Prevalence of Escherichia coli O157: H7 in foods in the MENA region between years 2000 and 2022: a review

Abstract

Escherichia coli O157:H7 is a pathogen associated with severe diarrhoeal diseases that lead to hemorrhagic colitis, hemolytic uremic syndrome, and death, especially in immunocompromised patients. The source of this bacterium is mainly beef cattle in which the pathogen seeps into the food chain from improper hygienic practices either during the slaughtering of the animal or during the after-processing. In addition, this pathogen gets introduced into other parts of the food chain such as vegetables and fruits, as well as dairy products. This document presents the first review of articles published on the prevalence of E. coli O157:H7 in foods in the MENA region in the last 20 years. Indeed E. coli O157:H7 was isolated from many foods of plant and animal origin. The presence of E. coli O157:H7 in meat products necessities good cooking procedures for these products to eliminate the pathogen and prevent foodborne illnesses. In addition, the presence of the pathogen in fruits and vegetables which are consumed mainly without cooking indicates a high risk of E. coli O157:H7 foodborne outbreaks. Thus, estimating the prevalence of this pathogen in these types of food will certainly help health authorities impose good agricultural and manufacturing practices to minimize the risk of such outbreaks. In this review, Google scholar and PubMed were used as the main search engines for our literature search using the following key words: Prevalence—E. coli O157:H7—Food—MENA—Meat—Chicken—Dairy products—Beef etc. and the names of the countries included in the MENA region were used as keywords.

Keywords:

• Chicken
• contamination
• dairy products
• E. coli
• fruits
• meat
• MENA
• milk
• O157:H7
• prevalence
• vegetables

Introduction

Escherichia coli is a Gram-negative facultative anaerobe, colonizing the gastrointestinal tract of humans and mammals (Jang et al., 2017). Majority of E. coli strains are harmless and are considered as intestinal normal flora, while other strains are considered as pathogenic, such as enterotoxigenic E. coli (EPEC), diffusely adhering E. coli (DAEC), enteroinvasive E. coli (EIEC), enterotoxigenic E. coli (ETEC), enteroaggregative E. coli (EAggEC), and enterohaemorrhagic E. coli (EHEC) (Amrieh, Hamze, Mallat, Achkar, & Dabboussi, 2014).

One of the pathogenic E. coli serotypes of concern in food and food animals is the E. coli O157:H7, which belongs to the EHEC group (Amrieh et al., 2014). It is a major cause of multiple human foodborne diseases, such as haemorrhagic diarrheal disease, which begins as watery diarrhea and could develop into bloody diarrhea (Shebib, Abdul Ghani, & Mahdi, 2003). If it remains untreated, this disease will manifest into haemolytic uremic syndrome (HUS) in children between 1 and 10 years old, and could eventually lead to thrombotic thrombocytopenic purpura (Novotna et al., 2005). Moreover, E. coli O157:H7 is known as verotoxin-producing E. coli because of its ability to produce toxins that have a cytopathic effect on mammalian Vero cells. It is also called Shiga-toxin producing E. coli (STEC), as it produces toxins that resemble the Shiga toxin (Stx) produced by Shigella dysenteriae 1, which are of two types, Stx1 and Stx2 (Navidinia et al., 2012).

Stx1 and Stx2 are considered the main virulence factors of E. coli O157:H7, with Stx2 being more toxic than Stx1 and more associated with HUS in human infections (Lim, Yoon, & Hovde, 2010). Other virulence factors, such as intimin (EaeA) and the plasmid-encoded enterohemolysin (EHIyA), plays a key role in the bacteria’s pathogenicity (Slanec, Fruth, Creuzburg, & Schmidt, 2009). Shiga-like toxins of E. coli can induce death in eukaryotic cells through the inhibition of protein synthesis (Lahti et al., 2001; Rey et al., 2006). Intimin protein encoded by the eaeA gene forms attaching and effacing lesions that helps E. coli O157:H7 induce actin rearrangement under the site of bacterial attachment (Frankel et al., 1998). Enterohemolysin acts on lysing erythrocytes, releasing iron that enables the bacterium to survive and flourish inside the intestine (Fu, Rogelj, & Kieft, 2005; Paton & Paton, 1998; Rey et al., 2006).

E. coli O157:H7 generally can be detected by several molecular, biochemical and immunological methods. One of the used conventional methods is based on the inability of E. coli O157:H7 to ferment sorbitol, leading to the formation of colourless colonies on sorbitol MacConkey agar (SMAC). The latex agglutination method can also be used for O157:H7 serotype detection using O157 and H7 antisera, which is considered simple and easy to use. The polymerase chain reaction is another important technique used for bacterial identification due to its relative accuracy and reproducibility. Amplifying the rfbE and fliC genes that code for O antigen and H7 E. coli flagellar protein, respectively, can also be used to identify the O157:H7 serotype (Deisingh & Thompson, 2004).

Moreover, a multiplex PCR reaction using primers amplifying regions of five genes (stx1, stx2, eaeA, fliC, and rfbE) is used to distinguish E. coli O157:H7 from the other serotypes, and the amplification of the stx2 gene alone helps in the detection of this serotype. In addition to the previous methods, immunomagnetic separation (IMS) using specific antibodies is considered a rapid, accurate, and reproducible method for isolation and identification of E. coli O157:H7 from different types of samples (Deisingh & Thompson, 2004).

Young cattle are considered a primary reservoir of E. coli O157:H7 as well as serving the source and considered asymptomatic carriers (Lim et al., 2010). This pathogen is also found in goats, sheep, cats, horses, rats, dogs and pigeons (Ghanbarpour, Sami, Salehi, & Ouromiei, 2011; Silva, Nicoli, Nascimento, & Diniz, 2009; Wallace, Cheasty, & Jones, 1997; Wani, Samanta, Bhat, & Nishikawa, 2004). Surveys conducted in the states of Washington and Wisconsin in the US, and in Ontario, Canada, revealed a higher prevalence of E. coli O157:H7 in calves than in adult cattle (Armstrong, Hollingsworth, & Morris, 1996; Wells et al., 1991).

E. coli O157:H7 was first identified as a human pathogen in 1982 (Lim et al., 2010). Cattles’s meat and milk are considered vehicles for E. coli O157:H7 transmission to humans. Consequently, consumption of food contaminated with cattle faeces during carcase evisceration have been associated with most of foodborne outbreaks caused by E. coli O157:H7 (Öksüz, Arici, Kurultay, & Gümüs, 2004).

The objective of this review, therefore, was to study the prevalence of E. coli O157:H7 in food in the MENA region between 2000 and 2022.

Methodology

This review presents a collection of studies on the prevalence of E. coli O157:H7 in foods in the MENA region between the years 2000 and 2022. Google scholar and PubMed were used as the main search engines for our literature search using the following keywords: Prevalence, E. coli O157:H7, Food, MENA, Meat, Chicken, Dairy products, Beef, Milk, Fruits, Vegetables, Jordan, Iraq, Egypt, Palestine, Syria, Saudi Arabia, UAE, Iran, Algeria, Turkey, Qatar and Libya etc.

Prevalence of E. coli O157: H7 in beef, meat and products

Since cattle act as a main reservoir of E. coli O157:H7, followed by sheep and goats (Abebe, Gugsa, & Ahmed, 2020; De Assis et al., 2021; Heredia & García, 2018), beef and meat products have been investigated for the prevalence rates of this pathogenic serotype. Cattle carcass usually become contaminated with the intestinal contents during evisceration and further processing after slaughtering the animal (Kirrella, Deeb, & Abdallah, 2017; Mahmoud, Zaki, & Abd-Elhafeez, 2020; Osaili, Alaboudi, & Rahahlah, 2013; Rahimi, Momtaz, & Nozarpour, 2010; Tahamtan, Hayati, & Mehdi Namavari, 2010). In addition, the cattle shed the pathogen normally with their faeces, which might contaminate vegetables by using the cows dung as natural fertilizers or irrigating vegetables using contaminated water with cattle faeces (Balali, Yar, Afua Dela, & Adjei-Kusi, 2020; Khandaghi, Razavilar, & Barzgari, 2010; Solomon, Yaron, & Matthews, 2002). Therefore, good hygienic practices and measures must be taken into consideration during slaughtering, as well as during post- slaughtering processing, such as storing, cutting, selling, cocking and serving these foods. Table 1 lists the studies that reported the isolation of E. coli O157:H7 from beef and meat products. The following sections will summarize the studies reported from the MENA region on the isolation of this pathogen based on country of the study.

Table 1. Prevalence rates of E. coli O157:H7 in meat and products.

Country	Sample type	Sample source	Prevalence	Reference
Iraq	Raw beef	Different butcher’s shops in Al-Karkh side of Baghdad city	6.7% (3/45)	Khalil and Gomaa (2016)
	Raw lamb		8.9% (4/45)
	Minced beef	Butcher’s shops in Baghdad city	4.4% (2/45)	Ghareeb and Al-Zaag (2009)
	Beef	Beef samples were collected from a slaughterhouse in Basra	0.89% (2/225)	Khudaier et al. (2012)
	Beef	Supermarkets of Baghdad	15.56% (7/45)	Al-Dragy & Baqer (2014)
	Imported beef	Local retails of Mosul city	16% (4/25)	Hassan & Hamad (2012)
	Meat contact surfaces		0% (0/25)
	Beef	Local markets in Wasit province	22.5% (9/40)	Al-Ruaby (2017)
	Chicken meat		7.5% (3/40)
	Minced meat	Areas surrounding Nineveh governorate	24% (12/50)	Hassan et al. (2012)
	Fresh minced beef	Different areas in Basra	5% (27/540)	Hussein et al.
	Meat	Butchers’ shops	4.8% (4/84)	Sheet et al., 2022
Jordan	Shawarma sandwiches	Five different cities in northern Jordan	6% (6/100)	Nimri et al. (2014)
	Ready-to-eat chicken samples	Large and small restaurants in all 12 Jordan governorates	0% (0/550)	Osaili et al. (2014)
	Ready-to-eat beef samples		0% (0/478)
Lebanon	Lahm-bi-ajeen (meat pies) and Shawarma	Five areas: Beirut, Mount Lebanon, north (mainly, Tripoli), south (mainly, Sidon), and east (the Bekaa valley)	8.3% (1/12)	Harakeh et al. (2005)
	Fresh ground and offal beef meat	Purchased as sold in the market in bulk from a local big supermarket chain	0% (0/20)	Yammine and Karam (2020)
Libya	Fresh sausage	Butcher’s shops in four different areas of Tripoli	48% (48/100)	Hamza (2017)
	Cooked spiced beef burger	Restaurants and fast-food places in Tripoli city and surrounding area	5.4% (5/92)	El Shrek et al. (2008)
	Uncooked spiced beef burger		27.1% (16/59)
	Cooked spiced chicken burger	Restaurants and fast-food places in Tripoli city and surrounding area	4.68% (3/64)	El Shrek and Ali (2012)
	Uncooked spiced chicken burger		19.6% (11/56)
Morocco	Ground beef	Markets in Casablanca, Morocco	1.43% (2/140)	Badri et al. (2011)
	Turkey		0% (0/200)
	Sausage		0% (0/120)
	Raw meats	Moroccan hospital catering service	0% (0/99)	Ghita et al. (2020)
	Raw vegetables		0% (0/33)
	Hot meals		0% (0/150)
	Salads		0% (0/54)
	Pastries		0% (0/24)
Qatar	Retail food (Cuts of meat, cheese, and samples of ready-to-eat food)	Collected from different types of produce that were processed and packaged in different sites	7.4% (8/108)	Gomez et al. (2020)
	Beef	Five major retail stores in Doha	10% (4/40)	Peters et al. (2017)
	Chicken		0% (0/21)
	Mutton		6.7% (2/30)
	Seafood		1.6% (1/62)
	Cheese		0% (0/11)
	Ready-to-eat food		5.5% (3/55)
Saudi Arabia	Raw meat	Different abattoirs located in Riyadh	4% (6/150)	Al-Zogibi et al. (2015)
	Beef	Fresh meat samples were collected from the abattoirs and minced meat were obtained from different retail establishment markets and abattoirs	0.93% (5/540)	Hemeg (2018)
	Chicken	Various markets	10% (10/100)	Nagawa et al.
	Raw beef	Abattoirs and markets located in Riyadh	2% (2/100)	Hessain et al. (2015)
	Raw mutton		2.5% (1/40)
	Raw chicken		2.5% (1/40)
	Raw turkey		0% (0/20)
	Ground beef		5% (4/80)
	Beef burger		10% (2/20)
	Beef sausage		0% (0/30)
	Ground chicken		5% (1/20)
	Chicken burger		0% (0/20)
Turkey	Raw minced beef	Retail markets	0% (0/255)	Noveir et al. (2000)
	Uncooked hamburger		0% (0/50)
	Soudjouk		0% (0/101)
	Soudjouk	Shops and markets in Afyon province	0% (0/100)	Sırıken et al. (2006)
	Ground beef	Markets and butchers	0.79% (1/126)	Elmali et al. (2005)
	Frozen raw beef	National companies and consumed in Ağrı	1.4% (1/70)	Ayaz et al. (2016)
	Fermented sausage	Different butcher shops and markets in the Elazig province of eastern Turkey	0% (0/96)	Ozbey et al. (2017)
	Sosis		0% (0/48)
	Salami		0% (0/48)
	Ground beef	Various local retail outlets in Kars city	6% (3/50)	Baran and Gulmez (2000)
	Chicken drumsticks		0% (0/50)
	Fresh ground beef	Butchers and supermarkets located in the Samsun Province in Turkey	0% (0/100)	Çadirci et al. (2010)
	Raw meatball		0% (0/100)
	Beef	Several butchers, markets, charcuteries, and local markets	0% (0/53)	Gökmen et al. (2022)
	Chicken		0% (0/31)
	Rinsing water of whole chicken samples	Butchers and whole chicken rinses	0% (0/75)	Telli et al., 2022
	Beef raw meat ball	–	12.5% (3/24)	Goncuoglu et al., 2021
	Vegetarian raw meat ball		6.9% (10/144)
UAE	Beef meat samples imported from South Africa	Randomly purchased from shopping malls at different districts in UAE	30% (14/47)	Dulaimi and Ibrahim (2018)
Iran	Raw beef’s meat	Selected butcheries in Fars and Khuzestan provinces	1.17% (1/85)	Rahimi et al. (2012)
	Raw camel’s meat		0% (0/50)
	Raw sheep’s meat		0% (0/62)
	Raw goat’s meat		0% (0/60)
	Water buffalo’s meat		0% (0/38)
	Fresh ground beef hamburger	Different meat factories across Khuzestan Province	1% (2/200)	Sheikh et al. (2013)
	Ground beef	Beef markets across Mashhad city, Khorasan province	1% (1/100)	Jamshidi et al. (2008)
	Hamburger	Four randomly selected factories in Isfahan	0.83% (1/120)	Miri et al. (2014)
	Chicken nugget		0% (0/70)
	Chicken meat	Shahrekord abattoir	0% (0/360)	Momtaz et al. (2012)
	Neck meat of camel	Najaf-Abad slaughterhouse, Isfahan	1% (3/282)	Rahimi et al. (2010)
Algeria	Imported frozen beef	North-western Algeria	0.40% (1/251)	Barka and Kihal (2010)
Egypt	Chicken sausages	Purchased from the retail markets in Cairo during 2006	26% (13/50)	Mahgoub and Sitohy (2013)
	Chicken sausages	Purchased from the retail markets in Cairo during 2010	41.11 (37/90)
	Cattle meat	Different Cairo abattoirs	2.78% (3/108)	Nawal et al. (2010)
	Frozen meat	Street vendors, retail markets	0% (0/160)	Ahmad and Shimamoto (2014)
	Fresh meat	Butchers	16.25% (13/80)
	Chicken breasts	Street vendors, retail markets	0% (0/160)
	Chicken legs	Street vendors, retail markets	2.5% (4/160)
	Meat products	Different supermarkets, butchers and retail shops	8.4% (18/215)	Elhadidy and Elkhatib (2015)
	Fresh beef	Different supermarkets and butchers’ shops	0% (0/27)	Sallam et al. (2013)
	Ground beef		13.3% (4/30)
	Beef burger		3.33% (1/30)
	Frozen beef liver	Imported beef collected at random from different supermarkets	4% (2/50)	Kirrella et al. (2017)
	Raw beef	Retail markets in Kafrelsheikh governorate	0% (0/100)	Elmonir et al. (2021)

Iran

Jamshidi et al. (2008) have conducted a study to isolate E. coli O157:H7 from ground beef samples. Among the 100 collected and tested samples, only one sample (1%) was positive for E. coli O157:H7. Similarly, Sheikh et al. (2013) reported E. coli O157:H7 contamination of only 1% of the ground beef samples tested in their study. Another study investigated the prevalence of E. coli O157:H7 in processed meats such as hamburger and chicken nuggets collected from different meat factories. Only one sample was contaminated with this pathogen, originating from a hamburger sample (Miri et al., 2014). The isolated O157:H7 strain was positive for stx1, stx2, eaeA and ehxA (encodes for enterohemolysin) genes. Momtaz, Rahimi, and Moshkelani (2012) investigated chicken meat for the contamination of the E. coli O157:H7 serotype, but this pathogen was not identified among other recovered E. coli isolates. Further, the presence of E. coli O157:H7 was investigated in 295 ruminant meat and camel meat collected from butcheries and retail markets (Rahimi, Khamesipour, Yazdi, & Momtaz, 2012). Out of 295 samples collected, 14 were positive for E. coli O157 serogroup, but only one isolate from retail beef meat (7.14%) was identified as E. coli O157:H7 serotype. In this study, camel meats were not contaminated by E. coli O157:H7, indicating that camel meat may not act as a potential source of E. coli O157:H7 infection (Rahimi et al., 2012). Lastly, camel neck meat was collected from a meat processing plant at three different stages: pre-evisceration, post-evisceration, and post-washing. The study showed that 1% of the collected neck meat samples were contaminated with E. coli O157:H7 during post-evisceration and the post-washing. This highlights the importance of monitoring programmes and routine inspection to prevent foodborne outbreaks, as contamination might occur during final steps of meat handling (Rahimi et al., 2010).

Turkey

The prevalence of E. coli O157:H7 in ground beef was investigated in Turkey. In 2010, Çadirci et al. have isolated 5 E. coli O157 strains from ground beef samples, however, none of these strains belonged to the E. coli O157:H7 serotype (Çadirci et al., 2010). In addition, Elmali et al. (2005) isolated only one O157:H7 strain from the 126 tested ground beef samples (Elmali et al., 2005). In contrast, Baran and Gulmez (2000) detected the presence of E. coli O157:H7 in 3 ground beef samples out of 50 samples tested. The same study reported the absence of E. coli O157:H7 contamination in 50 samples of drum stick chicken. In the year 2000, Noveir et al. have analysed different types of meat products, including raw minced meat, Soudjouk (traditional fermented meat product) and uncooked burgers. None of these products was contaminated with E. coli O157:H7 (Noveir, Dogan, & Kadir Halkman, 2000). Due to its popularity and the fact that shiga toxin-producing E. coli (STEC) can survive in fermented foods, Soudjouk was also investigated in a study conducted by Sırıken, Pamuk, Ozakın, Gedikoglu, and Eyigör (2006), and E. coli O157:H7 was not detected in any of the products analysed, which might be due to high acidity of these products and the antibacterial effect of the added garlic. Fermented sausages alongside sosis and salami were also examined for the presence of E. coli O157:H7 by Ozbey, Ozbey, and Kok (2017), and none of these food products harboured this serotype. In addition, frozen raw beef was investigated for contamination with E. coli O157:H7and only one isolate was recovered from 70 tested samples. This isolate harboured the stx1, stx2, eaeA and hly genes (Ayaz, Cufaoglu, Ormeci, & Oz, 2016). Cig kofte (CK; raw meat balls) that are commonly consumed in Turkey, the Middle Eastern countries and other parts of Asia were investigated for the prevalence of E. coli O157:H7 by Goncuoglu, Ayaz, Cufaoglu, and Cengiz (2021). The tested samples included 24 beef and 144 vegetarian CK. Out of 168 samples, the O157:H7 serotype was recovered from 3 beef samples (12.5%) and 10 vegetarian samples (6.9%). All isolates tested positive for the presence of the following virulence genes: eaeA, hly, fliCh7, espA, lpfA1-3. The stx1 and stx2 genes were present in all 3 isolates recovered from the beef CK, whereas vegetarian isolates, harboured at least one of the stx genes. Additionally, Gökmen et al. (2022) investigated a total of 294 samples (53 beef, 31 chicken, 100 raw milk, 35 white cheese, 22 lettuce, 34 parsley and 19 spinach samples) collected from several butchers, markets, charcuteries, and local markets to identify the contamination rate with E. coli O157:H7. None of the collected samples was positive for this serotype, however, other serotypes like O111 and O145 were detected in this study (Gökmen et al., 2022). Moreover, Telli et al. (2022) investigated the rate of contamination in the rinsing water of whole chicken samples with E. coli O157:H7. A total of 75 chicken samples were collected from butchers and whole chicken rinses were tested for the presence of E. coli O157:H7. It was found that 56 chicken samples were positive for E. coli, however none of the isolated strains were of the O157:H7 serotype (Telli et al., 2022).

Algeria

In the period from 2000 to 2022, Barka and Kihal (2010) have investigated the prevalence of E. coli O157:H7 in frozen meat imported from Argentina, Brazil, Uruguay, Ireland, New-Zealand, and Australia. Only one out of the 251 frozen samples examined was contaminated with this pathogen. This contaminated sample was imported from Uruguay. There were no other published studies from Algeria that document the presence of O157:H7 in food products.

Egypt

Due to the nature of the handling processes of meat and its products, there is always a high probability of cross contamination of these products with gut microbes during evisceration and subsequent processes. This problem is exacerbated by the low personal hygienic practices. Egypt is one of the largest countries in the MENA region and several studies from this country reported the contamination of meat products with E. coli O157:H7. In one study, a total of 50 chicken sausage samples were purchased from retail markets in Cairo in 2006 and another 90 samples were purchased during the year 2010. These samples were tested for the presence of E. coli O157:H7 after their production and 1 month before their expiry date. E. coli O157:H7 was detected in 26% (13/50) and 41.11% (37/90) of the samples collected during 2006 and 2010, respectively (Mahgoub & Sitohy, 2013). Their method of confirmation was based on calculating CFUs on selective media without including molecular or biochemical methods to confirm their high percentages, therefore their percentages seem to be not accurate. Another study examined buffalo and cow meat samples for contamination with O157:H7 serotype. A total of 108 meat samples (20 buffaloes and 88 cows) were collected, of which 3 samples (2.78%) were contaminated with E. coli O157:H7 (Nawal, Mohey, Raafat, Hassan, & Ashraf, 2010). Moreover, a study was conducted in 2014 using 800 meat samples and other 800 dairy and products samples collected from different cities and villages. E. coli O157:H7 was detected in 3.4% of all samples (54/1600) with slight increase in the incidence in various dairy products except the yoghurt samples, which could be attributed to the low pH of the yoghurt. The O157:H7 isolates obtained in the study were positive for stx1 and stx2 genes. Additionally, Elhadidy and Elkhatib (2015) have detected this pathogen in 8.4% (18/215) and 6.9% (15/215) of the collected meat products and dairy products, respectively. In meat products, the percentage of samples harbouring E. coli O157:H7 was the highest in ground beef samples (12%), followed by burgers (7.1%), and fresh beef (5.7%), while in the dairy samples, the highest percentage belonged to samples of cheese that was made from raw milk (7.8%), followed by raw milk samples (6%). Raw beef was also investigated by Elmonir et al. (2021) who collected and tested 100 samples, but the samples were devoid of any E. coli O157:H7 serotype. Moreover, fresh, ground or burger beef samples were tested by Sallam, Mohammed, Ahdy, and Tamura (2013). E. coli O157:H7 was detected in 4 out of 30 samples (13.3%) of ground beef, while only one burger sample (3.33%) tested positive for the E. coli O157:H7. However, all the tested unprocessed fresh beef samples (n = 27) were free of the O157:H7 serotype. These results imply that ground beef and beef burger samples might became contaminated during the production process. Out of the recovered five O157:H7 isolates, only one contained all the 4 virulence genes (stx1, stx2, eaeA, hlyA) (Sallam et al., 2013).

Furthermore, one study examined frozen beef liver samples for the presence of E. coli O157:H7. Of the collected 50 samples, only 2 samples (4%) were contaminated, indicating a possible contamination of the liver meat during the evisceration of the beef cattle. Thus, cooking the frozen liver very well is necessary to prevent this pathogen from causing infection (Kirrella et al., 2017).

Iraq

Several researchers from Iraq have examined the presence of E. coli O157:H7 in food and food products. Khalil and Gomaa (2016) has studied the prevalence rate of E. coli O157:H7 in fresh raw beef and lamb meat. Out of 90 collected samples (45 samples of each of raw beef and raw lamb), 6.7% of raw beef and 8.9% of lamb meat were contaminated with this particular serotype. In addition, this pathogen was detected in two beef samples out of 225 (0.89%) obtained from a slaughterhouse in Basra city (Khudaier, Abbas, & Khleel, 2012), while it was detected in 7 out of 45 samples (15.56%) of imported beef collected from Baghdad supermarkets (Al-Dragy & Baqer, 2014). Interestingly, imported frozen samples appeared to be heavily contaminated with this pathogen implying a loose restrictions on importing and storing meat products.

Furthermore, E. coli O157:H7 prevalence was investigated in minced meat collected from different local markets and butcher shops in Baghdad. This pathogen was detected in two out for 45 samples (4.4%) of minced beef collected from different butcher shops, while it was isolated from 12 out of 50 samples of imported minced meat collected from local markets (Ghareeb & Al-Zaag, 2009; Hassan, Abdulla, & Al-Dabbagh, 2012). Beef and chicken meat were also tested for the prevalence of E. coli O157:H7 by Al-Ruaby (2017), and again the contamination of beef meat with E. coli O157:H7 (9 out of 40) was significantly higher than that in chicken meat samples (3 out of 40), which is not uncommon as beef cattle is the reservoir for this pathogen. Moreover, the stx2 gene was present in 7 out of the 9 isolates (77.78%) recovered from beef samples but was absent in the chicken meat isolates. Furthermore, only 2 out of 9 isolates (22.22%) from beef carried the eaeA, while the gene was present in all the obtained chicken isolates. Lastly, Hassan and Hamad (2012) have investigated imported beef and their immediate contact surfaces for the contamination with E. coli O157:H7. This pathogen was detected in 4 out of 25 (16%) samples, whereas none of the 25 tested meat contact surfaces were contaminated with this pathogen, indicating that the pathogen originated in the meat and not a contamination.

In a very recent study, Sheet, Othman, and Alsanjary (2022), tested 504 samples from different surface swabs and meat samples collected from local markets. Out of the 504 tested samples, 13 only were found contaminated with E. coli O157:H7 including both meat and surfaces samples. All E. coli O157:H7 harbored uidA, stx1 and stx2, except one that didn’t carry the stx1 gene (Sheet et al., 2022).

Jordan

Shawarma or gyros (wrap of shredded meat; beef, lamb or chicken) is one of the most popular foods in Jordan, and is highly consumed by Jordanians. Nimri, Al-Dahab, and Batchoun (2014) described the predominance of E. coli contamination in both chicken (33/80) and beef (8/20) meats taken from shawarma sandwiches. Generally, the meat in these sandwiches is marinated, grilled and served with mayonnaise, pickles or tahini dip. Six out of the 41 (14.5%) isolated E. coli belonged to the O157:H7 serotype, with these isolates originating from the chicken shawarma sandwiches solely. This result is somewhat strange as chicken is not a common reservoir for this pathogen, however, it is likely the chicken meat was contaminated during the preparation due to the poor hygienic practices of the food handlers. In addition, the same store normally grills meat and chicken shawarma in a confined place leading to cross contamination. When tested for toxin genes, the isolates were positive for the stx1 gene (Nimri et al., 2014). Osaili et al. (2014) studied the prevalence of E. coli O157:H7 in ready to eat meat products including chicken or beef shawarma, beef pastry, beef kubba, meat kebab, beef in pita bread, roasted chicken, and burgers from chicken or beef that were collected from different restaurants. Although 1028 samples were collected and tested, none of them was contaminated with this pathogen, which might indicate that all collected meat samples were cooked at a temperature sufficient for eradicating this pathogen. Alternatively, the isolation protocol might not have been optimized to detect this pathogen.

Lebanon

In Lebanon, Shawarma sandwiches and meat pies were tested for the presence of E. coli O157:H7 by Harakeh et al. (2005). Meat pie samples were free of E. coli contamination, whereas 7 out of 12 shawarma samples (58%) were contaminated with E. coli. However, only one isolate was suspected to be E. coli O157:H7 and needed further confirmation. In another study, offal beef meat and fresh ground beef (local and imported) were investigated by Yammine and Karam (2020), but none of the tested beef samples tested positive for E. coli O157:H7 isolates.

Libya

In Libya, Hamza (2017) investigated the prevalence E. coli O157:H7 in 100 fresh sausages and spiced burger meat samples. Surprisingly, 48% (48/100) of the total samples were contaminated with this pathogen. The numbers of isolates E. coli O157:H7 is exceptionally high which could be due the low quality of beef that was used to fill the sausages or make the burgers. Further, it appears that high contamination might be also a reflection of the unhygienic practices during the production of these sausages. Nonetheless, the researchers have not used molecular techniques for further confirmation of the isolates identity, thus the above numbers are merely considered as presumptive E. coli O157:H7.

Two other groups have investigated the contamination of cooked and uncooked spiced beef and chicken burger meats with the O157:H7 serotype. El Shrek, Madi, El Bakoush, and El Tawil (2008) found that uncooked beef burger meats were highly contaminated with this pathogen, as 16 out of 59 samples (27.1%) tested positive for this serotype. On the other hand, only 5 samples out of the 92 tested cooked samples (5.4%) contained E. coli O157:H7. In a similar study, El Shrek and Ali (2012) examined cooked and uncooked spiced chicken burger meats for the presence of E. coli O157:H7. The authors reported that 11 out of 56 (19.6%) and 3 out of 64 (4.68%) of the uncooked and the cooked meat respectively, were found contaminated with E. coli O157:H7. As noted earlier, chicken is not considered a reservoir for this pathogen, thus the numbers of positive samples remain low signifying improper hygienic practices during handling these samples.

Morocco

Casablanca, a large Moroccan city has been associated with high numbers of food borne diseases and outbreaks originating from meat and meat products. This might be due to the touristic nature of the city reflecting high numbers of restaurants serving meals for large number of people in short times. Meat products such as ground beef, turkey and sausages were investigated for the presence of E. coli O157:H7 by Badri, Fassouane, Filliol, Hassar, and Cohen (2011). Surprisingly, they have only detected O157:H7 serotype in two out of 140 tested ground beef samples (1.43%), whereas sausages and turkey meat samples were free of E. coli O157:H7. The obtained O157:H7 isolates harbored stx1, stx2, eaeA and hylA virulence genes. Moreover, a study conducted by Ghita, Bennani, Berrada, Benboubker, and Bennani (2020) to look for E. coli O157:H7 in meals prepared to feed patients and medical staff. The types of food tested included raw meats, raw vegetables, hot meals, salads and pastries. All types were free of the O157:H7 serotype. However, non-O157 E. coli strains were isolated from 9 out of 300 tested samples. Furthermore, 91 samples of four different preparations of turkey meat (mild and spicy minced meat, mild and spicy sausages) were investigated for E. coli O157:H7, as it is being thought as poultry meat is main vehicle of foodborne diseases worldwide. A total of 70 E. coli strains were isolated that correspond to 70 different samples but only 2 of them (spicy sausages and mild minced meat) showed agglutination with the O157 and H7 anti-sera (El-Sharawy, Al-Zahrani, & El-Waseif, 2022).

Qatar

Various retail foods including beef, chicken, lamb, goat meat, camel meat, several seafood, cheese, and salads were tested for the presence of E. coli O157:H7 by Gómez et al. (2020). E. coli O157:H7 was detected in 8 out of 108 samples (7.4%) of these foods. The specific origins of these isolates were not mentioned in the study as they might be imported from different sources. Likewise, Peters et al. (2017) studied the prevalence of this pathogen in retail products including beef, chicken, mutton, seafood and cheese. E. coli O157:H7 was present in 4 out of 40 (10%) beef samples, and in 2 out of 30 (6.7%) of the mutton samples. In the same study, out of the 55 tested ready to eat food samples, only 3 samples (5.5%) were contaminated with O157:H7. In addition, only one out of the 62 tested sea food samples (1.6%) contained E. coli O157:H7 while both the tested 21 chicken samples and the 11 cheese samples were free of this pathogen.

Saudi Arabia

Different types of meats were investigated for the presence of E. coli O157:H7 by a number of research groups. Al-Zogibi, Mohamed, Hessain, El-Jakee, and Kabli (2015) have isolated six E. coli O157:H7 isolates out of 150 samples (4%) from raw meat, and they were all positive for stx2 and eaeA genes. In another study conducted by Hessain et al. (2015), the prevalence of E. coli O157:H7 was investigated in raw meat samples, including beef, mutton, chicken, turkey, ground beef, beef burgers, beef sausages, ground chicken and chicken burgers. Among all samples, 10% (2 out of 20) of the beef burgers contained E. coli O157:H7, followed by ground beef (5%; 4/80), ground chicken (5%; 1/20), raw mutton (2.5%; 1/40), raw chicken (2.5%; 1/40) and raw beef (2%; 2/100). The recovered isolates were screened for the presence of four virulence genes; stx1, stx2, eaeA and hlyA. In general, the isolates harbored some of these genes, but none of the isolates contained all the four virulence genes. Additionally, El-Sharawy et al. (2022) collected a total of 100 chicken samples from various markets to identify the prevalence rate of E. coli O157:H7. Almost half of the collected samples were contaminated with E. coli (40%). To investigate the specific contamination with E. coli O157:H7, the 20 isolated E. coli strains were serotyped by agglutination test and 50% of the E. coli strains were of the O157:H7 serotype (El-Sharawy et al., 2022).

Fresh beef and ground meat samples were also investigated for the presence of the E. coli O157:H7 serotype by Hemeg (2018). Different E. coli serotypes were detected, of which 5 isolates out of the 120 tested samples (4.2%) were identified as E. coli O157:H7. When screened for stx2 and eaeA genes, all the 5 isolates were positive.

UAE

Only one study conducted by Dulaimi and Ibrahim (2018) in UAE aimed at detecting E. coli O157:H7 in 47 South African imported meat samples. Interestingly, this pathogenic serotype was detected in 14 out of the 47 of the samples (30%), with all of them being positive for the stx2 gene. However, none of the other virulence genes were screened for in this study.

Table 1 summarizes the reports on the incidence of E. coli O157:H7 in meat and products published from the year 2000 till the time of preparation of this review.

Prevalence of E. coli O157: H7 in milk and dairy products

Milk is a product of dairy cattle that might acquire E. coli O157:H7 due to faecal contamination during the milking process or post-milking processing, as this pathogen is not a pathogen that might be found in the milk itself. Contaminated milk, therefore, could serve as a vehicle for transmitting E. coli O157:H7 serotype, especially if inefficient heat treatment was implemented during the processing of the milk. For instance, faecal contamination in the area where milking takes place is a potential problem that is amplified by poor hygienic practices. Table 2 lists the studies included in the milk and dairy products section and the prevalence rates of E. coli O157:H7. In the following sections the review will address research conducted in countries of the MENA region on the prevalence of E. coli O157:H7 in milk and its products.

Table 2. Prevalence rates of E. coli O157:H7 in milk and dairy products.

Country	Sample type	Sample source	Prevalence	Reference
Iraq	Cow’s soft cheese	Collected from retail markets in different locations of Babylon province in summer	20% (4/20)	Najim and Al-Isawi (2017)
	Buffalo’s soft cheese		10% (2/20)
	Cow’s soft cheese	Collected from retail markets in different locations of Babylon province in winter	25% (5/20)
	Buffalo’s soft cheese		15% (3/20)
	Raw milk	Different parts in Basra	0.44% (1/225)	Khudaier et al. (2012)
	Raw milk	10 dairy farms	10% (15/150)	Najd Obaid et al., 2022
	Soft cheese	Five popular markets in the city of Baquba	50% (30/60)	Al-Shedidi (2017)
	Bovine soft cheese	AL-Rasaffa districts	31.11% (14/45)	Khudhir (2017)
	Ovine soft cheese		37.77% (17/45)
	Ovine soft cheese	AL-Karkh district of Baghdad city	36.59% (15/41)	Najim (2017)
	Bovine soft cheese		31.37% (16/51)
	Raw milk	Different villages surrounding Baghdad province	51.54% (18/35)	Najim (2014)
	Raw milk	Local markets in Wasit province	17.5% (7/40)	Al-Ruaby (2017)
	Raw milk	Three different markets in Basra city	8.67% (13/150)	Abbas et al. (2012)
	Soft cheese	Three different markets in Basra city	10% (15/150)	Abbas and Khudor (2013)
	Raw cow’s milk	Different retail markets in Baghdad province	28% (7/25)	Khudhir (2013)
	Soft cheese		16% (4/25)
Libya	Fresh white cheese	Small factories in 7 areas around Tripoli, Libya	35.6% (31/87)	Abujnah and Magdoli (2016)
	Open ice cream and packed samples	Open ice cream samples were obtained from six small producers and packed samples were bought from two large manufactures located in different areas of Tripoli	1.33% (2/150)	El-Sharef et al. (2006)
Saudi Arabia	Milk	Five dairy farms were collected from different abattoirs located in Riyadh	4.81% (26/540)	Al-Zogibi et al. (2015)
Syria	Raw-unpasteurized bovine milk	Local markets in Lattakia city	30.9% (34/110)	Daood (2007)
	Hard cheese		77.41% (48/62)
	Sweet cheese		19% (8/42)
	Cream cheese		18% (9/50)
	Cream		10% (3/30)
Turkey	Ice cream	Retail markets	0% (0/73)	Yaman et al. (2006)
	Raw milk	Several butchers, markets, charcuteries, and local markets	0% (0/100)	Gökmen et al. (2022)
	White cheese		0% (0/35)
	Raw cow’s milk	Markets in Van city	0% (0/100)	Sancak et al. (2015)
	Herby cheese		2% (2/100)
	Van otlu cheese	Randomly collected from retail outlets	0% (0/50)	Tekinşen and Özdemir (2006)
Iran	Water buffalo’s milk	Obtained from Isfahan, Chaharmahal va Bakhtiari, Yazd and Khuzestan provinces	0.92% (1/109)	Rahimi et al. (2012)
	Camel’s milk		0% (0/87)
	Bovine milk		1.04% (2/192)
	Ovine milk		0% (0/138)
	Caprine milk		0% (0/208)
	Bulk tank milk	Various cow farms in Kermanshah	0% (0/206)	Mohammadi et al. (2013)
	Bovine subclinical mastitis milk	Kurdistan Province	0.75% (3/400)	Ahmadi et al. (2020)
	Cheese	Different supermarkets and retailer shops from Isfahan, Chaharmahal and Bakhtyari and Khuzestan provinces	0% (0/120)	Rahimi et al. (2011)
	Ice cream		0% (0/120)
	Yogurt		0% (0/50)
Egypt	Kareish cheese	Groceries, small markets, and supermarkets in Sohag city, Egypt	24% (12/50)	Hassanien and Shaker (2020)
	Labena		6% (3/50)
	Yoghurt		4% (2/50)
	Raw milk and raw milk dairy products	Different localities in Egypt	6.9% (15/215)	Elhadidy and Elkhatib (2015)
	Market raw milk	Collected from various supermarkets, shops and local groceries	4% (1/25)	El-afify et al. (2020)
	Bulk tank milk		0% (0/25)
	White soft cheese		0% (0/25)
	Kareish cheese		8% (2/25)
	Ice cream		0% (0/25)
	Raw Buffalo milk	Private farms, animal owners, street vendors, retail markets	6.67% (16/240)	Ahmad and Shimamoto (2014)
	Raw Bovine milk	Private farms, animal owners, street vendors, retail markets	1.67% (4/240)
	Kareish cheese	Street vendors	6.67% (8/120)
	Yogurt	Retail markets	0% (0/80)
	Domiati cheese	Retail markets	0.83% (1/120)
	Raw milk	Collected from different localities	0% (0/30)	Saad (2012)
	Damietta cheese		3.33% (1/30)
	Kareish cheese		6.67% (2/30)
	Milk	Collected from 180 She-camels	0% (0/720)	Diab et al. (2021)
	Cattle milk	Smallholder	0% (0/30)	Gwida et al. (2021)
	Cattle milk	Farm	0% (0/60)
	Milk	Retail markets in Kafrelsheikh governorate	0% (0/100)	Elmonir et al. (2021)

Iraq

Homemade and commercial soft cheese samples were collected and analysed to estimate the prevalence of E. coli O157:H7. These samples were collected in two different seasons, winter and summer. E. coli O157:H7 was detected in 20% and 10% of soft cheese originating from cows (n = 20) and buffaloes (n = 20), respectively, in the summer season. In winter, this pathogen was detected in 25% and 15% in cow (n = 20) and buffalo (n = 20) cheese samples, respectively. Results shows a seasonal variation in the prevalence of this pathogen, as it was more prevalent in winter than in summer (Najim & Al-Isawi, 2017). Isolates obtained in the summer season were screened for the presence of stx1 and stx2 genes, and 5 out of 6 isolates were positive for stx1, while only one isolate harboured the stx2 gene. Similarly, Al-Shedidi (2017) reported that 30 out of 60 (50%) collected local soft cheese samples were contaminated with E. coli O157:H7, and in contrast to results obtained by Najim and Al-Isawi (2017), the prevalence of this pathogen was higher in the summer (56.67%; n = 30) than in the winter (43.33%; n = 30). From the above two results, there might not be a ground for the seasonal variation of the E. coli distribution, it might be rather a variation in procedures of collection and testing. Furthermore, Khudhir (2017) and Najim (2017) have studied the prevalence of the E. coli O157:H7 serotype in soft cheese samples originating from cows and sheep. In both studies, this pathogen was highly prevalent in sheep’s soft cheese samples than the cow’s soft cheese samples; 37.77% (17/45) and 36.59% (15/41) for sheep’s soft cheese and 31.11% (14/45) and 31.37% (16/51) for cow’s soft cheese (Khudhir, 2017; Najim, 2017).

Additionally, Abbas and Khudor (2013) have tested 150 soft cheese samples for the presence of E. coli and found that 15 samples (10%) harboured E. coli O157:H7, while Khudhir (2013) found 4 out of 25 tested soft cheese samples (16%) positive for E. coli O157:H7. All the 15 isolates identified by Abbas and Khudor (2013) harboured the verotoxin producing genes, vt1 and vt2. Moreover, the incidence of E. coli O157:H7 was assessed in raw milk samples. Abbas, Khudor, and Abid Smeasem (2012) reported the presence of E. coli in 86 out of 150 tested raw milk samples, of which 13 isolates (8.67%) were identified as the O157:H7 serotype. These isolates were screened for vt1 and vt2 genes, and 7 of the 13 isolates (53.85%) were identified as VTEC E. coli O157:H7. In another study, 150 milk samples from 10 dairy farms were collected and tested for the prevalence of E. coli O157:H7 in dairy cattle. Fifteen out of 150 samples were contaminated with E. coli O157:H7. Furthermore, all obtained strains were screened for stx1 and stx2 genes and none of them harboured these virulence factors (Najd Obaid et al., 2022).

Furthermore, raw milk samples were positive for E. coli O157:H7 in studies conducted by Al-Ruaby (2017), Khudaier et al. (2012), Khudhir (2013) and Najim (2014) with the following percentages: 0.44% (n = 225), 17.5% (n = 40), 28% (n = 25) and 51.54% (n = 35), respectively. This large variation in the incidence of E. coli O157:H7 might be related to the hygienic practices during the milking and the distribution of the milk or simply due to a variation in the isolation and confirmation methods.

Saudi Arabia

In Saudi Arabia, 540 raw milk samples were tested by Al-Zogibi et al. (2015) for the presence of E. coli O157:H7. Different E. coli serotypes were isolated in this study, in which 26 (4.81%) isolates belonged to E. coli O157:H7 serotype. Interestingly, when the isolates were screened for stx2 and eaeA, all of them were positive for both virulence genes.

Syria

One study conducted by Daood (2007) investigated the prevalence of E. coli O157:H7 in raw milk and a variety of other dairy products. Among all samples, hard cheese was the most contaminated (77.41%; n = 62), followed by raw milk (30.9%; n = 110), sweet cheese (19%; n = 42), cream cheese (18%; n = 50) and soft cream (10%; n = 30). The high level of contamination of hard cheese suggests inefficient heat treatment and preservation, as well as the improper or complete absence of hygienic conditions and practices during preparation of this type of cheese. Further, this type of cheese is normally processed without any heat treatment and thus if was not properly heat-treated post processing, it might be heavily contaminated with a plethora of microorganisms not only E. coli O157:H7.

Turkey

Sancak, Sancak, and Isleyici (2015) reported that none of the 100 tested raw milk samples were positive for O157:H7 serotype, while 2 out of 100 cheese samples harboured E. coli O157:H7 indicating a proper hygienic practices. In another study conducted by Tekinşen and Özdemir (2006), none of the tested 50 samples of a traditional cheese derived from unpasteurized milk (Van otlu cheese) were positive for E. coli O157:H7. Although the unpasteurized milk was produced under poor hygienic practices in small dairy plants, it was devoid of this pathogen. Indeed, the absence of E coli O157:H7 might indicate that the starter milk was not contaminated with this pathogen, or the testing procedure was not sensitive enough so that no isolates were identified. As a product of milk and a popular product consumed in summer, ice cream was investigated by Yaman, Elmali, Ulukanli, Tuzcu, and Genctav (2006) to test for the presence of pathogens such as E. coli O157:H7. Even though 20.55% of the tested ice cream samples were contaminated with E. coli species, suggesting faecal contamination, none of the samples was positive for O157:H7 serotype.

Iran

A study was conducted by Rahimi et al. (2012) to estimate the prevalence of E. coli O157:H7 in raw milk from various sources, including water buffalo, camel, bovine, ovine and caprine. Only 3 raw milk samples from 109 water buffalo samples and 192 bovine samples were found contaminated with the O157:H7 serotype with percentages of 0.92% and 1.04%, respectively with all the isolates harboured the stx1, stx2, eaeA and hlyA virulence genes. A report by Mohammadi, Abiri, Rezaei, and Salmanzadeh-Ahrabi (2013) showed that a total of 206 raw milk samples collected from bulk tanks were free of this serotype, but still had other STEC species and other serotypes in 17.47% of the samples tested.

Furthermore, 400 milk samples from cattle with mastitis were analysed by Ahmadi, Mardani, and Amiri (2020) to evaluate the association of E. coli O157:H7 serotype with mastitis. These samples were collected and tested during all seasons of the year, with 100 samples collected during each season. Only 2 samples collected in autumn and 1 sample collected in winter were contaminated with E. coli O157:H7 indicating a weak association between mastitis and E. coli O157:H7. When tested for stx1, stx2 and eaeA, two of the three O157: H7 isolates tested positive for the stx1, stx2 and eaeA genes, while the one isolate had the stx1 and stx2 genes but not the eaeA gene. Rahimi, Chaleshtori, and Parsaei (2011) has studied a group of dairy products (120 cheese samples, 120 ice cream samples and 50 yoghurt) to investigate the prevalence of E. coli O157:H7. Out of 290 examined samples, 9 of them contained E. coli O157 strains, but none of these isolates belonged to the O157:H7 serotype. Collectively, the findings of these studies indicate that the dairy industry in Iran implements strict hygienic rules for the handling of milk and its products which is clearly indicated by the very low prevalence of E. coli 157:H7 in the milk and its products.

Egypt

A total of 150 samples of dairy products (50 kareish cheese, 50 labena and 50 yogurt) were collected and analysed by Hassanien and Shaker (2020) for the presence of E. coli O157:H7. Among all collected samples, 24% (12/50) of kareish samples were contaminated, followed by labena (6%; 3/50) and yoghurt (4%; 2/50). The reason why Kareish cheese harboured highest numbers of the pathogens is that it is usually prepared from raw milk in rural homes by traditional methods under improper hygienic conditions. In addition, it has lower acidity than labena and yoghurt, which might explain the difference in rates of contamination between these three dairy products. Similarly, two studies investigated kareish cheese for contamination with E. coli O157:H7. Elafify et al. (2020) reported the contamination in 2 out of 25 Kareish samples (8%), while 2 out of 30 samples (6.67%) examined by Saad (2012) were found to harbour this pathogen. The isolates obtained by Elafify et al. (2020) tested positive for the presence of the stx1, stx2 and eaeA genes. By the same token, Elhadidy and Elkhatib (2015) examined 215 dairy samples comprising of 100 raw milk, 60 Kareish cheese and 55 domiati cheese. The results showed that 15 samples (6.9%) were positive for E. coli O157:H7, with highest prevalence in cheese made from unpasteurized milk (Kareish cheese and domiati cheese) (7.8%), followed by raw milk samples (6%). This small difference in the incidences of E. coli O157:H7 between cheese and raw milk could be due the further processing of cheese which might have increased the chances of contamination compared to raw milk that was tested directly. Recently, Diab et al. (2021) collected 720 milk samples from 180 camels (4 udder quarters per animal), of which 18.9% (35 animals) were diagnosed with mastitis by testing their milk. E. coli was isolated from 25.6% and 19.7% of the milk collected from animals with clinical and subclinical mastitis, respectively. However, different E. coli serotypes were obtained; but none of the recovered isolates belonged to the O157:H7 serotype (Diab et al., 2021). Additionally, Elmonir et al. (2021) collected 100 milk samples from different retail markets and tested them for the prevalence of E. coli O157:H7. None of the samples was contaminated with this pathogen. Gwida, EL Mahmoudy, Elgohary, Mohamed, and Hassan (2021) investigated 90 cattle milk samples (smallholder milk = 30 and farm milk = 60) for the presence of shiga toxin producing E. coli. Out of the 90 samples, 12 (13.3%) were contaminated with E. coli, with 5 from smallholder milk (16.7%) and 7 from farm milk (11.7%). Despite the occurrence of O157 in both types of milk samples, isolates belonging to the O157:H7 serotype were not recovered. In a recent study, Mohamed et al. (2022) tested 267 cattle milk for the presence of Shiga toxin producing E. coli. Out of 86 isolated E. coli, 39 of them were Shiga toxin producing isolates. When tested for drug resistance, the STEC isolates were resistant to 11 out of the 16 antibiotics used in the experiment.

Libya

Exceptionally high numbers of E. coli O157:H7 contamination of cheese samples were reported by a study from Libya. Abujnah and Magdoli (2016) has studied fresh white cheese that are produced from raw unpasteurized milk and reported that 31 out of 87 of the samples (35.6%) were contaminated with E. coli O157:H7. These high numbers might indicate extremely poor hygienic practices employed during the milking process and the manufacturing of this type of cheese especially that milk is not a common reservoir for this pathogen. Further, with this high number of pathogenic E. coli O157:H7, there is certainly a high risk of consuming this type of cheese without prior heat treatment to eliminate this pathogen. In fact, traditionally, this type of cheese is boiled after the curd formation and prior to consumption. Another study conducted in Libya by El-Sharef, Ghenghesh, Abognah, Gnan, and Rahouma (2006) examined the bacteriological quality of a 150 samples of ice cream for the presence of E. coli O157:H7. Indeed, this pathogen was present only in 1.33% of all collected samples, originating from open ice cream samples. This percentage, although very low, certainly signify the dangers of consuming this ready to eat product by children or immunocompromised individuals. Table 2 summarizes the studies conducted on the prevalence of E. coli O157:H7 in milk and dairy products.

Prevalence of E. coli O157: H7 in vegetables and fruits

Despite not being of an animal source, vegetables and fruits were also investigated for the prevalence of E. coli O157:H7, which might be present due to faecal contamination from manured soil or low hygienic practices during handling and storing. Cross-contamination of vegetables with E. coli O157:H7-contaminated meat in big stores where vegetables and butchery departments are located side by side might occur frequently. In addition, irrigating vegetables and some fruits such as strawberry with inefficiently treated wastewater increases the chance that they become contaminated with this E. coli O157:H7. Consequently, these vegetables and some fruits act as potential vehicles of this pathogen, especially some of these vegetables are consumed without being cooked. Table 3 shows the studies included in the vegetables and fruits section and the rates of E. coli O157:H7 contamination.

Table 3. Prevalence rates of E. coli O157:H7 in vegetables and fruits.

Country	Sample type	Sample source	Prevalence	Reference
Palestine	Fresh vegetable salads	School canteens and different restaurants in Mid Zone, Khan Younis and Rafah governorates	1% (2/200)	El-Manama et al. (2017)
Saudi Arabia	Fresh Vegetables	Various local markets and street vendors in Taif city	0% (0/68)	Hassan et al. (2011)
Turkey	Fresh Vegetables and fruits	Supermarkets in Istanbul	0% (0/261)	Büyükünal et al. (2015)
	Lettuce	Several butchers, markets, charcuteries, and local markets	0% (0/22)	Gökmen et al. (2022)
	Parsley		0% (0/34)
	Spinach		0% (0/19)
Iran	Vegetables	27 farm producers in Tabriz city	0.71% (1/141)	Khandaghi et al. (2010)
	Soils		3.55% (5/141)
	Fresh-cut vegetables	Four chain supermarkets	0% (0/32)	Jeddi et al. (2014)
	Ready-to-eat salads		0% (0/20)
	Wheat sprouts		0% (0/32)
	Mung bean sprout		0% (0/32)
	Ready-to-eat salad vegetable	Purchased in the city of Tabriz, in 5 areas categorized as: North (N), South (S), West (W), East (E) and Central (C).	16.66% (10/60)	Kouchakkhani et al. (2016)
	Lettuce	Various farmlands in Tehran	1% (1/100)	Mazaheri et al. (2014)
Iraq	Lettuce	Supermarkets in Baghdad	20% (5/25)	Al-Dragy & Baqer (2014)
Egypt	Fruits	Collected from at least 3 local street markets (in densely populated areas)	1.17% (6/513)	Khalil and Gomaa (2016)
	Vegetables		7.87% (34/432)

The following section discusses the research findings on the prevalence of E. coli O157:H7 in vegetables and fruits in the MENA region.

Palestine

Aiming to assess the microbiological quality of fresh vegetable salads, El-Manama Abdelraouf, Arafa Hashem, and Abu-Owda Samia (2017) have analysed 200 fresh vegetable salads, of which 100 samples collected from school canteens, while the other 100 samples were collected from different restaurants. From both sources, the samples included three categories: single vegetable (n = 11), mixed vegetables (n = 167), and vegetables with additives (lemon, oil and/or vinegar) (n = 22). All three categories were contaminated with E. coli species, but only two samples originated from the mixed vegetables category were contaminated with the O157:H7 serotype.

Saudi Arabia

In search for enteric bacteria in vegetables, Hassan, Altalhi, Gherbawy, and El-Deeb (2011) have investigated 68 fresh vegetable samples (tomato, spinach, lettuce, parsley) purchased from local markets and street vendors for the presence of E. coli O157:H7. All samples were free of E. coli O157:H7 serotype.

Iran

Since the presence of E. coli in food is mainly due to faecal contamination, 141 vegetable samples collected from agriculture farms fertilized with cow manure were investigated for E. coli O157:H7 contamination (Khandaghi et al., 2010). Out of 41 lettuce samples tested, only one was contaminated (2.44%), while, cabbage, radish, sprout and carrots were free from this pathogen. In parallel, 141 soil samples were also collected from the farms where the vegetables were grown. E. coli O157:H7 was found in 3 out of 41 (7.32%) lettuce soil and 2 out of 38 (5.26%) cabbage soil. Furthermore, all obtained isolates harboured the shiga toxin genes, stx1 and stx2, as well as the eaeA gene. Since raw lettuce is used extensively in homemade meals and in restaurants without being heat processed, another group tested 100 lettuce samples for the presence of E. coli O157:H7. Indeed, 8 STEC isolates were identified, with one identified as E. coli O157:H7 (Mazaheri, Salmanzadeh Ahrabi, & Aslani, 2014). E. coli O157:H7 was not detected in ready-to-eat salads, bagged sprouts and fresh cut vegetable samples examined by Jeddi et al. (2014). However, Kouchakkhani et al. (2016) reported the contamination of 10 out of 60 tested ready to eat salad samples (16.66%) with E. coli O157:H7.

Egypt

In Alexandria, Egypt, a study was conducted to investigate the prevalence of E. coli O157:H7 in the most consumed leafy green vegetables. Whole vegetables (432 samples of bulbs, cucurbits, edible corms, edible roots, legumes, solanaceous vegetables, tubers and okra) and fruits (513 samples of berries, citrus, drupes, melons, pomes, and tropical fruits), were also analysed. E. coli O157:H7 was isolated from 34 vegetable samples (7.87%) and in 6 fruit samples (1.1%). In the vegetable category, carrots were the most contaminated (37.04%; 10/27), followed by pepper (29.63%; 8/27), tomato (18.52%; 5/27) and zucchini (18.52%; 5/27). The heavy contamination of these vegetables with E. coli O157:H7 might be due irrigating these plants with wastewater from drainage canals, as evidenced in Moneim et al. (2014) that Egyptian farmers use drainage water, even though this practice is forbidden, due to the limitation of clean water for irrigation purposes. Strawberries (11.1%; 3/27) were mostly contaminated with this pathogen, followed by dates (3.7%; 1/27), cantaloupe (3.7%; 1/27) and blueberries (3.7%; 1/27). Among the 40 (6/40; fruits and 34/40; vegetables) identified E. coli O157:H7 isolates, the stx1 was detected in 12 isolates, stx2 in 13 isolates, hlyA in 5 isolates, and 20 isolates harboured the eaeA gene (Khalil & Gomaa, 2016).

Turkey

One study by Büyükünal, Issa, Aksu, and Vural (2015) assessed the microbiological quality of 161 fresh vegetable samples and 100 fruit samples produced in either greenhouses or fields, and irrigated with either ground water or river water. None of the samples were contaminated with E. coli O157:H7, but still some were contaminated with other enteric bacteria, such as Salmonella, Campylobacter species and other types of E. coli, indicating a contamination probably by the irrigating water.

Iraq

Al-Dragy & Baqer (2014) investigated lettuce collected from supermarkets for the presence of E. coli O157:H7. Out of the 25 collected samples, 5 (20%) were found contaminated with this serotype.

Conclusions

This review focuses on the prevalence of E. coli O157:H7 in animal and plant food products in the MENA region. E. coli O157:H7 is associated with severe diarrheal diseases, and young cattle is considered the main reservoir of this bacterium. E. coli O157:H7 was found in a low percentage in the mentioned food and food animals’ stuff; meat and beef, dairy products, vegetables and fruits. The average percentage of O157:H7 prevalence in meat and beef was 4.7%, whereas in dairy products was 6.4%. Besides, this percentage was 3.11% in vegetables and fruits. From the results, it seems that contrary to the notion that cattle is the reservoir for the E. coli O157:H7, it appears that, it could be part of the microbiome of some cattle as indicated from the absence of this pathogen from fecal samples obtained from other cattle. This indicates a variation among the cattle microbiome which could be reflected by the diversity of the organisms in the microbiome which might also be affected by environmental conditions and the animal feed as well as the cleanness of the water. Nonetheless, other animals appear to harbor the pathogen in small but varying degrees such as sheep, chicken and some camel’s meat. However, it is not a must to find them in these samples. Instead, the pathogen appears surfacing around in all abattoirs and meat processing facilities. The improper hygienic practices and recklessness in dealing with meat carcasses and the packaging and shipping of these products plays a major role in contamination, therefore, good hygienic practices should be implemented to prevent infections caused by E. coli O157:H7.

Furthermore, we have noticed a big variation in the results of E. coli O157:H7 detection in between different studies and among different countries. We believe that the methods of isolation and confirmation might not be appropriate leading to missing the actual numbers of the pathogen in these samples or oppositely giving false positives in other studies.

Acknowledgments;

The authors would like to Acknowledge Jordan University of Science and Technology for their logistic support.

Disclosure statement

No potential conflict of interest was reported by the author(s).