Water Buffaloes (Bubalus bubalis) Identified as an Important Reservoir of Shiga Toxin-Producing Escherichia coli in Brazil

Abstract

The presence of Shiga toxin-producing Escherichia coli (STEC) in water buffaloes is reported for the first time in South America. The prevalence of STEC ranged from 0 to 64% depending on the farm. STEC isolates exhibiting the genetic profiles stx₁stx₂ehxA iha saa and stx₂ehxA iha saa predominated. Of the 20 distinct serotypes identified, more than 50% corresponded to serotypes associated with human diseases.

Shiga toxin-producing Escherichia coli (STEC) represents an important emerging group of food-borne pathogens. Domestic animals, particularly bovine and ovine, have been incriminated as natural reservoirs of STEC all over the world (2, 7). Although O157:H7 is known to be the most important STEC serotype in many industrialized countries, hundreds of distinct STEC serotypes have been isolated from human diseases in many geographic areas, including Brazil (3, 14, 16, 18, 29).

The presence of STEC in bovines and ovines in Brazil has been described previously (5, 17, 19, 31); however, the prevalence of STEC in Brazilian water buffaloes remains unknown, and all over the world very few data on the prevalence of STEC in water buffaloes are available (7, 10, 21). The aim of this study was to estimate the prevalence and the characteristics of STEC in healthy dairy water buffaloes in Brazil, where they are intensively reared and represent the most important herd in the Americas (11).

A total of 100 healthy dairy water buffaloes from nine farms located in the central area of Minas Gerais State, Brazil, were studied. These animals were randomly selected and represented at least 10% of the herd on each farm. Fecal samples that were collected with sterile swabs and dipped into 8 ml Cary Blair transport medium were directly inoculated into 5 ml of buffered peptone water and incubated overnight at 37°C. Cultures were streaked onto MacConkey sorbitol agar and incubated as described above. A total of 655 presumptively (24) identified E. coli colonies (sorbitol positive and negative) were investigated for the presence of the stx₁,stx₂, eae, and ehxA genes using colony hybridization assays and specific DNA probes (13).

STEC isolates were analyzed for the presence of genes coding for Saa, EfaI, ToxB, and Iha using colony hybridization assays and specific DNA probes (22, 23, 26, 27, 28). Cytotoxicity assays were performed as described by Gentry and Dalrymple (12), and the hemolytic activity was determined as described by Beutin et al. (1). stx₁ and stx₂ subtypes were determined for 56 strains representing one serotype from each animal. The genetic variant stx_1c was determined as described by Zhang et al., (32), and differentiation of the stx₂ variants was carried out as described previously (30).

The antimicrobial susceptibility pattern (6) was determined for 56 strains representing one serotype from each animal. The following antimicrobial agents were tested: ampicillin, amikacin, cefoxitin, chloramphenicol, streptomycin, gentamicin, kanamycin, nalidixic acid, tetracycline, tobramycin, trimethoprim-sulfamethoxazole, cefotaxime, ceftazidime, cefepime, piperacillin- tazobactam, amoxicillin-clavulanic acid, aztreonam, imipenem, and ceftriaxone.

O/H serotypes were identified by using standard methods (9) and O (O1 to O181) and H (H1 to H56) antisera.

In accordance with worldwide data (15), the prevalence of STEC in the present study ranged from 0 to 64.3% depending on the farm (Table 1). STEC strains were not detected in animals from two farms. However, more than 60% of the animals from two farms carried STEC strains. These remarkable differences in the prevalence of STEC depending on the farm probably could be associated with the on-farm management practices, such as manure handling, housing conditions, diet, etc.

TABLE 1.

Distribution of serotypes and genetic profiles of STEC strains isolated from healthy dairy water buffaloes on different farms of Minas Gerais, Brazil

Farma	No. of animals studied	No. of STEC-positive animals (%)	Animal	Serotypeb	No. of isolates	Genetic profile
A	16	2 (12.5)	29	O137:H41	1	stx1ehxA iha saa
				O74:H25	1	stx1stx2vhaehxA iha saa
			32	ONT:H7	1	stx2ehxA iha saa
B	6	2 (33.3)	52	O159:H21	1	stx1stx2vhbehxA iha saa
			54	O41:HNTc	1	stx1cehxA iha saa
C	14	6 (42.8)	3	O77:H18	8	stx1stx2vhbehxA iha saa
				ONT:H16c	1	stx2dehxA iha
			6	ONT:H21	2	stx1stx2ehxA iha saa
			61	ONT:H21	1	stx1stx2ehxA iha saa
			65	O88:H25c	3	stx1
			67	O116:H21	1	stx2ehxA iha saa
			70	O77:H18	3	stx1stx2/2vhbehxA iha saa
D	14	9 (64.3)	14	O141:H49	6	stx2vhaehxA iha saa
			15	O178:H19	1	stx2vhbehxA iha saa
			71	O23:H7	6	stx2ehxA iha saa
			73	O82:H8	2	stx2ehxA iha saa
				ONT:H7	1	stx2ehxA iha saa
			74	ONT:H18	2	stx1stx2ehxA iha saa
			75	ONT:HNTc	1	stx1
				O93:H19	2	stx1ehxA iha saa
			76	O116:H21	1	stx2ehxA iha saa
				O74:H25	1	stx1stx2vhaehxA iha saa
				ONT:H18	4	stx1stx2ehxA iha saa
			79	O74:H25	4	stx1stx2vhaehxA iha saa
			80	ONT:H7	1	stx2ehxA iha saa
E	10	4 (40.0)	81	O178:H19	2	stx2vhbehxA iha saa
				O77:H18	1	stx2vhbehxA iha saa
			83	O59:H8	1	stx2vhaehxA iha saa
			85	O74:H25	2	stx1stx2vhaehxA iha saa
			89	O82:H-	4	stx2ehxA iha saa
F	10	3 (30.0)	100	ONT:H42	1	stx1stx2vhbehxA iha saa
			103	O113:H21	1	stx1stx2vhaiha
			105	ONT:H7	1	stx1ehxA iha saa
G	18	11 (61.1)	110	O93:H19	1	stx2ehxA
				O156:H21	2	stx1stx2vhaiha
				ONT:H21	5	stx2untehxA iha saa
			111	O156:H21	1	stx1stx2vhaiha
			112	O74:H25c	3	stx1stx2vhaehxA iha saa
			114	O74:H25c	4	stx1stx2vhaehxA iha saa
				O22:H16	1	stx1stx2ehxA iha saa
				ONT:H2	1	stx2vhbehxA iha saa
			115	O82:H8	1	stx2ehxA iha saa
			117	O49:HNTc	1	stx1c
				O49:H21c	1	stx1c
			118	O77:H41	1	stx1ehxA iha saa
			119	O176:H2	1	stx2vhbehxA iha saa
				O49:H21c	1	stx1c
				ONT:H38c	2	stx1c
				ONT:H14c	1	stx1c
			121	O93:H16 (H-)	6	stx1iha saa
				ONT:HNT	1	stx1c
			123	O79:H14	1	stx1stx2vhbehxA iha saa
				ONT:H8	1	stx1c
			126	O176:H2	3	stx2vhbehxA iha saa
				ONT:H2	1	stx2vhbehxA iha saa
Total	100	37 (37)			109

^aSTEC was not detected on two farms (farms H and I).

^bBold type indicates serotypes associated with human diseases.

^cNegative for the cytotoxicity assay with Vero cells.

More than 70% of the STEC strains were typeable, and the strains belonged to 20 distinct serotypes. More than 50% of the serotypes and untypeable strains associated with H2, H7, H8, H14, H16, H18, and H21 antigens identified in this study had previously been described as serotypes and strains associated with human illness (www.microbionet.com.au/vtectable.htm). Seven serotypes, O23:H7, O74:H25, O77:H18, O82:H-, O93:H16, O141:H49, and O176:H2, accounted for 47.7% of the isolates. Serotype O74:H25 was the most frequent and was recovered from animals reared on four of seven farms, suggesting that there was widespread dissemination of certain serotypes among distinct farms. Other serotypes, such as O77:H18, O23:H7, and O141:H49, were recovered mostly from water buffaloes reared on one farm. Although in the area studied water buffaloes and other dairy and beef cattle are usually reared together, serotypes O23:H7, O49:H21, O59:H8, O74:H25, O77:H41, O93:H16, O93:H19, O137:H41, and O176:H2, which accounted for 45% of the serotypes found in water buffaloes, were not found among STEC strains from dairy and beef cattle (data not shown).

Of the 109 STEC isolates, 42 (38.5%) carried stx_2, 43 (39.5%) carried stx₁ and stx₂ sequences, and only 24 (22%) harbored the stx₁ sequence. The majority of STEC strains belonging to serotypes not previously reported to be serotypes associated with human illness and carrying the stx₁ sequence alone carried the stx_1c subtype. These strains were devoid of additional virulence factors and were negative for the cytotoxicity assay with Vero cells. In contrast, STEC serotypes associated with human diseases harbored stx₂ or stx₁ and stx₂ sequences and belonged to the stx₂, stx_2vha, or stx_2vhb subtype; one strain belonged to the stx_2d subtype, and one strain was untypeable with the method currently used. STEC isolates exhibiting the genetic profiles stx₁ stx₂ ehxA iha saa and stx₂ ehxA iha saa accounted for more than 70% of the isolates.

All STEC isolates were devoid of the eae gene. This result can be related to the serotypes found in these animals. According to Sandhu et al. (25), the presence of the eae gene is associated with certain O groups, such as O26, O103, O111, O145, and O157, none of which was identified in the STEC in the present study. The saa and iha gene sequences were detected alone or in association in 83.5% of the STEC strains. This wide distribution of Saa among eae-negative strains is in agreement with the findings of Paton et al. (23). With the lack of the eae gene, distinct adhesins other than intimin can have an important role in adherence to the intestinal epithelium and colonization of the gut. The ehxA gene sequence was detected in 86 of the 109 STEC strains (78.9%), and all of the strains expressed hemolytic activity after 18 to 24 h of incubation.

The majority of STEC strains were susceptible to all antimicrobials tested. Resistance to one drug (nalidixic acid, streptomycin, or ampicillin) and resistance to two antimicrobials (ampicillin plus streptomycin and nalidixic acid plus ampicillin) were found in 10 (17.8%) and 2 (3.6%) of the strains, respectively.

The prevalence of STEC strains having the genetic profile stx₁ stx₂ ehxA iha saa or stx₂ ehxA iha saa deserves great attention as STEC strains carrying the stx₂ gene are commonly associated with more severe disease (4, 8). Moreover, these strains carry other virulence genes, such as ehxA, iha, and saa, which can enhance their virulence.

This is the first report of the presence of STEC in water buffaloes in South America. Unpasteurized water buffalo milk may represent a potential risk to public health since it is used for homemade mozzarella production due to its high fat and casein content (20).

Because water buffalo farming is increasing as an important economic activity, control measures for hygienic practice, particularly in milking, surveillance, and legislation have to be improved in order to prevent fecal contamination of milk and dairy products.