolates were multidrug-resistant [21]. However, in the aforementioned study, the number of E. coli isolates (n = 29) and antibiotics tested was very limited. In comparison, we tested 118 E. coli against 19 antibiotics that belonged to 9 different classes. In our study, the resistance against penicillins; AMP (67% of isolates), cephalosporins; FEP (19%,), CTX (41%), LEX (67%), CFM (17%), aminoglycosides; GEN (41%), KAN (36%), STR (58%), tetracyclines; TET (40%), and sulphonamides; SXT (38%) (Figure 3) was concerning but also expected, as these classes of antibiotics are heavily used in dairy cattle farming [57,58]. Our observations were further corroborated by a previous study that suggested that elevated levels of resistance against gentamicin can be attributed to the high use of this antibiotic in feed additives on Lebanese farms [59]. Additionally, another study conducted on dairy farms reported that aminoglycosides, which are normally used in clinical settings, are frequently used on animal farms in Lebanon [60]. Similarly, tetracycline was also reported to be extensively used in the Lebanese agricultural sector [60].
We noted that 56.7% of the E. coli isolates in our study were resistant to carbapenems, which is highly concerning. Carbapenems are considered last-resort antibiotics that are used as a salvage therapy to treat complicated MDR and extensively drug-resistant (XDR) Gramnegative infections [61]. When bacterial isolates develop resistance to carbapenems, colistin, one of the highest priority critically important antibiotics, is used as a viable treatment option [61]. However, the wide-emergence of resistance to colistin, due to the presence of the mobile colistin-resistant gene (mcr), has been well-document in Lebanon in various agricultural and environmental matrices [28,29,[33][34][35][37][38][39]41,45,46]. The occurrence of resistance to critically important antibiotics in the food chain is a threat to public health and agriculture, heralding a rise in difficult-to-treat infections in both humans and farm animals. The high dissemination of antibiotic resistance in Lebanon has been attributed to the extensive reliance on and misuse of antibiotics in both animal farming and clinical settings [23]. It has been also established that there is a notable gap in knowledge about judicious antibiotic use among farmers in Lebanon [62]. For instance, dairy farmers believe that the use of antibiotics for prophylaxis and growth promotion is always beneficial, and that stopping the administration of antibiotics to their animals will lead to major economic losses [62]. Taken together, an immediate action plan is needed to address the dissemination of antibiotic resistance in Akkawi cheese and other food products.

Sample Collection
A total of 50 Akkawi cheese samples were aseptically collected from 16 major retail stores across Beirut, the capital of Lebanon, between September 2020 and October 2021. Forty-three cheese samples belonged to 9 brands, while 7 were unbranded. All samples were made from cows' milk and collected from different lots. While Akkawi is normally made from pasteurized milk, this process was not evident in cases of unbranded cheese samples. The cheese samples were transported in a cooler with ice and processed immediately upon arrival at the laboratory (within 2 h of collection).

Sample Processing and Enumeration of E. coli and S. aureus
Each sample was aseptically processed by transferring 25 g of cheese to a sterile stomacher bag (Fisher Scientific, New Hampshire, USA) filled with 225 mL of buffered peptone water (BPW) (Oxoid, Hampshire, UK) [36]. The samples were then homogenized for 1 min at 250 rpms using a stomacher (Thomas Scientific, NJ, USA). The homogenates were 10-fold serially diluted (10 −1 , 10 −2 , 10 −3 , 10 −4 ), and 100 µL of each dilution was plated in duplicate on the selective Chromogenic Tryptone Bile X-Glucuronic (TBX) agar (Oxoid, Hampshire, UK), and Baird-Parker (BPA) agar (Millipore-Sigma, Darmstadt, Germany) for the enumeration and isolation of E. coli and S. aureus, respectively. TBX plates were incubated at 37 • C for 24 h, and BPA plates were left for 48 h at 37 • C under aerobic conditions. After incubation, colonies that matched the diagnostic phenotypes of E. coli (Blue/Green colonies on TBX) and S. aureus (black shiny colonies with a halo on BPA) were counted, and the bacterial densities were reported as colony-forming units per gram of cheese (CFU/g) by averaging the counts from the duplicates.

Polymerase Chain Reaction (PCR) for the Confirmation of Identity of E. coli and S. aureus
Putative E. coli colonies were suspended in 100 µL of DNase-free water and placed in a water bath at 95 • C for 10 min to extract DNA [63]. The E. coli isolates were screened for a species-specific 16S rRNA gene fragment [64,65]. Briefly, PCR reactions (20 µL) were prepared, as reported elsewhere [38,39], and subjected to the following program: an initial denaturation at 95 • C followed by 38 cycles, each consisting of denaturation at 95 • C for 30 s, annealing at 58 • C for 45 s and extension at 72 • C for 1 min, followed by a final extension at 72 • C for 10 min [65]. The amplified PCR amplicons (544 bp) were visualized by gel electrophoresis. Reactions with DNase-free water instead of bacterial DNA were used as a negative control, while DNA extracted from E. coli DH5α was used as a positive control. Similarly, the PCR detection of a specific fragment of femB was used to confirm the identity of the S. aureus isolates [66]. However, the PCR program consisted of an initial denaturation at 95 • C for 4 min, followed by 30 cycles of denaturation at 94 • C for 1 min, annealing at 50 • C for 45 s, and extension 72 • C for 3 min, which was followed by a final extension at 72 • C for 10 min. S. aureus ATCC ® 6538 was used as a positive control.

Statistical Analysis
A simple linear regression analysis was conducted using SPSS 23.0 (IBM ® SPSS ® Statistics, NY, USA) to evaluate the correlation between the bacterial densities in the Akkawi cheese samples. Statistical tests were two-sided, with a type I error set at α = 0.05.

Conclusions
This study highlights the frequent occurrence of a fecal indicator, E. coli, and S. aureus in Akkawi cheese collected from retail stores across Beirut, Lebanon.
Given the popularity of Akkawi cheese among the Lebanese population, the high contamination with these bacterial indicators raises public health concerns and highlights the need for robust food safety systems. Subsequently, microbial pathogens such as L. monocytogenes, which have been previously associated with cheese samples [14], must be fully investigated in order to identify and mitigate the potential sources of contamination. Furthermore, the occurrence of MDR E. coli in the cheese samples emphasizes a paramount need to strengthen antimicrobial stewardship, revise agricultural practices, and adopt policies to curb the spread of antibiotic resistance. The latter is not only a local problem, because AMR can spread across national borders, causing a wide threat to multiple stakeholders. Consequently, efforts must focus on decreasing the over reliance on antibiotics in both agriculture and medicine to preserve the efficacy of these highly important dugs, while seeking viable and sustainable alternatives that do not promote the spread of resistance.