Low frequency of enterohemorrhagic, enteroinvasive and diffusely adherent Escherichia coli in children under 5 years in rural Mozambique: a case-control study

The study was conducted by the Manhiça Health Research Centre (Centro de Investigação em Saúde de Manhiça - CISM), in Manhiça district, a rural area in Maputo Province, southern Mozambique. The district is located approximately 80 km north of Maputo city with two distinct seasons, rainy and hot season from November to April and dry and cool season from May to October, as detailed elsewhere [18, 19]. The CISM has been running a continuous Demographic Surveillance System (DSS) for vital events and migrations since 1996 [18], currently covering the entire Manhiça district with an estimated population of 183,000 inhabitants in 43,000 households with two regular visits a year to update demographic events (migration, births, deaths, etc.).

We analyzed a subset of E. coli isolates collected under the GEMS protocol conducted between two distinct periods, that aimed to quantify the burden, etiology and sequelae of diarrhea in sub-Saharan Africa and South Asia [16]. The first period (December 10, 2007 and October 31, 2011) comprised only the enrolment of children with MSD and respective controls without diarrhea [16], while the second one (November 3, 2011 and November 2, 2012) comprised two parallel case-control studies, one assessing MSD and the other LSD; each with their respective controls [17]. In the GEMS study, children less than 5 years with diarrhea (≥3 loose stools within the previous 24 h) belonging to the catchment area under DSS who sought care at the sentinel health centers (SHCs) were assessed for study criteria and enrolled in LSD or MSD group. MSD was defined as a new (onset after ≥7 diarrhea-free days) and acute (onset within the previous 7 days) diarrheal episode presenting one of the following criteria: sunken eyes, loss of skin turgor, intravenous hydration administered or prescribed, dysentery, or hospitalization [20]. LSD was defined as a new acute diarrhea case seen at SHCs that did not meet the definition of MSD. For each case, 1–3 community based controls (children free of diarrhea in the last 7 days) were enrolled matched by age stratum (infants [0–11 months], toddlers [12–23 months] and young children [24–59 months]), neighborhood and gender as described elsewhere [20]. At enrolment, each participant provided fresh stool that was placed in cold chain and transport media according to the protocol. If antibiotics were to be administered to participants with diarrhea before stool was produced, we obtained two rectal swabs for bacterial culture pending passage of the whole stool for the remaining assays [20].

In GEMS protocol, the stool samples from cases and controls were assessed for numerous enteropathogens (bacterial, protozoal and viral agents) using microbiological and molecular methods [17, 21]. For E. coli isolation, stool samples were plated onto MacConkey (MAC) agar and incubated at 37 °C for 18–24 h. Afterward, putative lactose-fermenting bacterial colonies resembling E. coli were picked and tested using Motility Indole Ornithine medium. Up to 3 lactose and indole positive E. coli colonies per sample were selected and stored at − 80 °C in single colony for further analysis [21]. In GEMS protocol, the E. coli colonies were analyzed by multiplex PCR that targeted only for ETEC, EAEC and EPEC [21]. Therefore, in this investigation, the E. coli colonies from GEMS were screened by conventional PCR targeting additional pathotypes of DEC as EHEC (stx1 and stx2), DAEC (daaE) and EIEC (ial and ipaH). Briefly, the three putative E. coli colonies isolated from stools in GEMS study were retrieved onto MAC and incubated at 37 °C for 18–24 h. Afterward, the E. coli colonies from same stool were pooled, the DNA extracted [21] and analyzed by multiplex PCRs using primers previously designed for detection of ial [22], ipaH [23], daaE [24], stx1 and stx2 genes [21]. We performed two multiplex PCRs; the first one detected genes ial, ipaH and daaE and the second one the genes stx1 and stx2. For the first multiplex, 3 μl of DNA template was added to the PCR mix containing 12.5 μl of PCR Mix 2X (Qiagen), 5 μl of Q-solution 10X (Qiagen), 0.5 μl of 10 μM of each primer and 1.5 μl of RNase-free water to a final volume of 25 μl. The second multiplex contained: 12.5 μl of PCR Mix 2X (Qiagen), 5 μl of Q-solution 10X (Qiagen), 0.2 μl of 25 μM of each primer, 3.7 μl of RNase-free water and 3 μl of DNA template to a final volume of 25 μl. A single cycling protocol was applied for both multiplex, with following parameters: preheating at 95 °C for 15 min; and 35 cycles of denaturation at 94 °C for 30 s, annealing at 57 °C for 90 s, elongation at 72 °C for 90 s; with final extension at 72 °C for 10 min in an Eppendorf Mastercycler Gradient thermal cycler (Eppendorf, Hamburg, Germany). The amplification products were separated through a 2% agarose gel stained with ethidium bromide. The 1-kb plusA 100-bp DNA ladder (Bio-Rad) was used as a molecular size marker in gel.

We analyzed 297 stools samples from cases with LSD matched to 297 controls (children without diarrhea), and 89 stool samples from cases with MSD matched to 222 controls, collected between November 3, 2011 and November 2, 2012. Among the cases with LSD, 120 children aged 12–23 months, 104 were infants, and 73 were young children (Table 1). In contrast, the MSD study showed that, diarrhea was most frequent in infants (42 cases matched to 99 controls), followed in children aged 12–23 months (29 cases matched to 79 controls) and lastly in young children (18 cases matched to 44 controls). We also found that diarrhea was more frequently reported in male children than in female children, both in LSD (54.2%, 161/297) and in the MSD study (64.2%, 57/89).

DEC were more common among LSD cases (2.7%, [8/297] of cases vs. 1.3% [4/297] of controls; p = 0.243]) than in MSD cases (0%, [0/89] of cases vs. 0.4%, [1/222] of controls; p = 1.000). Detailed analysis of DEC circulation for the above period showed a low frequency of EHEC, EIEC and DAEC in all age strata with no significant association with neither LSD nor MSD. Although the low frequency of pathotypes among cases in the LSD group, EIEC was predominant in children aged 24–59 months (4.1% [3/73] for cases vs. 0% for controls). This was followed by DAEC in children from 0–11  months (1.9% [2/104] for cases vs. 1.9% [2/104] for controls) and lastly EHEC being isolated in one sample from control (Table 1). Among the MSD group, DAEC was only detected in one sample from control (0–11 months), while EIEC and EHEC were absent.

The analysis of a subset of E. coli (118 from cases with MSD and from 154 matched controls) isolated between December 10, 2007 to October 31, 2011 revealed remarkable variation of pathotypes EHEC, EIEC and DAEC frequencies when compared to the period between November 3, 2011 and November 2, 2012. We observed high frequency of DEC in controls compared to MSD cases (16.2%, [25/154] vs. 11.9%, [14/118], p = 0.383, respectively) between December 10, 2007 to October 31, 2011. Among these, DAEC was the most common pathotype, being detected in a similar proportion in cases (8.4%; 6/71) and controls (8.4%; 8/95) during infancy. Nonetheless, DAEC frequency was higher in controls aged 12–23 months (0% for cases vs. 7.1% for controls) and 24–59 months (7.7% for cases vs. 17.6% for controls). In contrast, EIEC was found more frequent in cases than in controls for all age strata, although no association was observed (Table 2). EHEC was only found in three samples; one sample from case (infant) and one from control (infant) and in one sample from a control aged 24–59 months.